Optimization of process parameters for scanning human face using hand-held scanner

Optimization of process parameters for scanning human face using hand-held scanner | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Optimization of process parameters for scanning human face using hand-held scanner Ashish Kaushik, Upender Punia, Ramesh Kumar Garg, Mohit Yadav, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2051093/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Three-dimensional surface scanning of several anatomical areas or human body has gained popularity in current decades due to pre-surgical planning and improved workflow of patient diagnosis and treatment Living surfaces, such as the human face, have various degrees of surface complexity to account for, as well as a range of process parameters to consider. In the proposed work, the face of a person was scanned in various combinations of input parameters using a handheld laser scanner, SENSE 3D (3D system, Rock Hill, SC/USA). Scanner to surface distance, angular orientation, and illumination intensity are considered significant input parameters while using laser scanners for 3D facial data. A number of twenty experimental runs and input parameter combination were suggested by face centered central composite design. The human face has been scanned on these twenty runs to retrieve 3D CAD model and FID score of each model has been completed to investigate the quality/accuracy of the captured data. A model has been trained among input and output using a neural network and further, it is optimized using a genetic algorithm to maximize accuracy The minimum, FID score achieved 270.24, obtained with a scanning distance of 22 inches, the angular orientation of 67.5 degrees, and ambient lightning condition of 16 watt/meter square in twenty experimental runs. The accuracy is maximized by minimizing the FID score utilizing a heuristic GA-ANN technique having 28 inches as scanning distance, 48.041 degrees as angular orientation, and 18 watt/meter square as the ambient lighting condition. Digital fabrication 3D scanning 3D laser scanner Industry 4.0 Design of experiments GA-ANN Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Introduction The challenge of accurately evaluating the size and shape of the human face has always sparked the interest of both researchers and doctors. The utilization of 3D surface scanning technique to create digitized models of human anatomical parts that can assist with altering the way a huge variety of products are planned and manufactured [1]. One of the fundamental problems which any novel user encounters during 3D laser scanning is the choice of suitable process parameters to obtain the full scan with very few steps. There is always an option associated with input process parameters like angular orientation, relative scanner distance, and the effect of light intensity, which is not always too intuitive but also strongly influences the scan results. The demand for accurate assessment and visualization of the facial bone and other tissues increases due to advancements in dentistry, maxillo-facial and plastic surgery. Face scanning is also examined to evaluate the outcomes of facial plastic surgery [2]. 3D Scanning is the technique of detailed analysis of real-world capture to gather data on its specifications related to dimensions appearance. From the past few years, the majority of reports are published on the utilization of 3D scanners in the medical, dental, and healthcare sectors is considerably increased, which might prove helpful for plastic surgery [3–7] . Scanning technology during its infancy period is limited to industrial applications as the scanning of the human face requires a sequence of specific conditions compared to industrial objects [4] . Various studies focused on improving the performance and optimization of process parameters of handheld scanners studying the face of person, including a progression of exceptional circumstances. As the face of living person can't be immobilized, the scanner must feature a brief recording span. Hence a handheld 3D scanner is used for this purpose. Moreover, a portable handheld scanner can be moved in various directions and enables us to scan from different orientations. A study has been performed to create a laser triangulation 3D scanner whose sole purpose is to solve the limitation of occlusion. It uses two distinct laser colors, namely green and red, along with a charge coupled device camera (CCDC) and helps enhance the reliability of respective system [8]. Several research focused on design of a time-to-digital converter and measurement of time of flight (ToF), in order to utilize same in portable scanners [9]. The dependency of ambient lighting as an essential factor in influencing the quality of the captured signal by Charge Couple Devices (CCD), and the impact of surface roughness the object during measurement was also recorded in few studies. Optimum results are obtained in case of absence or limiting ambient lighting conditions [10]. The research was performed to worked with the diffuse reflections of the object’s surface being scanned and an effective method to restrict the outcome of ambient illumination by implementing the distinct light filters upon the CCD sensors [11] . Moreover, a general solution to use the coating spray for covering the object by matty white layer is also introduced. The effects of other influential factors such as incident angle, distance from the scanner, and object color are also studied on a Computer Numerically Controlled (CNC) laser scanning process [12] . A primary structured light pattern for 3D structured light scanner is implemented in a research, during development, the suggested system's accuracy and resilience were evaluated on artificial items with established surface geometry, followed by assessments on human individuals [ 13 ]. Several researches indicated the problems in data acquisition of dark, translucent, and glossy surfaces. Generally, in the case of shining surfaces, noise is eventually added during the measurement, and the points exclusive to the actual surface can also be provoked. In both instances, sensors cannot achieve data locally (lost data), or the scan measurement is adversely affected [14]. These shiny surfaces are covered with coating sprays that help scan the objects' dark, glossy, and translucent surfaces. It was found that for 5–15 micrometers, the respective additional thickness and variation is about 45 micrometers using comparatively thin coating sprays. A low cost scanner with a hemispherical workspace has been designed and implemented with least square minimization approach to realign the parameters for achieving optimized results[15]. Further, a modified particle swarm technique, to optimize the significant morphological parameters in a contactless laser scanning method in various research [16]. Few of them investigate the effectiveness of positioning aids for obtaining 3D data in various clothing postures and configurations, through a superior quality body scanner [17] . The quality of digitized points with the help of 3D laser scanning is evaluated by several criteria, namely density, completeness, noise, and accuracy [18,19] . Several types of research are published regarding the performance of 3D scanners at various surface-to-scanner comparative orientations, and altogether examined scanners in normal position w.r.t the surface as the best condition [20–22]. A handheld, cost-effective 3D laser scanner was used to scan a human face under three significant input factors, i.e., scanning distance, angular orientation, and light intensity. A combination of input factors for twenty experiments has been designed based on face-centered central composite design. Accordingly, twenty CAD models have been retrieved on the twenty combinations of input factors. The accuracy of scan models is investigated through the Frechet Inception Distance (FID) score, which is taken as output. A model has been trained among input and output using a neural network, and further, it is optimized using a genetic algorithm. Materials And Methods To test new scanning technologies, it has been suggested that high accuracy and resolution laser scanners be used [23] . 3D Sense (3D system, Rock Hill, SC/USA) is a minimal expense, a convenient 3-dimensional surface scanner with, as per the maker, an exactness of near millimeters with resolution at Scanning range exists between 177.8 and 1828.8mm along with the color resolution of 1920 x1080 pixels. In contrast, the range of operation lies between 0.2 to 1.6 m. The field of view:45° horizontally, 57.5° in the vertical direction, and 69° diagonally with the physical configuration of the scanner is 129 (w) x 179.8(h) x 33(d) mm and maximal image of 30 /fps. 3.1 Proposed method of scanning 3.1.1 Scanning the person To capture the scan, get a person seated in normal posture and directs the scanner in their general direction. Although it is possible to scan virtually anything, clothing and accessories can still affect the scanning, so avoid wearing brighter and darker clothes. If the materials are difficult to scan, the scan can be performed by increasing the sensitivity. Also, spectacles or sunglasses can affect the texture, so these should be removed before scanning. Enhanced mesh quality, stabilized tracking, and proper orientation of trimming plane can be obtained by the Object Recognition feature, which includes object, body, and head. Figure 1 represents the object recognition features available in the SENSE software. The green color in our preview indicates either the presence of data within the scan volume representing data build-up or correct distance from proper focus, so our aim should be to adjust the sensor's reach to get the subject in the green zone. To begin with, firstly focus lies in capturing the face of the person for optimal trajectory and should keep away from capturing the face further as they turn around. Start the scan from ear to nose, lower the scanner to capture the chin from below, and further raise it to scan the area above the forehead. Return the scanner to nose level and then scan the face by moving it to the second ear. After completing the face-scanning, we moved towards hair scanning. Remember to scan from all the possible angles. The person sitting on a rotating or swivel table will spin slowly. This may take several seconds to several minutes, so the person must be in the proper posture to avoid geometry distortion during the scanning. Keep the shoulders and back in the scanner field of view. Attention is mainly focused on the target in green color for accurate capturing. If the tracking is lost, return the scanner to its previous position or the person to its last orientation to recover the lost scan data. The scanner will continue to scan further once it finds its left-off place. As the person completes one turn, the scanner should stop scanning. Experimental setup for real-time capturing of human faces is represented in Fig. 2 3.1.2 Processing the scan At this stage, some preliminary processing to reduce the number of faces and colorize the scan is performed to simplify the geometry so that it is easy to work within the domain of mesh software. Figure 3 indicates the several tools required during the preliminary processing of scan models. Once the scan is completed, several tools are available to make any modifications in the scan, such as: Crop to remove any excess scan data captured during the scan Trim, which is like crop, but the difference lies in the fact that smaller of two sections obtained by trim line is removed in case of trimming, Color - The brightness and contrast of the model can be adjusted by the Color tool, and a preview is obtained after making the final adjustments before applying. Fig. 4 describes the color tools used in SENSE software for regulating the image sensitivity and contrast of objects. Erase tool to remove or eliminate unwanted scan portions, which can be performed by moving the cursor over the unwanted scan portion or by mouse button, and the unwanted scan will be removed. 