Computer Vision Based Detection and Quantification of Extraneous Water in Raw Milk

Computer Vision Based Detection and Quantification of Extraneous Water in Raw Milk | 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 Computer Vision Based Detection and Quantification of Extraneous Water in Raw Milk Bezuayehu Gutema Asefa, Legesse Hagos, Tamirat Kore, Shimelis Admassu Emire This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-625039/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 A rapid method based on digital image analysis and machine learning technique is proposed for the detection of milk adulteration with water. Several machine learning algorithms were compared, and SVM performed best with 89.48 % of total accuracy and 95.10 % precision. An increase in the classification performance was observed in extreme classes. Better quantitative determination of the extraneous water was achieved using SVMR with R 2 (CV) and R 2 (P) of 0.65 and 0.71 respectively. The proposed technique can be used to screen raw milk based on the level of added extraneous water without the necessity of any additional reagent. Food Science & Technology Milk adulteration Multivariate classification Support vector machine Validation Figures Figure 1 1. Introduction With high nutritive value, providing macronutrients (proteins, fat and minerals) and micronutrients (vitamins and trace elements), cow milk is among the recognized contributor to a balanced diet of many populations. Due to the high nutritional composition, a high rate of milk consumption with an increasing demand exists worldwide (Handford et al., 2016 ). Despite the role of milk in food and nutrition security, the increase in demand has amplified fraudulent activities, subsequently making milk the second most vulnerable product to adulteration (Moore et al., 2012 ). Milk adulteration could be dilution with water with the intention to increase economic gain or addition of substances ( e.g. , Sucrose, sodium chloride, vegetable oil and surfactants) that improve the physicochemical and visual characteristics of milk (Poonia et al., 2017 ). Besides, the addition of substances that extend the shelf life of milk, such as formaldehyde, hydrogen peroxide and hypochlorite is becoming a serious issue of adulteration in the dairy industry (Das et al., 2016 ; Handford et al., 2016 ; Nascimento et al., 2017 ). Assessments on the prevalence of milk adulteration in several countries found water as the most frequently added adulterants (Faraz et al., 2013 ; Kandpal et al., 2012 ; Shabir Barham, 2014 ; Soomro et al., 2014 ). Water is added to grow economic gain by increasing the volume of milk through dilution. However, the addition of water to milk dilutes the constituents in milk and could cause potential public health risk of acute malnutrition (stunting, wasting and underweight) which leads to nutrition-related child mortality (Handford et al., 2016 ; Park et al., 2013 ; Shabir Barham, 2014 ). According to experts, next to educating farmers about the consequences of milk fraud, the need for improved detection is key to address the prevailing risk of fraud in milk (Handford et al., 2016 ). Several studies have shown the possibility of determining the presence of water as an adulterant in milk samples using different techniques. Newly developing techniques that are robust, green, simple and cost-effective are gaining increasing importance in food quality monitoring. Digital image-based procedures that use the power of machine learning algorithms are increasingly used to assess adulteration in agro-food products including milk. In recent years, several studies were conducted to develop digital image-based techniques for the determination of adulterants in milk. However, the newly developed techniques lack representative sampling during imaging of milk samples. In addition, indicator chemicals were used to bring the desired classification result before the imaging process (Dos Santos & Pereira-Filho, 2013 ; Kobek, 2017 ). This brings limitations in the utilization of those techniques since users of such methods are required to have technical knowledge of the procedure. Considering the limitation in the existing methods, this paper proposed a clean method based on digital image processing coupled with a machine-learning algorithm to test milk adulteration with water. The proposed technique is fast, robust and doesn’t require sample preparation and use of any chemicals. 