3.1.3 Exporting the scan Several file formats are available to export the raw data obtained during scanning. However, the STL format is the most widely used for exporting data because of the ubiquity of its use. 3.1.4 Edit and evaluation of data Data exported in the above stage consists of some noise unnecessary data errors in mesh surfaces such as tiny openings or some areas with faulty normal and needs to be repaired or removed before further processing. 3.2 Frechet inception distance (FID) The FID is a standard metric for assessing the worth of produced images and, more explicitly, created to estimate the execution of productive adversarial structure [24] . Martin used FID score technique in the study that suggest a two time- scale update rule (TTUR) for training generative adversarial network(GANs) by stochastic approximation. It has been noticed that the convergence of TTUR is directly depend on the assumption of local nash equilibrium. The mention score has been advancement above the prevailing inception score, abbreviated as IS. Inception score determines the condition of an accumulation of synthetic images centered on the top performance of the image. The scores incorporate equally the certainty of the restricted class forecasts for every manufactured picture and basic of marginal probability of forecasted classes. This score doesn't compare how synthesized pictures contrast with authentic pictures. The mean to foster FID score was to assess synthetic pictures fixated on the insights of gathering engineered images recognized to measure an assortment of original images from the objective space. The goal of machine learning (ML) technology is to make medical procedures relatively easy, efficient, and superior.. [ 25 ]. Experimental Design Matrix The statistical technique of design of experiments (DoE) is implemented in the present work which optimize the systems performance through previously known input variables. The process generally evaluates the effect of combination of several factors at a time which is also refereed as interactions using factorial designs. The experimental table is derived using face centered composite design approach. Figure 5 clearly demonstrates the flow chart of the proposed work which is completed in five phases. The raw data collected in phase I is used for creating the experimental design matrix using DoE methodology. All the experiments as uggested by DoE with combination of input factors are now performed in phase 3. Data can be analysed and corresponding models are created in phase IV followed by model validation in subsequent stage. Aforesaid phases in this research are further discussed in detail. 4.1 Phase 1: Collection of raw data Raw data is collected to set a feasible range of significant input parameters required to optimize. This research mainly needs to optimize three input parameters for scan models. The parameters mentioned below are taken into consideration: Angle (A) - corresponds to the angular orientation of global rotation. Ambient Lightning Conditions (I) - in indoor surroundings. Scanning distance (D) is the relative distance of the scanner from the object's orientation. For consideration, scanning distance lies between 18 inches to 28 inches, angular variation lies between 45 degrees to 90 degrees, and light intensity lies in 12 watts per meter square to 18 watts per meter square. Table 1 showcase the variables assigned as A,B,C with their range intervals. Table 1 Significant process parameter with their range in central composite design Process Parameter Range in face centered central composite design A-Scanning Distance (in inches) 18 28 B-Angular Variation (in degrees) 45 90 C-Light Intensity (in Watt/meter square) 12 18 4.2. Phase 2: Creating Design Table A method of Design of Experiments (DoE) is determined according to the objectives of experiments. Face centered composite design approach was adopted to analyze the influence of operating variables on accuracy of scan models. In order to achieve the optimum conditions, three process factors were taken in to account i.e. scanning distance(A), angular orientation (B), and light intensity(C). The present approach includes 2 n factorial runs, n c center runs and 2n axial runs. The reproduction of data and experimental error are controlled by center points [26] . The total number of experimental runs that are required to perform can be obtained using Eq. (1) [27]. N = 2 n + 2n + n c (1) Here n represents the number of process factor, N denotes the total number of runs and n c represents the center point number. The present approach specifies that twenty experimental runs are needed for the process of optimization including six axial experiments, eight factorial experiments, and six center experiments. Table 2 demonstrates the CCD approach adopted in this proposed work. Table 2 Input Parameters Obtained by Using DoE Std Run Scanning distance (D in Inches) Scanning angle (A in degrees) Intensity ( I in watt/meter sq.) 1 1 19.6216 54.1214 14.8108 14 2 26 67.5 16 9 3 22 67.5 14 2 4 19.6216 54.1214 17.1892 17 5 22 67.5 16 12 6 22 90 16 6 7 24.3784 54.1214 17.18192 16 8 22 67.5 16 7 9 24.3784 80.876 14.8108 19 10 22 67.5 16 15 11 22 67.5 16 18 12 22 67.5 16 3 13 19.6216 80.8786 14.8108 11 14 22 45 16 8 15 24.3784 80.8786 17.1892 13 16 18 67.5 16 10 17 22 67.5 18 4 18 19.6216 80.8786 17.1892 5 19 24.3784 54.1214 14.8108 20 20 22 67.5 16 4.3 Phase 3: Experiments Twenty scans are conducted according to the values of input parameters suggested by Design of Experiments software. The obtained scan images are the results of the respective set of input values presented by software, as depicted in Fig. 6 as model M1, M2, M3 and so on. Results And Discussion Twenty CAD models were retrieved according to the combination of significant factors for twenty experimental runs based upon the combination of significant input parameters ie. scanning distance, light intensity and angular orientation. Accuracy evaluation of retrieved models helps us in identification of suitable model or provides a scope for further improvement in models. The performance metric adopted for evaluating accuracy of scan models is FID as discussed above. Table 3 . represents the FID score of scan models with their corresponding parameters. Model M11 is the best-suited model according to the FID technique. Table 3 FID Scores of Scan Models Run Scanning distance (d in inches) Scanning angle (A in degrees) Intensity ( I in watt/meter sq.) FID score M1 19.6216 54.1214 14.8108 333.87 M2 26 67.5 16 310.978 M3 22 67.5 14 375.576 M4 19.6216 54.1214 17.1892 347.023 M5 22 67.5 16 274.891 M6 22 90 16 346.22 M7 24.3784 54.1214 17.18192 297.061 M8 22 67.5 16 313.939 M9 24.3784 80.876 14.8108 290.511 M10 22 67.5 16 271.894 M11 22 67.5 16 270.24 M12 22 67.5 16 273.637 M13 19.6216 80.8786 14.8108 298.579 M14 22 45 16 334.062 M15 24.3784 80.8786 17.1892 284.83 M16 18 67.5 16 319.741 M17 22 67.5 18 278.021 M18 19.6216 80.8786 17.1892 322.404 M19 24.3784 54.1214 14.8108 332.678 M20 22 67.5 16 272.361 Figure 7 depicts the variation of the accuracy of scan models concerning FID score where scan models are depicted on the X-axis and FID score is plotted on Y-axis of cartesian coordinates. The curve indicates variation of scan models to FID score 5.1 Artificial neural network The mechanism of the brain serves as the basis for neural networks. .The neural structure is made up of an arranged layer of synthesized neurons which are linked to all remaining artificial neurons using a coefficient (weight function) particularly known as the element of processing. Weighted information is used in the neural network model. First-layer neurons acquire weighted information in the neural network. The solution to the problem is determined based on the data generated by the most recent layer of neurons as well as the hidden layer that lies underlying it. Across numerous training procedures, the inter-unit connectors were optimized until the prediction error reached a minimum value; evidently, the system obtained the precise value. The ANN can foresee instances that haven't been shown to the system before owing to generalization. The approach of FCCD developed a design matrix (shown in Table 3 ) with three input factors and one experimental output result that has a non-linear relationship and is utilized for training, testing, and validation. The root mean square error and related correlation coefficients were used as assessing parameters for the ANN model with different layouts, which was trained with 70% of the data, tested with 15% of the data, and validated with 15% of the data [28]. The error is calculated through forward propagation by first putting all of the inputs together, then multiplying arbitrary weights and biases, and finally applying this calculation to the output of the tangential sigmoid activation function. During backward propagation of the error with updated weight, the method of gradient descent was utilized, resulting in a feed-forward back propagation network (FFBP) [29–30]. Although more computing is required, the guided algorithm Levenberg Marquardt (LM) back propagation was employed since it is the precise and most feasible of all algorithms. The FCCCD presented a design matrix with three input parameters (Scanning Distance, Angular orientation, Light intensity) and one experimental output responses (FID Score) that have a nonlinear relationship and have been utilized for training, testing, and validation. Forward propagation estimates the inaccuracy by adding together inputs, multiplying weights, and applying preferences to the activation function of the tangent sigmoid. During backward propagation, the gradient descent approach was employed for the error with updated weight generating feed-forward back propagation (FFBP) network. This ANN generated a model based on structure, as shown in Fig. 8 that was trained with 20 sets of three input variables and one output responses, and the training state for the ANN model is demonstrated in Fig. 9 , at epoch 6. The controlled algorithm Levenberg Marquardt (LM) backpropagation yielded the highest overall value of R (0.95), as shown in Fig. 10 . This approach was chosen because it is the best ever and most stable of all other algorithms however, more processing is needed. At the sixth iteration of training, the input data stretches its highest optimal solution, and when the mean square error (MSE) of evidence samples starts to climb, the epochs involuntarily stop, as depicted in the MSE graph in Fig. 12. The finest performance for validation at epoch 0 was 0.034044. 5.2 Heuristic tool for optimization in genetic algorithms and artificial neural networks (GA-ANN) The goal is to optimize accuracy by minimising FID score, and the model built using the ANN heuristic tool is executed as a fitness function in the genetic algorithm (GA-ANN). Table 4 . displays the Genetic algorithm Heuristic tool's parameters. The population size is the numeral of chromosomes produced at arbitrary in a single iteration to examine the output response by using a fitness function. Mutation and Crossover functions create children from parents, leading in the development of innovative chromosomes to determine the best output. The population range can be any value that yields the largest chromosomes inside a particular bound. A population size of 50 was selected for optimization. The suggested approach makes use of a constraint dependent as crossover and mutation function. The crossover percentage has been fixed to 0.8, whereas rate of mutation has been set as 0.01. The number of chromosomes which best match is carried over to the following generation is known as the elite count.. In the proposed work, the value of elite count is taken to be 2.5. Genetic algorithms fitness function is based on the ANN model. The results of the GA are presented in Table 8, and they include the best input values that ensure the output response is balanced. The model was further validated by conducting confirmative experiments using the GA-ANN's optimum arrangement sets of taken input process parameters. Table 4 Heuristic tool genetic algorithm parameters Parameters of GA Value Parameters of GA Value No. of input variables 3 Mutation rate 0.01 No. of output responses 1 Mutation function Constraint dependent Population size 50 Stopping criteria No. of generations Crossover fraction 0.8 Elite count 2.5 Cross over function Constraint dependent Fitness function ANN model To verify that the results are accurate, each