2. Materials And Methods 2.1. Milk Samples Raw milk samples were obtained from two different dairy farms found in Sebeta and Debre Zeit Agricultural Research Centers of the Ethiopian Institute of Agricultural Research (EIAR). Known research dairy farms were selected to ensure the purity of the milk before spiking the adulterant. A batch of milk was used to acquire images of pure milk and modified milk with water as an adulterant in a range from 10 to 40%. Since image acquisition was performed in sampling locations, all milk samples used in the study were neither refrigerated nor subjected to transportation longer than one km. The volume of milk sample for each image acquisition was kept constant at 25 ml, which was quantitatively transferred to a petri dish to acquire images from the top surface. Adulterated milk samples were simulated by spiking water in the whole sample used in one day to avoid differences in image intensities due to spiking individual samples. 2.2. Image Acquisition A conventional image acquisition chamber having a dimension L x W x H of (40 x 40 x 60) , made from aluminum sheet was used. Uniform lighting was maintained using twelve fluorescent lumps mounted to four sides of the imaging chamber at a height 40 cm above the bottom surface. A digital camera (EOS, 6D Mark II, Canon, Japan) installed with an image stabilizer of 24–105 mm was set at the top of the image acquisition chamber heading down to the petri dish containing milk sample at a height of around 55 cm. The process of image acquisition was fully monitored using EOS utility software. Fifty samples were prepared for each sample group from the two sampling sites. Image of each sample was captured in duplicate, making a total of two hundred images for each group of samples. 2.3. Feature extraction Images acquired from all samples were processed using a batch processor in ImageJ. All the captured images were treated with a global processing stage, that takes the region of interest from the bulk image. The central area of each image was cropped with a pixel size of 250 x 250. Further processing such as converting to different color spaces (Lab* and HSI) and filtering were performed as summarized in (Table 1 ). The mean and modal grey values, minimum and maximum grey values, standard deviation, median and center of mass were calculated for each processed image. After calculating processed image parameters, some values indicated in the ‘-’ sign in (Table 1 ) are found irrelevant and were not included as a predictor variable due to the fact that similar output values were obtained for all sample groups. Totally, 125 variables were included as a predictor in the development of multivariate models. Table 1 Description of the image processing and parameters included as a variable for the development of classification model Image process description Measurement parameters Mean grey value Standard deviation Modal grey value Median grey value Minimum grey value Maximum grey value Center of mass (X maximum) Center of mass (Y maximum) Skewness Kurtosis Resizing (250 x 250 pixels) + + + + + + + + + + Filtering (Gaussian) + + + + + + + + + + Filtering (Median) + + + + + + + + + + Filtering (Kwahara) + + + + + + + + + + Filtering (FFT) + + + + - - + + + + Filtering (Convolve) + + - - - + + + + + Splitting RGB (R) + + + + + + + + + + Splitting RGB (G) + + + + + + + + + + Splitting RGB (B) + + + + + + + + + + Convert to HSI (H) - - - - - - - - - - Convert to HSI (S) - - - - - - - - - - Convert to HSI (I) + + + + + + + + + + Convert to Lab* (L) + + + + + + + + + + Convert to Lab* (a*) + + + + + + + + + + Convert to Lab* (b*) + + + + + + + + + + Image processing description: ‘+’ signs indicate parameters used as a variable, whereas ‘-’ signs refer to parameters excluded from the variable list 2.4. Multivariate procedure Numerical values generated from the processed images were used to develop classification and regression models based on the level of added water into the pure milk. Multivariate procedures were Performed using MATLAB software (R2020b, PLS Toolbox, Eigenvector). Characteristics of the different multivariate procedures used in the current study are briefly described in Table (2). Table 2 : Summary of machine learning algorithms used for the classification task Table 2 Summary of machine learning algorithms used for the classification task Algorithm Description K-nearest neighbor (K-NN) K-NN-based classification works by identifying the distances between an unknown object and each of the objects of the training set mostly based on the Euclidean distance. A decision is made based on the majority rule after the selection of the k-nearest objects to the unknown sample (Berrueta et al., 2007 ). Soft independent modeling of class analogy (SIMCA) SIMCA calculates the geometric distance from the principal component model and determines the class distance. In addition, the modeling and discriminatory powers are determined (Brereton, 2003 ). Support vector machine (SVM) SVM-based classification works by obtaining the ‘optimal’ boundary of two classes in a vector space independently on the probabilistic distributions of training vectors in the data set (Berrueta et al., 2007 ). Partial least square discriminant analysis (PLS-DA) PLS-based classification works by finding the components in the input matrix (X) that describe the relevant variations at most in the input variables and have a maximal correlation with the target value in Y (Massart et al., 1998 ). 