set of trials is repeated three times, and the average of those values is calculated. The experimental and expected values were quite similar in the confirmation experiments. Conclusion The proposed research work has proficiently explored the methods for data acquisition and retrieving the CAD models through reverse engineering/scanning. The proposed study demonstrated that using handheld scanner with proper optimized process parameters will result in a valid and accurate models /data acquisition of irregular surfaces like human face. The current research successfully examined the impact of significant input parameters that predominately affect the scanning of human faced. The best-achieved accuracy was found for experimental model M11 with the lowest FID score, i.e., 270.24, obtained with a scanning distance of 22 inches, the angular orientation of 67.5 degrees, and ambient lightning condition of 16 watt/meter square. Parametric optimization is achieved with the help of hybrid GA-ANN for improving accuracy of scan models using FID score. The accuracy is maximized by decreasing the FID score to 186.3569 through heuristic GA-ANN tool having 28 inches as scanning distance, 48.041 degrees as angular orientation, and 18 watt/meter square as the ambient lighting condition. In summary, the proposed method using handheld 3D laser scanners for retrieving human face can be implemented in clinical utilities, craniofacial research, pre surgical treatment, and assessment of soft tissues. Declarations Compliance with Ethical Standards: Disclosure of potential conflicts of interest : The author hereby confirms that there are no potential conflicts of interest. Research involving Human Participants and/or Animals: The research involves a person voluntarily interested in face scanning. No animals are involved during research. Informed consent :Not applicable Ethical Approval and Consent to Participate: The study requires no ethical approval. Face of a male colleague has been scanned who willingly participate during research. Human Ethics : Not applicable. Consent for Publication: Yes, all the authors and other person who are involved during the research are agreed to permission for publication. Availability of Supporting Data : Not appropriate. Competing Interests: Authors disclose that there is no competing interest of proposed work. Funding : There is no funding/any grant for the proposed research. Authors' Contributions : Conceptualization and Supervision: Ramesh Kumar Garg, Mannu Rathee, Ravinder Kumar Sahdev. Draft writing, figures: Ashish Kaushik, Upender Punia. Data contribution and tool analysis: Mohit Yadav, Rajat Vashistha, Deepak Chhabra Acknowledgements Author sincerely acknowledges CAD/CAM & Additive manufacturing lab of MD University for providing tools/equipments facilities. Authors' Information Ashish Kaushik graduated from M.D. University, Rohtak, Haryana, India in 2018 with a B.Tech. and an M.Tech. in Mechanical Engineering. He is currently enrolled in the Department of Mechanical Engineering at the DCRUST Murthal, India, for the purpose of obtaining his doctorate degree.. His area of research are Additive Manufacturing, Biomaterials, Scanning, Reverse Engineering, Designing and Optimization. Upender Punia graduated from GJUST, Hisar, with a B.Tech in Mechanical Engineering in 2016 and from M.D. Univetrsity Rohtak, Haryana, with an M.Tech in 2020. He is in the process of getting his Ph.D. in Mechanical Engineering from India's DCRUST, Murthal. His area of research are Additive Manufacturing, Biomaterials, Scanning, Reverse Engineering, Designing and Optimization. Dr. Ramesh Kumar Garg is currently employed as a Chairman and Professor in the Department of Mechanical Engineering at the DCRUST, Murthal, India . More than sixty of his research papers have been published in either International or National Journals, and he has been visited to a number of nations as part of an academic programme. Among his scientific interests include production, system design, multi-criteria decision-making methods, and industrial engineering. Mohit Yadav has obtained the Masters of science & Master of philosphy degree in Applied Sciences, Mathematics from M.D. University, Rohtak, Haryana, India. He is continuing to pursue a doctorate in Mathematics at the University Institute of Engineering & Technology. His areas of research are Fluid Mechanics, Renewable Energy Harvesting, Mathematical Modeling, Simulation and Optimization. Rajat Vashistha earned his Master of Technology in Mechanical Engineering with a speciality in biomechanics in 2018 . His study focuses on the implementation of AI-based techniques. A CIBIT scholarship was awarded to him in recognition of the merit of his MPhil project, which involved the implementation of deep learning to medical imaging in order to detect brain tumours at a preliminary phase. At Vigyan Prasar, an independent institute of the Department of Science and Technology of the Government of India, Rajat worked as a Senior Project Officer before joining CIBIT. He also served as an Assistant Professor at NIMS University Rajasthan before that. Dr. Manu Rathee is the Head of the Department of Prosthodontics at the P.G.I of Dental Sciences at the P.B.D.S University of Health Sciences in Rohtak, Haryana. He is also serving as a Senior Professor there. Co-PI has experience of working in the field of additive manufacturing and has conducted/guided original research on 3D printing in dentistry via in-vivo and in-vitro study designs. She has conducted a randomized controllable clinical testing on comparative assessment of the consequence of conventionally fabricated versus 3D printed provisional restorations for fixed dental implant prosthesis and randomized controlled clinical trial on comparative evaluation of the effect of CAD/CAM versus 3D printed provisional restorations for fixed dental implant prosthesis. Another original research with in-vitro study design was conducted on comparative evaluation of fracture resistance of anterior provisional restorations fabricated using conventional, CAD/CAM and 3D printed techniques. She has published various articles including original research, case reports and review articles in peer reviewed indexed National and International Journals. She has successfully delivered various intraoral fixed and removable prosthesis using 3D printing technology including single unit as well as and multiple unit fixed prosthesis. Dr. Ravinder Kumar Sahdev is currently serving as an Assistant Professor in Mechanical Engineering Department, UIET, M.D. University, Rohtak. He has published more than 35 publications in reputed Conferences and Journals. He has completed one research project. His major areas of research areas are Additive Manufacturing, Solar Thermal, Heat Transfer, Drying, Solar Energy Conversion, Drying Technology, Grain Drying, Solar Energy, PVT & PCM. Dr. Deepak Chhabra is employed at M.D University Rohtak as an Assistant Professor in the Mechanical Engineering Department of UIET, India. More than Seventy of their scientific articles have been published in International or National Journals or as Book Chapters by Elsevier, and he has visited to a variety of nations in preparation for International Conferences. His areas of interest in research include Mechanobiology, Active Vibration Control, 3D Printing, Finite Element Modeling, Scanning, Biomaterials & Optimization etc. References S. L. Kovacs, A. Zimmermann, G. Brockmann, M. Gühring, H. Baurecht, N. A. Papadopulos, H. F. Zeilhofer, Three-dimensional recording of the human face with a 3D laser scanner, J Plast Reconstr Aesthet Surg 59 (11) (2006) 1193-1202, https://doi.org/10.1016/j.bjps.2005.10.025. S. C. Aung, R. C. K. Ngim, S. T. Lee, Evaluation of the laser scanner as a surface measuring tool and its accuracy compared with direct facial anthropometric measurements, Br. J. Plast. Surg. 48 (8) (1995) 551-558,https://doi.org/10.1016/0007-1226(95)90043-8. C. L. Istook, S. J. Hwang, 3D body scanning systems with application to the apparel industry, J. Fash. Mark. Manag. 5(2) (2001) 120-132, https://www.emerald.com/insight/content/doi/10.1108/EUM0000000007283/full/pdf. Haleem, M. Javaid, 3D scanning applications in medical field: a literature-based review, Clin. Epidemiol. Glob. Health 7(2) (2019) 199-210, https://doi.org/10.1016/j.cegh.2018.05.006. G. Bhatia, M. W. Vannier, K. E. Smith, P. K. Commean, J. Riolo, V. L. Young, Quantification of facial surface change using a structured light scanner, Plast. Reconstr. Surg. 94(6) (1994) 768-774, https://doi.org/10.1097/00006534-199411000-00003. K. Bush, O. Antonyshyn, Three-dimensional facial anthropometry using a laser surface scanner: validation of the technique, Plast. Reconstr. Surg 98 (2) (1996) 226-235, https://doi.org/10.1097/00006534-199608000-00004. C. B. Cutting, J. G. McCarthy , D. B. Karron, Three-dimensional input of body surface data using a laser light scanner, Ann. Plast. Surg. 21 (1) (1988) 38-45, https://doi.org/10.1097/00000637-198807000-00008. A. Peiravi, B. Taabbodi, A reliable 3D laser triangulation-based scanner with a new simple but accurate procedure for finding scanner parameters, J Am Sci 6(5) (2010) 80-85. P. Napolitano, F. Alimenti, P. Carbone, A novel sample-and-hold-based time-to-digital converter architecture, IEEE Trans. Instrum. Meas. 59 (5) (2010) 1019-1026, http://dx.doi.org/10.1109/TIM.2010.2044078. E. Cuesta, J. C. Rico, P. Fernández, D. Blanco, G. Valiño, Influence of roughness on surface scanning by means of a laser stripe system, Int. J. Adv. Manuf. Technol 43(11) (2009) 1157-1166, https://link.springer.com/article/10.1007/s00170-008-1794-9. N. Vukašinović, J. Možina, J. Duhovnik, Correlation between incident angle, measurement distance, object colour and the number of acquired points at CNC laser scanning, Stroj. Vestn./J. Mech. Eng. 58(1) (2012) 23-28, http://dx.doi.org/10.5545/sv-jme.2011.053. K. Ouji, M. Ardabilian, L. Chen, F. Ghorbel, 3D deformable super- resolution for multi-camera 3D face scanning. J. Math. Imaging Vis. 47(1) (2013) 124-137 , DOI : 10.1007/s10851-012-0399-y . Stančić, J. Musić, & V. Zanchi, Improved structured light 3D scanner with application to anthropometric parameter estimation, Measurement 46 (1) (2013) 716-726, https://ui.adsabs.harvard.edu/link_gateway/2013Meas...46..716S/doi:10.1016/j.measurement.2012.09.010. S. Martínez, E. Cuesta, J. Barreiro, & B. Álvarez, Analysis of laser scanning and strategies for dimensional and geometrical control, Int. J. Adv. Manuf. Technol 46(5) (2010) 621-629, https://link.springer.com/article/10.1007/s00170-009-2106-8. M. A. Isa, I. Lazoglu, Design and analysis of a 3D laser scanner, Measurement 111 (2017) 122-133, http://dx.doi.org/10.1016/j.measurement.2017.07.028. V. K. Pathak, A. K. Singh, Optimization of morphological process parameters in contactless laser scanning system using modified particle swarm algorithm, Measurement 109 (2017) 27-35, https://ui.adsabs.harvard.edu/link_gateway/2017Meas..109...27P/doi:10.1016/j.measurement.2017.05.049. F. Schwarz-Müller, R. Marshall, S. Summerskill, C. Poredda, Measuring the efficacy of positioning aids for capturing 3D data in different clothing configurations and postures with a high-resolution whole-body scanner, Measurement 169 (2021) 108519, https://doi.org/10.1016/j.measurement.2020.108519. Lartigue, A. Contri, P. Bourdet, Digitised point quality in relation with point exploitation. Measurement 32(3) (2002) 193- 203, http://dx.doi.org/10.1016/S0263-2241(02)00008-8. Contri, P. Bourdet, C. Lartigue, Quality of 3D digitised points obtained with non-contact optical sensors, CIRP Ann Manuf Technol 51(1) (1999) 443- 446, http://dx.doi.org/10.1016/S0007-8506(07)61556-X. M. Mahmud, D. Joannic, M. Roy, A. Isheil, J. F. Fontaine, 3D part inspection path planning of a laser scanner with control on the uncertainty. Comput. Aided Des, 43 (4) (2011) 345-355, http://dx.doi.org/10.1016/j.cad.2010.12.014. H. Y. Feng, Y. Liu, F. Xi, Analysis of digitizing errors of a laser scanning system, Precis. Eng. 25 (3) (2001) 185-191, http://dx.doi.org/10.1016/S0141-6359(00)00071-4. N. Van Gestel, S. Cuypers, P. Bleys, J. P. Kruth, A performance evaluation test for laser line scanners on CMMs, Opt Lasers Eng . 47 (3-4) (2009) 336-342, https://doi.org/10.1016/J.OPTLASENG.2008.06.001. E. Sanders, & S. Fatone, Residual limb volume change: systematic review of measurement and management, J . Rehabil. Res. Dev. 48(8) (2011) 949, https://doi.org/10.1682/jrrd.2010.09.0189 Y. Yu , W. Zhang, Y. Deng, Frechet Inception Distance (FID) for Evaluating GANs, 2021. Oala, L., Murchison, A. G., Balachandran, P., Choudhary, S., Fehr, J., Leite, A. W., & Wiegand, T. (2021). Machine learning for health: algorithm auditing & quality control. Journal of medical systems , 45 (12), 1-8. M. Mourabet, A. El Rhilassi, H. El Boujaady, M. Bennani-Ziatni, A. Taitai, Use of response surface methodology for optimization of fluoride adsorption in an aqueous solution by Brushite, Arab. J. Chem., (2017) 10 S3292-S3302. R. U. Owolabi, M. A. Usman, & A. J. Kehinde, (2018). Modelling and optimization of process variables for the solution polymerization of styrene using response surface methodology, J. King Saud Univ. Eng. Sci 30 (1) 22-30, https://doi.org/10.1016/j.jksues.2015.12.005. V. Kumar, A. Kumar, D. Chhabra, P. Shukla, Improved biobleaching of mixed hardwood pulp and process optimization using novel GA-ANN and GA-ANFIS hybrid statistical tools, Bioresour. Technol. 271 (2019) 274-282, https://doi.org/10.1016/j.biortech.2018.09.115. M. Yadav, D. Yadav, R. K. Garg, R. K. Gupta, S. Kumar, & D. Chhabra, Modeling and Optimization of Piezoelectric Energy Harvesting System Under Dynamic Loading, Advances in Fluid and Thermal Engineering (2021) 339-353, https://www.springerprofessional.de/en/modeling-and-optimization-of-piezoelectric-energy-harvesting-sys/19092572. A. Sharma, D. Chhabra, R. Sahdev, A. Kaushik, U. Punia, Investigation of wear rate of FDM printed TPU, ASA and multi-material parts using heuristic GANN tool. Materials Today: Proceedings.(2022), https://doi.org/10.1016/j.matpr.2022.04.015. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-2051093","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":136078934,"identity":"502ae2c7-b919-4fee-a492-fc024d6c2de0","order_by":0,"name":"Ashish Kaushik","email":"","orcid":"","institution":"Deenbandhu Chhotu Ram University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ashish","middleName":"","lastName":"Kaushik","suffix":""},{"id":136078936,"identity":"f7ce468a-34d9-4d20-bc89-dbbab2c4ed7f","order_by":1,"name":"Upender Punia","email":"","orcid":"","institution":"Deenbandhu Chhotu Ram University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Upender","middleName":"","lastName":"Punia","suffix":""},{"id":136078940,"identity":"2df74201-4ae3-4388-991d-87ca1fdb97a0","order_by":2,"name":"Ramesh Kumar Garg","email":"","orcid":"","institution":"Deenbandhu Chhotu Ram University of Science and Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ramesh","middleName":"Kumar","lastName":"Garg","suffix":""},{"id":136078950,"identity":"bc134a16-8184-4bf0-94ce-0ead32b85106","order_by":3,"name":"Mohit Yadav","email":"","orcid":"","institution":"Maharshi Dayanand University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Mohit","middleName":"","lastName":"Yadav","suffix":""},{"id":136078954,"identity":"07c3663c-9b57-4627-9c6c-15ab54cbecd8","order_by":4,"name":"Rajat Vashistha","email":"","orcid":"","institution":"University of Queensland","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Rajat","middleName":"","lastName":"Vashistha","suffix":""},{"id":136078959,"identity":"b0b25701-ad42-4de7-ba13-df2074fc360c","order_by":5,"name":"Mannu Rathee","email":"","orcid":"","institution":"Pandit Bhagwat Dayal Sharma University of Health Sciences","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Mannu","middleName":"","lastName":"Rathee","suffix":""},{"id":136078962,"identity":"824a884d-dc9d-414f-b427-eea5f91bfd99","order_by":6,"name":"Ravinder Kumar Sahdev","email":"","orcid":"","institution":"University Institute of Engineering and Technology, MDU","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ravinder","middleName":"Kumar","lastName":"Sahdev","suffix":""},{"id":136078965,"identity":"ed8cc5d3-24e7-4048-9279-abaadb351460","order_by":7,"name":"Deepak Chhabra","email":"data:image/png;base64,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","orcid":"","institution":"University Institute of Engineering and Technology, MDU","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Deepak","middleName":"","lastName":"Chhabra","suffix":""}],"badges":[],"createdAt":"2022-09-10 08:14:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2051093/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2051093/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":26474952,"identity":"d07d8175-1cce-42ea-8326-2dfd4af4cf86","added_by":"auto","created_at":"2022-09-14 21:20:56","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":112513,"visible":true,"origin":"","legend":"\u003cp\u003eScanning software feature for recognition\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/2746ea8488f179863549292e.png"},{"id":26475359,"identity":"3cc477de-0ca0-472a-8e2b-72004cf51d6e","added_by":"auto","created_at":"2022-09-14 21:25:56","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":256497,"visible":true,"origin":"","legend":"\u003cp\u003eReal-time capturing of human faces\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/7b2c6894b87e01a596f3e66c.png"},{"id":26474950,"identity":"a639edc2-9125-4475-a3f9-568e63faf80e","added_by":"auto","created_at":"2022-09-14 21:20:56","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":96239,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentation of Tools for Scan Processing\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/6e8559d47ebd349b50aba2b6.png"},{"id":26474951,"identity":"270bf4be-ff19-45a3-9529-2dfc3b23cf31","added_by":"auto","created_at":"2022-09-14 21:20:56","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":69805,"visible":true,"origin":"","legend":"\u003cp\u003eColour tool in sense 3D software\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/3bd83f3c2241554a1149d685.png"},{"id":26474954,"identity":"1b07e5f3-cdc3-4f59-a339-2ed60f72519d","added_by":"auto","created_at":"2022-09-14 21:20:56","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":28622,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/b4f7e1156425b5e2a67ed2a7.png"},{"id":26475360,"identity":"c623999c-6c63-4b77-bc08-4fdb17482a68","added_by":"auto","created_at":"2022-09-14 21:25:56","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":508098,"visible":true,"origin":"","legend":"\u003cp\u003eModels of scanned human faces\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/c36ad13d6744508b3c5b42f2.png"},{"id":26474956,"identity":"0847c61a-1b83-4f4a-9fb1-2242eee7d459","added_by":"auto","created_at":"2022-09-14 21:20:56","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":36870,"visible":true,"origin":"","legend":"\u003cp\u003eFID score variations\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/2151027ea6bb1b597a2afab8.png"},{"id":26474958,"identity":"efc83abd-f0b4-4079-b483-d4ed6b5bac4a","added_by":"auto","created_at":"2022-09-14 21:20:56","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":240104,"visible":true,"origin":"","legend":"\u003cp\u003eANN structure representation\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/fd3cdcdcd192803e1a3c0296.png"},{"id":26474962,"identity":"f25fbc00-ee35-4f34-8c10-e96b891d6489","added_by":"auto","created_at":"2022-09-14 21:20:57","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":39317,"visible":true,"origin":"","legend":"\u003cp\u003eIllustration of training and validation states for ANN\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/7112703e42886bf11dc867b5.png"},{"id":26476083,"identity":"e2c6fa42-e36a-4121-b635-d225505a8924","added_by":"auto","created_at":"2022-09-14 21:30:56","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":106524,"visible":true,"origin":"","legend":"\u003cp\u003eANN- based correlation graph depicting training, validation, testing and overall results\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/062cf07d2cab97354bd0e52c.png"},{"id":26475358,"identity":"4e9680ce-b50a-4907-b2f0-7068772ffe13","added_by":"auto","created_at":"2022-09-14 21:25:56","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":39429,"visible":true,"origin":"","legend":"\u003cp\u003eConvergence of heuristic GA-ANN\u003c/p\u003e","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/0eab48e202f9fb78aa890517.png"},{"id":26474961,"identity":"a4f8906a-fc41-4207-bfc3-b205bd5c7368","added_by":"auto","created_at":"2022-09-14 21:20:56","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":123033,"visible":true,"origin":"","legend":"\u003cp\u003eMSE plot\u003c/p\u003e","description":"","filename":"floatimage12.png","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/6d5c6df8fa129c02a3a7c76a.png"},{"id":28062682,"identity":"22905ca9-f811-4ed5-80ef-607066cd75b2","added_by":"auto","created_at":"2022-10-20 20:29:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2399002,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2051093/v1/3f7b9dcf-001f-490b-bb58-db7627c448c6.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Optimization of process parameters for scanning human face using hand-held scanner","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe challenge of accurately evaluating the size and shape of the human face has always sparked the interest of both researchers and doctors. The utilization of 3D surface scanning technique to create digitized models of human anatomical parts that can assist with altering the way a huge variety of products are planned and manufactured \u003cb\u003e[1].\u003c/b\u003e One of the fundamental problems which any novel user encounters during 3D laser scanning is the choice of suitable process parameters to obtain the full scan with very few steps. There is always an option associated with input process parameters like angular orientation, relative scanner distance, and the effect of light intensity, which is not always too intuitive but also strongly influences the scan results. The demand for accurate assessment and visualization of the facial bone and other tissues increases due to advancements in dentistry, maxillo-facial and plastic surgery. Face scanning is also examined to evaluate the outcomes of facial plastic surgery \u003cb\u003e[2].\u003c/b\u003e\u003c/p\u003e \u003cp\u003e3D Scanning is the technique of detailed analysis of real-world capture to gather data on its specifications related to dimensions appearance. From the past few years, the majority of reports are published on the utilization of 3D scanners in the medical, dental, and healthcare sectors is considerably increased, which might prove helpful for plastic surgery\u003cb\u003e[3\u0026ndash;7]\u003c/b\u003e. Scanning technology during its infancy period is limited to industrial applications as the scanning of the human face requires a sequence of specific conditions compared to industrial objects\u003cb\u003e[4]\u003c/b\u003e. Various studies focused on improving the performance and optimization of process parameters of handheld scanners studying the face of person, including a progression of exceptional circumstances. As the face of living person can't be immobilized, the scanner must feature a brief recording span. Hence a handheld 3D scanner is used for this purpose.\u003c/p\u003e \u003cp\u003eMoreover, a portable handheld scanner can be moved in various directions and enables us to scan from different orientations. A study has been performed to create a laser triangulation 3D scanner whose sole purpose is to solve the limitation of occlusion. It uses two distinct laser colors, namely green and red, along with a charge coupled device camera (CCDC) and helps enhance the reliability of respective system\u003cb\u003e[8].\u003c/b\u003e Several research focused on design of a time-to-digital converter and measurement of time of flight (ToF), in order to utilize same in portable scanners\u003cb\u003e[9].\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe dependency of ambient lighting as an essential factor in influencing the quality of the captured signal by Charge Couple Devices (CCD), and the impact of surface roughness the object during measurement was also recorded in few studies. Optimum results are obtained in case of absence or limiting ambient lighting conditions \u003cb\u003e[10].