2.5. Model performance evaluation The performance of each model was assessed using a total accuracy method which was computed using the True Positive (TP) and True Negative (TN) values obtained from the confusion matrix (Eq. 1) (Tang et al., 2014 ). Besides, the precision (Eq. 2) recall (Eq. 3) was calculated based on False Negative (FN) and False Positive (FP) values to support the classification model effectiveness (Lopes et al., 2019 ). Accuracy = TP/(TP + FN) (1) Precision = TP/(TP + FP) (2) Recall = TP/(TP + FN) (3) 3. Results 3.1. Exploratory Analysis A total of 25 predictor variables from 900 image data (i.e., 180 x 5 groups) were inspected visually from the excel file to identify potential outliers. Based on the observation, 29 image data were removed and the remaining 871 image data were used to develop the classification models. Before the analysis, Kenard stone technique was employed to randomly separate 80% of the data into the training set and the remaining 20% into a test set. The effect of variation in feature size was corrected by autoscaling the predictor variables. Principal Component Analysis (PCA) was applied to reduce data dimensionality and new variables that are linear combinations of the original image feature values were generated. The selection of an optimal number of PCs was done based on the lowest prediction error in cross-validation (Venetian blinds). 3.2. Multivariate Classification The result table indicating the performance of each classification algorithm is given in Table (3). Of the four classification algorithms, SIMCA provided the worst performance with less than 60% total accuracy in a training dataset. Next to SIMCA, poor classification performance was obtained with the PLSDA algorithm. In contrast to the two classifiers, KNN and SVM achieved fair classification with total accuracy of 79.45 and 89.48 respectively. SVM generally achieved superior results compared to all the classifiers with 89.48% accuracy, 95.10% precision, and 83.24% recall values. Table 3 : Performance measures of different classification algorithms over the training, cross-validation and prediction dataset. Table 3 Performance measures of different classification algorithms over the training, cross-validation and prediction dataset. Algorithm Performance measures Training set Cross-validation set Testing set KNN Accuracy 79.43 81.78 79.45 Precision 89.22 90.35 88.46 Recall 67.16 71.12 67.59 SVM Accuracy 100 86.47 89.48 Precision 100 93.38 95.10 Recall 100 78.58 83.24 PLS-DA Accuracy 66.94 66.12 66.97 Precision 75.91 74.39 75.98 Recall 48.07 46.72 47.90 SIMCA Accuracy 58.49 Precision 41.50 Recall 87.12 Table 4 : Performance measures for class prediction of KNN, PLS-DA and SVM algorithms Table 4 Performance measures for class prediction of KNN, PLS-DA and SVM algorithms Algorithm M: W Training Cross-validation Testing KNN 0% 81.91 81.95 78.04 10% 78.13 79.93 79.38 20% 76.75 80.85 78.59 30% 76.43 77.88 72.04 40% 83.95 88.28 89.19 PLS-DA 0% 63.98 63.75 68.74 10% 67.13 66.56 67.57 20% 58.31 56.58 62.73 30% 56.94 55.82 55.48 40% 88.37 87.88 80.35 SVM 0% 100 87.07 91.95 10% 100 85.27 88.26 20% 100 84.72 88.99 30% 100 82.82 86.18 40% 100 92.47 92.04 3.3. Estimation of adulteration level The dataset was also used to develop a prediction model for the level of adulteration. The prediction performance of Partial Least Squares Regression (PLSR), Principal Component Regression (PCR), and SVMR algorithms was evaluated. The summary of quantitative prediction performance measures is presented in Table (5). Table 5 : Performance measures of regression models developed for quantitative adulterant prediction Table 5 Performance measures of regression models developed for quantitative adulterant prediction No. Method preprocess. LV/PC RMSEC RMSECV R 2 (Cal) R 2 (CV) R 2 (P) 2 PCR >> 3 11.23 11.28 0.31 0.30 0.16 3 PLSR >> 6 9.84 10.05 0.47 0.44 0.44 4 SVMR >> 4.93 8.02 0.87 0.65 0.71 4. Discussion The exploratory analysis showed that the first three PCs explained more than 75% of the data variance as indicated in a 3-dimensional PCA score-plot obtained from three PCs (Fig. 1 ). The change in color intensity can be observed from the score-plot. Increasing the amount of added water could be related to the diminishing color density of the images which is illustrated in reduced scores in PC 1. Since milk color is influenced by the composition, the addition of water to pure milk can affect the intensity. Detecting such minor differences in the intensity of milk color using the human eye could be difficult unless digital technologies are used with the support of numerical software. Further analysis on the model’s prediction performance for each class of samples exhibited efficient classification performance of SVM algorithms in extreme classes Table (4). This means milk samples with no adulteration and milk samples that have 40 % added water were identified with better classification performance compared to other samples. Correct identification of pure milk sample was achieved using the same algorithm with an accuracy of 91.95% in prediction set samples. Also, SVM achieved the highest classification accuracy (92.04) in milk samples adulterated with 40% water. This result outperformed the previously developed procedure (Kobek, 2017 ), who found total classification accuracy of 81.66 using an Artificial Neural Network (ANN) based classification model. In another research, SIMCA and KNN classification algorithms were applied to distinguish milk adulterated with water from pure milk, and total accuracy of 82 and 92% respectively for SIMCA and KNN were found (Dos Santos & Pereira-Filho, 2013 ). However, indicator chemicals were used to bring the desired color change in the two findings. Given these facts, our finding verified the possibility of using digital images to determine milk adulteration with water without the necessity of adding indicator chemicals. Except for the SVMR algorithm, inadequate prediction performance was found in predicting the level of extraneous water with prediction R 2 of 0.16, 0.44 and 0.52 in PCR, PLSR and MLR respectively. Interestingly, SVMR achieved better performance in predicting the amount of adulterated water in the milk samples with R 2 (CV) and R 2 (P) of 0.65 and 0.71 respectively. 5. Conclusion The change in color of milk due to dilution by water has proved to be useful to detect adulteration through the use of processed images coupled with machine learning algorithms. SVM classification model discriminated milk samples based on the level of added water with accuracy and precision of 89.48 % and 95.10%, respectively. The performance of the proposed technique is satisfactory to apply for screening of raw milk samples at dairy processing industry. The proposed technique can be used for the rapid determination of extraneous water in raw milk without the necessity of any additional reagent. Abbreviations SVM: Support Vector Machine; SVMR: Support Vector Machine Regression; CV: Cross validation; P: Prediction; HIS: Hue, Intensity, Saturation; TP: True positive; TN: True negative; FN: False negative FP: False positive PCA: Principal Component analysis SIMCA: Soft Independent Modeling of Class Analogies KNN: K- nearest neighbors PLSDA: Partial Least Squares Discriminant Analysis PCR: Principal Component Regression PLS: Partial Least Squares PLSR: Partial Least Square Regression EIAR: Ethiopian Institute of agricultural Research Declarations Ethical approval and consent to participate Not applicable Consent for publication Not applicable Availability of data and materials Data can be shared upon request Competing interests The authors declare that they have no known competing interests. Funding Not applicable Authors’ contributions Bezuayehu G. Asefa : Conceptualization, Methodology, Data analysis, Writing - original draft. Legesse Hagos : Data acquisition, Writing - review & editing. Tamirat Kore : Data acquisition, Writing - review & editing. Shimelish A. Emiru : Methodology, Writing - review & editing, supervision. Acknowledgments The authors wish to thank the Ethiopian Institute of Agricultural Research for providing the necessary facilities for carrying out the study. Author’s information 1 Food Science and Nutrition Research, National Fishery and Aquatic Life Research Center, Ethiopian Institute of Agricultural Research, P. O. Box 64, Sebeta, Ethiopia; 2 Food Science and Nutrition Research, Debre Zeit Agricultural Research Center, Ethiopian Institute of Agricultural Research, Debre Zeit, Ethiopia; 3 Department of Food Engineering, School of Chemical and Bioengineering, Addis Ababa University, P.O. Box 33381, Addis Ababa, Ethiopia. References Berrueta LA, Alonso-Salces RM, Héberger K (2007) Supervised pattern recognition in food analysis. J Chromatogr A 1158(1–2):196–214 Brereton RG (2003) Chemometrics: Data analysis for the laboratory and chemical plant. John Wiley & Sons Das S, Goswami B, Biswas K (2016) Milk Adulteration and Detection: A Review. Sensor Letters 14(1):4–18. https://doi.org/10.1166/sl.2016.3580 Dos Santos PM, Pereira-Filho ER (2013) Digital image analysis–an alternative tool for monitoring milk authenticity. Anal Methods 5(15):3669–3674 Faraz A, Lateef M, Mustafa MI, Akhtar P, Yaqoob M, Rehman S (2013) Detection of adulteration, chemical composition and hygienic status of milk supplied to various canteens of educational institutes and public places in Faisalabad. JAPS Journal of Animal Plant Sciences 23(1 Supplement):119–124 Handford CE, Campbell K, Elliott