\u003c/b\u003e The research was performed to worked with the diffuse reflections of the object\u0026rsquo;s surface being scanned and an effective method to restrict the outcome of ambient illumination by implementing the distinct light filters upon the CCD sensors\u003cb\u003e[11]\u003c/b\u003e. Moreover, a general solution to use the coating spray for covering the object by matty white layer is also introduced. The effects of other influential factors such as incident angle, distance from the scanner, and object color are also studied on a Computer Numerically Controlled (CNC) laser scanning process \u003cb\u003e[12]\u003c/b\u003e. A primary structured light pattern for 3D structured light scanner is implemented in a research, during development, the suggested system's accuracy and resilience were evaluated on artificial items with established surface geometry, followed by assessments on human individuals [\u003cb\u003e13\u003c/b\u003e].\u003c/p\u003e \u003cp\u003eSeveral researches indicated the problems in data acquisition of dark, translucent, and glossy surfaces. Generally, in the case of shining surfaces, noise is eventually added during the measurement, and the points exclusive to the actual surface can also be provoked. In both instances, sensors cannot achieve data locally (lost data), or the scan measurement is adversely affected [14]. These shiny surfaces are covered with coating sprays that help scan the objects' dark, glossy, and translucent surfaces. It was found that for 5\u0026ndash;15 micrometers, the respective additional thickness and variation is about 45 micrometers using comparatively thin coating sprays. A low cost scanner with a hemispherical workspace has been designed and implemented with least square minimization approach to realign the parameters for achieving optimized results[15]. Further, a modified particle swarm technique, to optimize the significant morphological parameters in a contactless laser scanning method in various research \u003cb\u003e[16].\u003c/b\u003e Few of them investigate the effectiveness of positioning aids for obtaining 3D data in various clothing postures and configurations, through a superior quality body scanner \u003cb\u003e[17]\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eThe quality of digitized points with the help of 3D laser scanning is evaluated by several criteria, namely density, completeness, noise, and accuracy \u003cb\u003e[18,19]\u003c/b\u003e. Several types of research are published regarding the performance of 3D scanners at various surface-to-scanner comparative orientations, and altogether examined scanners in normal position w.r.t the surface as the best condition \u003cb\u003e[20\u0026ndash;22].\u003c/b\u003e\u003c/p\u003e \u003cp\u003eA handheld, cost-effective 3D laser scanner was used to scan a human face under three significant input factors, i.e., scanning distance, angular orientation, and light intensity. A combination of input factors for twenty experiments has been designed based on face-centered central composite design. Accordingly, twenty CAD models have been retrieved on the twenty combinations of input factors. The accuracy of scan models is investigated through the Frechet Inception Distance (FID) score, which is taken as output. A model has been trained among input and output using a neural network, and further, it is optimized using a genetic algorithm.\u003c/p\u003e"},{"header":"Materials And Methods","content":"\u003cp\u003eTo test new scanning technologies, it has been suggested that high accuracy and resolution laser scanners be used \u003cstrong\u003e[23]\u003c/strong\u003e. 3D Sense (3D system, Rock Hill, SC/USA) is a minimal expense, a convenient 3-dimensional surface scanner with, as per the maker, an exactness of near millimeters with resolution at Scanning range exists between 177.8 and 1828.8mm along with the color resolution of 1920 x1080 pixels. In contrast, the range of operation lies between 0.2 to 1.6 m. The field of view:45\u0026deg; horizontally, 57.5\u0026deg; in the vertical direction, and 69\u0026deg; diagonally with the physical configuration of the scanner is 129 (w) x 179.8(h) x 33(d) mm and maximal image of 30 /fps.\u003c/p\u003e\n\u003cdiv\u003e\n \u003ch2\u003e3.1 Proposed method of scanning\u003c/h2\u003e\n \u003cdiv\u003e\n \u003ch2\u003e3.1.1 Scanning the person\u003c/h2\u003e\n \u003cp\u003eTo capture the scan, get a person seated in normal posture and directs the scanner in their general direction. Although it is possible to scan virtually anything, clothing and accessories can still affect the scanning, so avoid wearing brighter and darker clothes. If the materials are difficult to scan, the scan can be performed by increasing the sensitivity. Also, spectacles or sunglasses can affect the texture, so these should be removed before scanning. Enhanced mesh quality, stabilized tracking, and proper orientation of trimming plane can be obtained by the Object Recognition feature, which includes object, body, and head. Figure 1 represents the object recognition features available in the SENSE software.\u003c/p\u003e\n \u003cp\u003eThe green color in our preview indicates either the presence of data within the scan volume representing data build-up or correct distance from proper focus, so our aim should be to adjust the sensor\u0026apos;s reach to get the subject in the green zone. To begin with, firstly focus lies in capturing the face of the person for optimal trajectory and should keep away from capturing the face further as they turn around. Start the scan from ear to nose, lower the scanner to capture the chin from below, and further raise it to scan the area above the forehead. Return the scanner to nose level and then scan the face by moving it to the second ear. After completing the face-scanning, we moved towards hair scanning. Remember to scan from all the possible angles. The person sitting on a rotating or swivel table will spin slowly. This may take several seconds to several minutes, so the person must be in the proper posture to avoid geometry distortion during the scanning. Keep the shoulders and back in the scanner field of view. Attention is mainly focused on the target in green color for accurate capturing. If the tracking is lost, return the scanner to its previous position or the person to its last orientation to recover the lost scan data. The scanner will continue to scan further once it finds its left-off place. As the person completes one turn, the scanner should stop scanning. Experimental setup for real-time capturing of human faces is represented in Fig. 2\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv\u003e\n \u003ch2\u003e3.1.2 Processing the scan\u003c/h2\u003e\n \u003cp\u003eAt this stage, some preliminary processing to reduce the number of faces and colorize the scan is performed to simplify the geometry so that it is easy to work within the domain of mesh software. Figure 3 indicates the several tools required during the preliminary processing of scan models.\u003c/p\u003e\n \u003cp\u003eOnce the scan is completed, several tools are available to make any modifications in the scan, such as:\u003c/p\u003e\n \u003col style=\"list-style-type: lower-alpha;\"\u003e\n \u003cli\u003eCrop to remove any excess scan data captured during the scan\u003c/li\u003e\n \u003cli\u003eTrim,\u0026nbsp;which\u0026nbsp;is\u0026nbsp;like\u0026nbsp;crop,\u0026nbsp;but\u0026nbsp;the\u0026nbsp;difference\u0026nbsp;lies\u0026nbsp;in\u0026nbsp;the\u0026nbsp;fact\u0026nbsp;that\u0026nbsp;smaller\u0026nbsp;of\u0026nbsp;two\u0026nbsp;sections\u0026nbsp;obtained\u0026nbsp;by\u0026nbsp;trim\u0026nbsp;line\u0026nbsp;is removed\u0026nbsp;in\u0026nbsp;case of\u0026nbsp;trimming,\u003c/li\u003e\n \u003cli\u003eColor - The\u0026nbsp;brightness\u0026nbsp;and contrast of the model can\u0026nbsp;be\u0026nbsp;adjusted\u0026nbsp;by the Color\u0026nbsp;tool, and\u0026nbsp;a\u0026nbsp;preview\u0026nbsp;is\u0026nbsp;obtained\u0026nbsp;after\u0026nbsp;making\u0026nbsp;the\u0026nbsp;final\u0026nbsp;adjustments\u0026nbsp;before\u0026nbsp;applying.\u0026nbsp;Fig. 4 describes\u0026nbsp;the\u0026nbsp;color\u0026nbsp;tools\u0026nbsp;used\u0026nbsp;in\u0026nbsp;SENSE\u0026nbsp;software\u0026nbsp;for\u0026nbsp;regulating\u0026nbsp;the\u0026nbsp;image\u0026nbsp;sensitivity\u0026nbsp;and\u0026nbsp;contrast\u0026nbsp;of\u0026nbsp;objects.\u003c/li\u003e\n \u003cli\u003eErase tool to remove or eliminate unwanted scan portions, which can be performed by moving the cursor over the unwanted scan portion or by mouse button, and the unwanted scan will be removed.\u003c/li\u003e\n \u003c/ol\u003e\n \u003c/div\u003e\n \u003cdiv\u003e\n \u003ch2\u003e3.1.3 Exporting the scan\u003c/h2\u003e\n \u003cp\u003eSeveral file formats are available to export the raw data obtained during scanning. However, the STL format is the most widely used for exporting data because of the ubiquity of its use.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv\u003e\n \u003ch2\u003e3.1.4 Edit and evaluation of data\u003c/h2\u003e\n \u003cp\u003eData exported in the above stage consists of some noise unnecessary data errors in mesh surfaces such as tiny openings or some areas with faulty normal and needs to be repaired or removed before further processing.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv\u003e\n \u003ch2\u003e3.2 Frechet inception distance (FID)\u003c/h2\u003e\n \u003cp\u003eThe FID is a standard metric for assessing the worth of produced images and, more explicitly, created to estimate the execution of productive adversarial structure \u003cstrong\u003e[24]\u003c/strong\u003e. Martin used FID score technique in the study that suggest a two time- scale update rule (TTUR) for training generative adversarial network(GANs) by stochastic approximation. It has been noticed that the convergence of TTUR is directly depend on the assumption of local nash equilibrium. The mention score has been advancement above the prevailing inception score, abbreviated as IS. Inception score determines the condition of an accumulation of synthetic images centered on the top performance of the image. The scores incorporate equally the certainty of the restricted class forecasts for every manufactured picture and basic of marginal probability of forecasted classes. This score doesn\u0026apos;t compare how synthesized pictures contrast with authentic pictures. The mean to foster FID score was to assess synthetic pictures fixated on the insights of gathering engineered images recognized to measure an assortment of original images from the objective space. The goal of machine learning (ML) technology is to make medical procedures relatively easy, efficient, and superior.. [\u003cstrong\u003e25\u003c/strong\u003e].\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Experimental Design Matrix","content":"\u003cp\u003eThe statistical technique of design of experiments (DoE) is implemented in the present work which optimize the systems performance through previously known input variables. The process\u003c/p\u003e\n\u003cp\u003egenerally evaluates the effect of combination of several factors at a time which is also refereed as interactions using factorial designs. The experimental table is derived using face centered composite design approach. Figure\u0026nbsp;5 clearly demonstrates the flow chart of the proposed work which is completed in five phases. The raw data collected in phase I is used for creating the experimental design matrix using DoE methodology. All the experiments as uggested by DoE with combination of input factors are now performed in phase 3. Data can be analysed and corresponding models are created in phase IV followed by model validation in subsequent stage. Aforesaid phases in this research are further discussed in detail.\u003c/p\u003e\n\u003cdiv class=\"Section2\" id=\"Sec10\"\u003e\n \u003ch2\u003e4.1 Phase 1: Collection of raw data\u003c/h2\u003e\n \u003cp\u003eRaw data is collected to set a feasible range of significant input parameters required to optimize. This research mainly needs to optimize three input parameters for scan models.\u003c/p\u003e\n \u003cp\u003eThe parameters mentioned below are taken into consideration:\u003c/p\u003e\n \u003col style=\"list-style-type: lower-roman;\"\u003e\n \u003cli\u003eAngle (A) - corresponds to the angular orientation of global rotation.