CT (2016) Impacts of Milk Fraud on Food Safety and Nutrition with Special Emphasis on Developing Countries. Comprehensive Reviews in Food Science Food Safety 15(1):130–142. https://doi.org/10.1111/1541-4337.12181 Kandpal SD, Srivastava AK, Negi KS (2012) ESTIMATION OF QUALITY OF RAW MILK (OPEN & BRANDED) BY MILK ADULTERATION TESTING KIT. Indian Journal of Community Health 24(3):188–192 Kobek JA (2017) Vision based model for identification of adulterants in milk. Strathmore University Lopes JF, Ludwig L, Barbin DF, Grossmann MVE, Barbon S (2019) Computer vision classification of barley flour based on spatial pyramid partition ensemble. Sensors 19(13):2953 Massart DL, Vandeginste BG, Buydens LM, Lewi PJ, Smeyers-Verbeke J, Jong SD (1998) Handbook of chemometrics and qualimetrics. Elsevier Science Inc Moore JC, Spink J, Lipp M (2012) Development and Application of a Database of Food Ingredient Fraud and Economically Motivated Adulteration from 1980 to 2010. J Food Sci 77(4):R118–R126. https://doi.org/10.1111/j.1750-3841.2012.02657.x Nascimento CF, Santos PM, Pereira-Filho ER, Rocha FR (2017) Recent advances on determination of milk adulterants. Food Chem 221:1232–1244 Park YW, Haenlein GF, Ag DS (2013) Milk and dairy products in human nutrition. Wilet-Blackwell. A John Wiley & Sons, Ltd., Publication , 700 Poonia A, Jha A, Sharma R, Singh HB, Rai AK, Sharma N (2017) Detection of adulteration in milk: A review. Int J Dairy Technol 70(1):23–42. https://doi.org/10.1111/1471-0307.12274 Shabir Barham G (2014) Detection and Extent of Extraneous Water and Adulteration in Milk Consumed at Hyderabad, Pakistan. Journal of Food Nutrition Sciences 2(2):47. https://doi.org/10.11648/j.jfns.20140202.15 Soomro AA, Khaskheli M, Memon MA, Barham GS, Haq IU, Fazlani SN, Khan IA, Lochi GM, Soomro RN (2014) Study on adulteration and composition of milk sold at Badin. Intl J Res Appl Nat Social Sci 2(9):57–70 Tang J, Alelyani S, Liu H (2014) Data classification: Algorithms and applications. Data Mining and Knowledge Discovery Series, CRC Press (2014) , 37–64 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-625039","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":35873325,"identity":"7bccc1f4-aebd-4f96-8e77-3067bd162438","order_by":0,"name":"Bezuayehu Gutema Asefa","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYBACAyidwMDA//EBkMHDR4IWBmMQh4eNFC1mEiAWQS3m7O3PJH622eQZnF+QVvk1x06GjYH54aMbeLRY9hxIk+xtSys2uPHg2G3ZbclAh7EZG+fgc9iNhGMSvG2HEzfcONh2W3IbM1ALD5s0Xi33H7ZJ/m37D9RymK1Ycls9EVpuMLNJ87YdSNxwvo2N8eO2w4S1WPakMVvLnEtOnHmDh1macdtxHjZmAn4xZz/+8OabMrvEvvNnGD/+3FZtz8/e/PAxPi1gwAiKC4kEBmYeEI+ZkHIw+APE/AcYGH8QpXoUjIJRMApGGgAAK4xLqWCoF9UAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-1525-2054","institution":"Ethiopian Institute of Agricultural Research","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Bezuayehu","middleName":"Gutema","lastName":"Asefa","suffix":""},{"id":35873326,"identity":"9db6b2d2-31e9-439f-b5cc-b70148507437","order_by":1,"name":"Legesse Hagos","email":"","orcid":"","institution":"Ethiopian Institute of Agricultural Research","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Legesse","middleName":"","lastName":"Hagos","suffix":""},{"id":35873327,"identity":"b0222478-0392-4632-ba0d-cef13768f31c","order_by":2,"name":"Tamirat Kore","email":"","orcid":"","institution":"Ethiopian Institute of Agricultural Research","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Tamirat","middleName":"","lastName":"Kore","suffix":""},{"id":35873328,"identity":"4810af2c-c016-4f86-a887-2e8a54e60416","order_by":3,"name":"Shimelis Admassu Emire","email":"","orcid":"","institution":"Addis Ababa University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Shimelis","middleName":"Admassu","lastName":"Emire","suffix":""}],"badges":[],"createdAt":"2021-06-15 17:59:51","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-625039/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-625039/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":10909671,"identity":"b3632e79-5fab-4918-a14a-64777dd75cce","added_by":"auto","created_at":"2021-06-29 14:33:17","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":26599,"visible":true,"origin":"","legend":"3-D score plot of adulterated milk samples","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-625039/v1/e1417d722bc53a7d63d7c8b6.png"},{"id":13701326,"identity":"128e0a95-c6b9-4c02-bd36-1804ae794b9c","added_by":"auto","created_at":"2021-09-17 13:29:03","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":424407,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-625039/v1/1449b1f5-cfdb-43cd-8bdd-45ebede430c7.pdf"}],"financialInterests":"","formattedTitle":"Computer Vision Based Detection and Quantification of Extraneous Water in Raw Milk","fulltext":[{"header":"1. Introduction","content":" \u003cp\u003eWith high nutritive value, providing macronutrients (proteins, fat and minerals) and micronutrients (vitamins and trace elements), cow milk is among the recognized contributor to a balanced diet of many populations. Due to the high nutritional composition, a high rate of milk consumption with an increasing demand exists worldwide (Handford et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Despite the role of milk in food and nutrition security, the increase in demand has amplified fraudulent activities, subsequently making milk the second most vulnerable product to adulteration (Moore et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2012\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMilk adulteration could be dilution with water with the intention to increase economic gain or addition of substances (\u003cem\u003ee.g.\u003c/em\u003e, Sucrose, sodium chloride, vegetable oil and surfactants) that improve the physicochemical and visual characteristics of milk (Poonia et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Besides, the addition of substances that extend the shelf life of milk, such as formaldehyde, hydrogen peroxide and hypochlorite is becoming a serious issue of adulteration in the dairy industry (Das et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Handford et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Nascimento et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAssessments on the prevalence of milk adulteration in several countries found water as the most frequently added adulterants (Faraz et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Kandpal et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Shabir Barham, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Soomro et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Water is added to grow economic gain by increasing the volume of milk through dilution. However, the addition of water to milk dilutes the constituents in milk and could cause potential public health risk of acute malnutrition (stunting, wasting and underweight) which leads to nutrition-related child mortality (Handford et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Park et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Shabir Barham, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). According to experts, next to educating farmers about the consequences of milk fraud, the need for improved detection is key to address the prevailing risk of fraud in milk (Handford et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSeveral studies have shown the possibility of determining the presence of water as an adulterant in milk samples using different techniques. Newly developing techniques that are robust, green, simple and cost-effective are gaining increasing importance in food quality monitoring. Digital image-based procedures that use the power of machine learning algorithms are increasingly used to assess adulteration in agro-food products including milk. In recent years, several studies were conducted to develop digital image-based techniques for the determination of adulterants in milk. However, the newly developed techniques lack representative sampling during imaging of milk samples. In addition, indicator chemicals were used to bring the desired classification result before the imaging process (Dos Santos \u0026amp; Pereira-Filho, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Kobek, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). This brings limitations in the utilization of those techniques since users of such methods are required to have technical knowledge of the procedure.\u003c/p\u003e \u003cp\u003eConsidering the limitation in the existing methods, this paper proposed a clean method based on digital image processing coupled with a machine-learning algorithm to test milk adulteration with water. The proposed technique is fast, robust and doesn\u0026rsquo;t require sample preparation and use of any chemicals.\u003c/p\u003e "},{"header":"2. Materials And Methods","content":" \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Milk Samples\u003c/h2\u003e \u003cp\u003eRaw milk samples were obtained from two different dairy farms found in Sebeta and Debre Zeit Agricultural Research Centers of the Ethiopian Institute of Agricultural Research (EIAR). Known research dairy farms were selected to ensure the purity of the milk before spiking the adulterant. A batch of milk was used to acquire images of pure milk and modified milk with water as an adulterant in a range from 10 to 40%. Since image acquisition was performed in sampling locations, all milk samples used in the study were neither refrigerated nor subjected to transportation longer than one km. The volume of milk sample for each image acquisition was kept constant at 25 ml, which was quantitatively transferred to a petri dish to acquire images from the top surface. Adulterated milk samples were simulated by spiking water in the whole sample used in one day to avoid differences in image intensities due to spiking individual samples.