\u003c/li\u003e\n \u003cli\u003eAmbient Lightning Conditions (I) - in indoor surroundings.\u003c/li\u003e\n \u003cli\u003eScanning distance (D) is the relative distance of the scanner from the object\u0026apos;s orientation.\u003c/li\u003e\n \u003c/ol\u003e\n \u003cp\u003eFor consideration, scanning distance lies between 18 inches to 28 inches, angular variation lies between 45 degrees to 90 degrees, and light intensity lies in 12 watts per meter square to 18 watts per meter square. Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e showcase the variables assigned as A,B,C with their range intervals. \u0026nbsp;\u003c/p\u003e\n \u003ctable border=\"1\" id=\"Tab1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eSignificant process parameter with their range in central composite design\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eProcess Parameter\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eRange in face centered central composite design\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eA-Scanning Distance (in inches)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e28\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eB-Angular Variation (in degrees)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eC-Light Intensity\u003c/p\u003e\n \u003cp\u003e(in Watt/meter square)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003cp\u003e\u003c/p\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Section2\" id=\"Sec11\"\u003e\n \u003ch2\u003e4.2. Phase 2: Creating Design Table\u003c/h2\u003e\n \u003cp\u003eA method of Design of Experiments (DoE) is determined according to the objectives of experiments. Face centered composite design approach was adopted to analyze the influence of operating variables on accuracy of scan models. In order to achieve the optimum conditions, three process factors were taken in to account i.e. scanning distance(A), angular orientation (B), and light intensity(C). The present approach includes 2\u003csup\u003en\u003c/sup\u003e factorial runs, n\u003csub\u003ec\u003c/sub\u003e center runs and 2n axial runs. The reproduction of data and experimental error are controlled by center points \u003cstrong\u003e[26]\u003c/strong\u003e. The total number of experimental runs that are required to perform can be obtained using Eq. (1) \u003cstrong\u003e[27].\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eN\u0026thinsp;=\u0026thinsp;2\u003csup\u003en\u003c/sup\u003e\u0026thinsp;+\u0026thinsp;2n\u0026thinsp;+\u0026thinsp;n\u003csub\u003ec\u003c/sub\u003e (1)\u003c/p\u003e\n \u003cp\u003eHere n represents the number of process factor, N denotes the total number of runs and n\u003csub\u003ec\u003c/sub\u003e represents the center point number. The present approach specifies that twenty experimental runs are needed for the process of optimization including six axial experiments, eight factorial experiments, and six center experiments. Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e demonstrates the CCD approach adopted in this proposed work.\u003c/p\u003e\n \u003ctable border=\"1\" id=\"Tab2\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eInput Parameters Obtained by Using DoE\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eStd\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRun\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eScanning distance (D in Inches)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eScanning angle (A in degrees)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eIntensity ( I in watt/meter sq.)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19.6216\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54.1214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14.8108\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19.6216\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54.1214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17.1892\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24.3784\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54.1214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17.18192\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24.3784\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e80.876\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14.8108\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19.6216\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e80.8786\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14.8108\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24.3784\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e80.8786\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17.1892\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19.6216\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e80.8786\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17.1892\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24.3784\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54.1214\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14.8108\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Section2\" id=\"Sec12\"\u003e\n \u003ch2\u003e4.3 Phase 3: Experiments\u003c/h2\u003e\n \u003cp\u003eTwenty scans are conducted according to the values of input parameters suggested by Design of Experiments software. The obtained scan images are the results of the respective set of input values presented by software, as depicted in Fig. 6 as model M1, M2, M3 and so on.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Results And Discussion","content":"\u003cp\u003eTwenty CAD models were retrieved according to the combination of significant factors for twenty experimental runs based upon the combination of significant input parameters ie. scanning distance, light intensity and angular orientation. Accuracy evaluation of retrieved models helps us in identification of suitable model or provides a scope for further improvement in models. The performance metric adopted for evaluating accuracy of scan models is FID as discussed above. Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. represents the FID score of scan models with their corresponding parameters. Model M11 is the best-suited model according to the FID technique.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFID Scores of Scan Models\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRun\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eScanning distance (d in inches)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eScanning angle (A in degrees)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eIntensity ( I in watt/meter sq.)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFID score\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19.6216\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e54.1214\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.8108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e333.87\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e310.978\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e375.576\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19.6216\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e54.1214\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.1892\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e347.023\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e274.891\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e346.22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24.3784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e54.1214\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.18192\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e297.061\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e313.939\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24.3784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80.876\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.8108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e290.511\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e271.894\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e270.24\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e273.637\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19.6216\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80.8786\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.8108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e298.579\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e334.062\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24.3784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80.8786\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.1892\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e284.83\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e319.741\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e278.021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19.6216\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80.8786\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.1892\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e322.404\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24.3784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e54.1214\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14.8108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e332.678\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e272.361\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e7\u003c/span\u003e depicts the variation of the accuracy of scan models concerning FID score where scan models are depicted on the X-axis and FID score is plotted on Y-axis of cartesian coordinates. The curve indicates variation of scan models to FID score\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Artificial neural network\u003c/h2\u003e \u003cp\u003eThe mechanism of the brain serves as the basis for neural networks. .The neural structure is made up of an arranged layer of synthesized neurons which are linked to all remaining artificial neurons using a coefficient (weight function) particularly known as the element of processing. Weighted information is used in the neural network model. First-layer neurons acquire weighted information in the neural network. The solution to the problem is determined based on the data generated by the most recent layer of neurons as well as the hidden layer that lies underlying it. Across numerous training procedures, the inter-unit connectors were optimized until the prediction error reached a minimum value; evidently, the system obtained the precise value. The ANN can foresee instances that haven't been shown to the system before owing to generalization. The approach of FCCD developed a design matrix (shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) with three input factors and one experimental output result that has a non-linear relationship and is utilized for training, testing, and validation. The root mean square error and related correlation coefficients were used as assessing parameters for the ANN model with different layouts, which was trained with 70% of the data, tested with 15% of the data, and validated with 15% of the data \u003cb\u003e[28].\u003c/b\u003e The error is calculated through forward propagation by first putting all of the inputs together, then multiplying arbitrary weights and biases, and finally applying this calculation to the output of the tangential sigmoid activation function. During backward propagation of the error with updated weight, the method of gradient descent was utilized, resulting in a feed-forward back propagation network (FFBP)\u003cb\u003e[29\u0026ndash;30].\u003c/b\u003e Although more computing is required, the guided algorithm Levenberg Marquardt (LM) back propagation was employed since it is the precise and most feasible of all algorithms.\u003c/p\u003e \u003cp\u003eThe FCCCD presented a design matrix with three input parameters (Scanning Distance, Angular orientation, Light intensity) and one experimental output responses (FID Score) that have a nonlinear relationship and have been utilized for training, testing, and validation. Forward propagation estimates the inaccuracy by adding together inputs, multiplying weights, and applying preferences to the activation function of the tangent sigmoid. During backward propagation, the gradient descent approach was employed for the error with updated weight generating feed-forward back propagation (FFBP) network. This ANN generated a model based on structure, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e8\u003c/span\u003e that was trained with 20 sets of three input variables and one output responses, and the training state for the ANN model is demonstrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e9\u003c/span\u003e, at epoch 6.\u003c/p\u003e \u003cp\u003eThe controlled algorithm Levenberg Marquardt (LM) backpropagation yielded the highest overall value of R (0.95), as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e10\u003c/span\u003e. This approach was chosen because it is the best ever and most stable of all other algorithms however, more processing is needed. At the sixth iteration of training, the input data stretches its highest optimal solution, and when the mean square error (MSE) of evidence samples starts to climb, the epochs involuntarily stop, as depicted in the MSE graph in Fig.\u0026nbsp;12. The finest performance for validation at epoch 0 was 0.034044.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Heuristic tool for optimization in genetic algorithms and artificial neural networks (GA-ANN)\u003c/h2\u003e \u003cp\u003eThe goal is to optimize accuracy by minimising FID score, and the model built using the ANN heuristic tool is executed as a fitness function in the genetic algorithm (GA-ANN).\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. displays the Genetic algorithm Heuristic tool's parameters. The population size is the numeral of chromosomes produced at arbitrary in a single iteration to examine the output response by using a fitness function. Mutation and Crossover functions create children from parents, leading in the development of innovative chromosomes to determine the best output. The population range can be any value that yields the largest chromosomes inside a particular bound. A population size of 50 was selected for optimization. The suggested approach makes use of a constraint dependent as crossover and mutation function. The crossover percentage has been fixed to 0.8, whereas rate of mutation has been set as 0.01. The number of chromosomes which best match is carried over to the following generation is known as the elite count.. In the proposed work, the value of elite count is taken to be 2.5. Genetic algorithms fitness function is based on the ANN model. The results of the GA are presented in Table\u0026nbsp;8, and they include the best input values that ensure the output response is balanced. The model was further validated by conducting confirmative experiments using the GA-ANN's optimum arrangement sets of taken input process parameters.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eHeuristic tool genetic algorithm parameters\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameters of GA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eParameters of GA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo. of input variables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMutation rate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo. of output responses\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMutation function\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eConstraint dependent\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePopulation size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStopping criteria\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo. of generations\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCrossover fraction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eElite count\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCross over function\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eConstraint dependent\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFitness function\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eANN model\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTo verify that the results are accurate, each set of trials is repeated three times, and the average of those values is calculated. The experimental and expected values were quite similar in the confirmation experiments.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe proposed research work has proficiently explored the methods for data acquisition and retrieving the CAD models through reverse engineering/scanning. The proposed study demonstrated that using handheld scanner with proper optimized process parameters will result in a valid and accurate models /data acquisition of irregular surfaces like human face. The current research successfully examined the impact of significant input parameters that predominately affect the scanning of human faced. The best-achieved accuracy was found for experimental model M11 with the lowest FID score, i.e., 270.24, obtained with a scanning distance of 22 inches, the angular orientation of 67.5 degrees, and ambient lightning condition of 16 watt/meter square. Parametric optimization is achieved with the help of hybrid GA-ANN for improving accuracy of scan models using FID score. The accuracy is maximized by decreasing the FID score to 186.3569 through heuristic GA-ANN tool having 28 inches as scanning distance, 48.041 degrees as angular orientation, and 18 watt/meter square as the ambient lighting condition. In summary, the proposed method using handheld 3D laser scanners for retrieving human face can be implemented in clinical utilities, craniofacial research, pre surgical treatment, and assessment of soft tissues.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompliance with Ethical Standards:\u003c/strong\u003e\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eDisclosure of potential conflicts of interest :\u0026nbsp;The author hereby confirms that there are no potential conflicts of interest.\u003c/li\u003e\n \u003cli\u003eResearch involving Human Participants and/or Animals: The research involves a person voluntarily interested in face scanning. No animals are involved during research.\u003c/li\u003e\n \u003cli\u003eInformed consent :Not applicable\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003eEthical Approval and Consent to Participate:\u003c/strong\u003e The study requires no ethical approval. Face of a male colleague has been scanned who willingly participate during research.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHuman Ethics\u003c/strong\u003e \u003cstrong\u003e:\u003c/strong\u003e Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication:\u003c/strong\u003e Yes, all the authors and other person who are involved during the \u0026nbsp; \u0026nbsp; research are agreed to permission for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Supporting Data\u003c/strong\u003e \u003cstrong\u003e:\u0026nbsp;\u003c/strong\u003eNot appropriate.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests:\u003c/strong\u003e Authors disclose that there is no competing interest of proposed work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e: There is no funding/any grant for the proposed research.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; Contributions :\u003c/strong\u003e\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eConceptualization and Supervision: Ramesh Kumar Garg, Mannu Rathee, Ravinder Kumar Sahdev.\u003c/li\u003e\n \u003cli\u003eDraft writing, figures: Ashish Kaushik, Upender Punia.\u003c/li\u003e\n \u003cli\u003eData contribution and tool analysis: Mohit Yadav, Rajat Vashistha, Deepak Chhabra\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthor sincerely acknowledges CAD/CAM \u0026amp; Additive manufacturing lab of MD University for providing tools/equipments facilities.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; Information\u003c/strong\u003e\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003eAshish Kaushik\u003c/strong\u003e graduated from M.D. University, Rohtak, Haryana, India in 2018 with a B.Tech. and an M.Tech. in Mechanical Engineering. He is currently enrolled in the Department of Mechanical Engineering at the DCRUST Murthal, India, for the purpose of obtaining his doctorate degree.. His area of research are Additive Manufacturing, Biomaterials, Scanning, Reverse Engineering, Designing and Optimization.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eUpender Punia\u003c/strong\u003e graduated from GJUST, Hisar, with a B.Tech in Mechanical Engineering in 2016 and from M.D. Univetrsity Rohtak, Haryana, with an M.Tech in 2020.\u0026nbsp;He is in the process of getting his Ph.D. in Mechanical Engineering from India\u0026apos;s DCRUST, Murthal. His area of research are Additive Manufacturing, Biomaterials, Scanning, Reverse Engineering, Designing and Optimization.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003eDr. Ramesh Kumar Garg\u003c/strong\u003e is currently employed as a Chairman and Professor in the Department of Mechanical Engineering at the DCRUST, Murthal, India\u003cstrong\u003e.\u003c/strong\u003e More than sixty of his research papers have been published in either International or National Journals, and he has been visited to a number of nations as part of an academic programme. Among his scientific interests include production, system design, multi-criteria decision-making methods, and industrial engineering.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eMohit Yadav\u0026nbsp;\u003c/strong\u003ehas obtained the Masters of science \u0026amp; Master of philosphy degree in Applied Sciences, Mathematics from M.D. University, Rohtak, Haryana, India. He is continuing to pursue a doctorate in Mathematics at the University Institute of Engineering \u0026amp; Technology. His areas of research are Fluid Mechanics, Renewable Energy Harvesting, Mathematical Modeling, Simulation and Optimization.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eRajat Vashistha\u003c/strong\u003e earned his Master of Technology in Mechanical Engineering with a speciality in biomechanics in 2018\u003cstrong\u003e.\u003c/strong\u003e His study focuses on the implementation of AI-based techniques. A CIBIT scholarship was awarded to him in recognition of the merit of his MPhil project, which involved the implementation of deep learning to medical imaging in order to detect brain tumours at a preliminary phase. At Vigyan Prasar, an independent institute of the Department of Science and Technology of the Government of India, Rajat worked as a Senior Project Officer before joining CIBIT. He also served as an Assistant Professor at NIMS University Rajasthan before that.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDr. Manu Rathee\u0026nbsp;\u003c/strong\u003eis the Head of the Department of Prosthodontics at the P.G.I of Dental Sciences at the P.B.D.S University of Health Sciences in Rohtak, Haryana. He is also serving as a Senior Professor there. Co-PI has experience of working in the field of additive manufacturing and has conducted/guided original research on 3D printing in dentistry via in-vivo and in-vitro study designs. She has conducted a randomized controllable clinical testing on comparative assessment of the consequence of conventionally fabricated versus 3D printed provisional restorations for fixed dental implant prosthesis and randomized controlled clinical trial on comparative evaluation of the effect of CAD/CAM versus 3D printed provisional restorations for fixed dental implant prosthesis. Another original research with in-vitro study design was conducted on comparative evaluation of fracture resistance of anterior provisional restorations fabricated using conventional, CAD/CAM and 3D printed techniques. She has published various articles including original research, case reports and review articles in peer reviewed indexed National and International Journals. She has successfully delivered various intraoral fixed and removable prosthesis using 3D printing technology including single unit as well as and multiple unit fixed prosthesis.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDr. Ravinder Kumar Sahdev\u003c/strong\u003e is currently serving as an Assistant Professor in Mechanical Engineering Department, UIET, M.D. University, Rohtak. He has published more than 35 publications in reputed Conferences and Journals. He has completed one research project. His major areas of research areas are Additive Manufacturing, Solar Thermal, Heat Transfer, Drying, Solar Energy Conversion, Drying Technology, Grain Drying, Solar Energy, PVT \u0026amp; PCM.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDr. Deepak Chhabra\u003c/strong\u003e is employed at M.D University Rohtak as an Assistant Professor in the Mechanical Engineering Department of UIET, India. More than Seventy of their scientific articles have been published in International or National Journals or as Book Chapters by Elsevier, and he has visited to a variety of nations in preparation for International Conferences. His areas of interest in research include Mechanobiology, Active Vibration Control, 3D Printing, Finite Element Modeling, Scanning, Biomaterials \u0026amp; Optimization etc.\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eS. L. Kovacs, A. Zimmermann, G. Brockmann, M. G\u0026uuml;hring, H. Baurecht, N. A. Papadopulos, H. F. Zeilhofer, Three-dimensional recording of the human face with a 3D laser scanner, J Plast Reconstr Aesthet Surg \u003cem\u003e59\u003c/em\u003e(11) (2006) 1193-1202, https://doi.org/10.1016/j.bjps.2005.10.025.\u003c/li\u003e\n\u003cli\u003eS. C. Aung, R. C. K. Ngim, S. T. Lee, Evaluation of the laser scanner as a surface measuring tool and its accuracy compared with direct facial anthropometric measurements, Br. J. Plast. Surg.\u003cem\u003e 48\u003c/em\u003e\u003cem\u003e(8)\u003c/em\u003e (1995) 551-558,https://doi.org/10.1016/0007-1226(95)90043-8.\u003c/li\u003e\n\u003cli\u003eC. L. Istook, S. J. Hwang, 3D body scanning systems with application to the apparel industry, J. Fash. Mark. Manag.\u003cem\u003e 5(2)\u003c/em\u003e (2001) 120-132, https://www.emerald.com/insight/content/doi/10.1108/EUM0000000007283/full/pdf.\u003c/li\u003e\n\u003cli\u003eHaleem, M. Javaid, 3D scanning applications in medical field: a literature-based review, Clin. Epidemiol. Glob. Health\u003cem\u003e 7(2)\u003c/em\u003e (2019) 199-210, https://doi.org/10.1016/j.cegh.2018.05.006.\u003c/li\u003e\n\u003cli\u003eG. Bhatia, M. W. Vannier, K. E. Smith, P. K. Commean, J. Riolo, V. L. Young, Quantification of facial surface change using a structured light scanner, Plast. Reconstr. Surg.\u003cem\u003e 94(6)\u003c/em\u003e (1994) 768-774, https://doi.org/10.1097/00006534-199411000-00003.\u003c/li\u003e\n\u003cli\u003eK. Bush, O. Antonyshyn, Three-dimensional facial anthropometry using a laser surface scanner: validation of the technique, Plast. Reconstr. Surg\u003cem\u003e 98\u003c/em\u003e\u003cem\u003e(2)\u003c/em\u003e (1996) 226-235, https://doi.org/10.1097/00006534-199608000-00004.\u003c/li\u003e\n\u003cli\u003eC. B. Cutting, J. G. McCarthy , D. B. Karron, Three-dimensional input of body surface data using a laser light scanner, Ann. Plast. Surg.\u003cem\u003e 21\u003c/em\u003e\u003cem\u003e(1)\u003c/em\u003e (1988) 38-45, https://doi.org/10.1097/00000637-198807000-00008.\u003c/li\u003e\n\u003cli\u003eA. Peiravi, B. Taabbodi, A reliable 3D laser triangulation-based scanner with a new simple but accurate procedure for finding scanner parameters, J Am Sci 6(5) (2010) 80-85.\u003c/li\u003e\n\u003cli\u003eP. Napolitano, F. Alimenti, P. Carbone, A novel sample-and-hold-based time-to-digital converter architecture, IEEE Trans. Instrum. Meas. \u003cem\u003e59\u003c/em\u003e(5) (2010) 1019-1026, http://dx.doi.org/10.1109/TIM.2010.2044078.\u003c/li\u003e\n\u003cli\u003eE. Cuesta, J. C. Rico, P. Fern\u0026aacute;ndez, D. Blanco, G. Vali\u0026ntilde;o, Influence of roughness on surface scanning by means of a laser stripe system, Int. J. Adv. Manuf. Technol\u003cem\u003e 43(11)\u003c/em\u003e (2009) 1157-1166, https://link.springer.com/article/10.1007/s00170-008-1794-9.\u003c/li\u003e\n\u003cli\u003eN. Vuka\u0026scaron;inović, J. Možina, J. Duhovnik, Correlation between incident angle, measurement distance, object colour and the number of acquired points at CNC laser scanning,\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eStroj. Vestn./J. Mech. Eng.\u003cem\u003e 58(1)\u003c/em\u003e (2012) 23-28, http://dx.doi.org/10.5545/sv-jme.2011.053.\u003c/li\u003e\n\u003cli\u003eK. Ouji, M. Ardabilian, L. Chen, F. Ghorbel, 3D deformable super- resolution for multi-camera 3D face scanning. J. Math. Imaging Vis. 47(1) (2013) 124-137 , DOI : 10.1007/s10851-012-0399-y\u003cstrong\u003e.\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003e\u003cstrong\u003e \u003c/strong\u003eStančić, J. Musić, \u0026amp; V. Zanchi, Improved structured light 3D scanner with application to anthropometric parameter estimation, \u003cem\u003eMeasurement\u003c/em\u003e\u003cem\u003e46\u003c/em\u003e(1) (2013) 716-726, https://ui.adsabs.harvard.edu/link_gateway/2013Meas...46..716S/doi:10.1016/j.measurement.2012.09.010.\u003c/li\u003e\n\u003cli\u003eS. Mart\u0026iacute;nez, E. Cuesta, J. Barreiro, \u0026amp; B. \u0026Aacute;lvarez, Analysis of laser scanning and strategies for dimensional and geometrical control, Int. J. Adv. Manuf. Technol\u003cem\u003e 46(5)\u003c/em\u003e (2010) 621-629, https://link.springer.com/article/10.1007/s00170-009-2106-8.\u003c/li\u003e\n\u003cli\u003eM. A. Isa, I. Lazoglu, Design and analysis of a 3D laser scanner, \u003cem\u003eMeasurement\u003c/em\u003e\u003cem\u003e111\u003c/em\u003e (2017) 122-133, http://dx.doi.org/10.1016/j.measurement.2017.07.028.\u003c/li\u003e\n\u003cli\u003eV. K. Pathak, A. K. Singh, Optimization of morphological process parameters in contactless laser scanning system using modified particle swarm algorithm, \u003cem\u003eMeasurement\u003c/em\u003e\u003cem\u003e109\u003c/em\u003e (2017) 27-35, https://ui.adsabs.harvard.edu/link_gateway/2017Meas..109...27P/doi:10.1016/j.measurement.2017.05.049.\u003c/li\u003e\n\u003cli\u003eF. Schwarz-M\u0026uuml;ller, R. Marshall, S. Summerskill, C. Poredda, Measuring the efficacy of positioning aids for capturing 3D data in different clothing configurations and postures with a high-resolution whole-body scanner, \u003cem\u003eMeasurement\u003c/em\u003e\u003cem\u003e169\u003c/em\u003e (2021) 108519, https://doi.org/10.1016/j.measurement.2020.108519.\u003c/li\u003e\n\u003cli\u003eLartigue, A. Contri, P. Bourdet, Digitised point quality in relation with point exploitation. \u003cem\u003eMeasurement 32(3)\u003c/em\u003e (2002) 193- 203, http://dx.doi.org/10.1016/S0263-2241(02)00008-8.\u003c/li\u003e\n\u003cli\u003eContri, P. Bourdet, C. Lartigue, Quality of 3D digitised points obtained with non-contact optical sensors,\u003cem\u003e \u003c/em\u003eCIRP Ann Manuf Technol\u003cem\u003e 51(1)\u003c/em\u003e (1999) 443- 446, http://dx.doi.org/10.1016/S0007-8506(07)61556-X.\u003c/li\u003e\n\u003cli\u003eM. Mahmud, D. Joannic, M. Roy, A. Isheil, J. F. Fontaine, 3D part inspection path planning of a laser scanner with control on the uncertainty. Comput. Aided Des, \u003cem\u003e43\u003c/em\u003e\u003cem\u003e(4)\u003c/em\u003e (2011) 345-355, http://dx.doi.org/10.1016/j.cad.2010.12.014.\u003c/li\u003e\n\u003cli\u003eH. Y. Feng, Y. Liu, F. Xi, Analysis of digitizing errors of a laser scanning system, Precis. Eng. \u003cem\u003e25\u003c/em\u003e\u003cem\u003e(3)\u003c/em\u003e (2001) 185-191, http://dx.doi.org/10.1016/S0141-6359(00)00071-4.\u003c/li\u003e\n\u003cli\u003eN. Van Gestel, S. Cuypers, P. Bleys, J. P. Kruth, A performance evaluation test for laser line scanners on CMMs, Opt Lasers Eng .\u003cem\u003e 47\u003c/em\u003e\u003cem\u003e(3-4)\u003c/em\u003e (2009) 336-342, https://doi.org/10.1016/J.OPTLASENG.2008.06.001.\u003c/li\u003e\n\u003cli\u003eE. Sanders, \u0026amp; S. Fatone, Residual limb volume change: systematic review of measurement and management, \u003cem\u003eJ\u003c/em\u003e. Rehabil. Res. Dev.\u003cem\u003e 48(8)\u003c/em\u003e (2011) 949, https://doi.org/10.1682/jrrd.2010.09.0189\u003c/li\u003e\n\u003cli\u003eY. Yu , W. Zhang, Y. Deng, Frechet Inception Distance (FID) for Evaluating GANs, 2021.\u003c/li\u003e\n\u003cli\u003eOala, L., Murchison, A. G., Balachandran, P., Choudhary, S., Fehr, J., Leite, A. W., \u0026amp; Wiegand, T. (2021). Machine learning for health: algorithm auditing \u0026amp; quality control. \u003cem\u003eJournal of medical systems\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(12), 1-8.\u003c/li\u003e\n\u003cli\u003eM. Mourabet, A. El Rhilassi, H. El Boujaady, M. Bennani-Ziatni, A. Taitai, Use of response surface methodology for optimization of fluoride adsorption in an aqueous solution by Brushite, Arab. J. Chem., (2017) 10 S3292-S3302.\u003c/li\u003e\n\u003cli\u003eR. U. Owolabi, M. A. Usman, \u0026amp; A. J. Kehinde, (2018). Modelling and optimization of process variables for the solution polymerization of styrene using response surface methodology, J. King Saud Univ. Eng. Sci \u003cem\u003e30\u003c/em\u003e(1) 22-30, https://doi.org/10.1016/j.jksues.2015.12.005.\u003c/li\u003e\n\u003cli\u003eV. Kumar, A. Kumar, D. Chhabra, P. Shukla, Improved biobleaching of mixed hardwood pulp and process optimization using novel GA-ANN and GA-ANFIS hybrid statistical tools,\u003cstrong\u003e\u003cbr\u003e\u003c/strong\u003eBioresour. Technol. \u003cem\u003e271\u003c/em\u003e (2019) 274-282, https://doi.org/10.1016/j.biortech.2018.09.115.\u003c/li\u003e\n\u003cli\u003eM. Yadav, D. Yadav, R. K. Garg, R. K. Gupta, S. Kumar, \u0026amp; D. Chhabra, Modeling and Optimization of Piezoelectric Energy Harvesting System Under Dynamic Loading, Advances in Fluid and Thermal Engineering (2021) 339-353, https://www.springerprofessional.de/en/modeling-and-optimization-of-piezoelectric-energy-harvesting-sys/19092572.\u003c/li\u003e\n\u003cli\u003eA. Sharma, D. Chhabra, R. Sahdev, A. Kaushik, U. Punia, Investigation of wear rate of FDM printed TPU, ASA and multi-material parts using heuristic GANN tool. Materials Today: Proceedings.(2022), https://doi.org/10.1016/j.matpr.2022.04.015.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Digital fabrication, 3D scanning, 3D laser scanner, Industry 4.0, Design of experiments, GA-ANN","lastPublishedDoi":"10.21203/rs.3.rs-2051093/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2051093/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThree-dimensional surface scanning of several anatomical areas or human body has gained popularity in current decades due to pre-surgical planning and improved workflow of patient diagnosis and treatment Living surfaces, such as the human face, have various degrees of surface complexity to account for, as well as a range of process parameters to consider. In the proposed work, the face of a person was scanned in various combinations of input parameters using a handheld laser scanner, SENSE 3D (3D system, Rock Hill, SC/USA). Scanner to surface distance, angular orientation, and illumination intensity are considered significant input parameters while using laser scanners for 3D facial data. A number of twenty experimental runs and input parameter combination were suggested by face centered central composite design. The human face has been scanned on these twenty runs to retrieve 3D CAD model and FID score of each model has been completed to investigate the quality/accuracy of the captured data. A model has been trained among input and output using a neural network and further, it is optimized using a genetic algorithm to maximize accuracy The minimum, FID score achieved 270.24, obtained with a scanning distance of 22 inches, the angular orientation of 67.5 degrees, and ambient lightning condition of 16 watt/meter square in twenty experimental runs. The accuracy is maximized by minimizing the FID score utilizing a heuristic GA-ANN technique having 28 inches as scanning distance, 48.041 degrees as angular orientation, and 18 watt/meter square as the ambient lighting condition.\u003c/p\u003e","manuscriptTitle":"Optimization of process parameters for scanning human face using hand-held scanner","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-09-14 21:20:54","doi":"10.21203/rs.3.rs-2051093/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3f9f3bf0-e454-447c-83f3-4f2244ec06ae","owner":[],"postedDate":"September 14th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2022-10-20T20:29:19+00:00","versionOfRecord":[],"versionCreatedAt":"2022-09-14 21:20:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-2051093","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2051093","identity":"rs-2051093","version":["v1"]},"buildId":"7rjqhiLT3MXkJMwkYKINL","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}