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Image Acquisition\u003c/h2\u003e \u003cp\u003eA conventional image acquisition chamber having a dimension \u003cem\u003eL x W x H\u003c/em\u003e of \u003cem\u003e(40 x 40 x 60)\u003c/em\u003e, made from aluminum sheet was used. Uniform lighting was maintained using twelve fluorescent lumps mounted to four sides of the imaging chamber at a height 40 cm above the bottom surface. A digital camera (EOS, 6D Mark II, Canon, Japan) installed with an image stabilizer of 24\u0026ndash;105 mm was set at the top of the image acquisition chamber heading down to the petri dish containing milk sample at a height of around 55 cm. The process of image acquisition was fully monitored using EOS utility software. Fifty samples were prepared for each sample group from the two sampling sites. Image of each sample was captured in duplicate, making a total of two hundred images for each group of samples.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Feature extraction\u003c/h2\u003e \u003cp\u003eImages acquired from all samples were processed using a batch processor in ImageJ. All the captured images were treated with a global processing stage, that takes the region of interest from the bulk image. The central area of each image was cropped with a pixel size of 250 x 250. Further processing such as converting to different color spaces (Lab* and HSI) and filtering were performed as summarized in (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The mean and modal grey values, minimum and maximum grey values, standard deviation, median and center of mass were calculated for each processed image. After calculating processed image parameters, some values indicated in the \u0026lsquo;-\u0026rsquo; sign in (Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) are found irrelevant and were not included as a predictor variable due to the fact that similar output values were obtained for all sample groups. Totally, 125 variables were included as a predictor in the development of multivariate models.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescription of the image processing and parameters included as a variable for the development of classification model\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"11\"\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eImage process description\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"10\" nameend=\"c11\" namest=\"c2\"\u003e \u003cp\u003eMeasurement parameters\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMean grey value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStandard deviation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eModal grey value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMedian grey value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMinimum grey value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMaximum grey value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eCenter of mass (X maximum)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eCenter of mass (Y maximum)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eSkewness\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003eKurtosis\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResizing (250 x 250 pixels)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFiltering (Gaussian)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFiltering (Median)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFiltering (Kwahara)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFiltering (FFT)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFiltering (Convolve)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSplitting RGB (R)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSplitting RGB (G)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSplitting RGB (B)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConvert to HSI (H)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConvert to HSI (S)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConvert to HSI (I)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConvert to Lab* (L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConvert to Lab* (a*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConvert to Lab* (b*)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c11\"\u003e \u003cp\u003e+\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"11\"\u003e\u003cem\u003eImage processing description: \u0026lsquo;+\u0026rsquo; signs indicate parameters used as a variable, whereas \u0026lsquo;-\u0026rsquo; signs refer to parameters excluded from the variable list\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4. Multivariate procedure\u003c/h2\u003e \u003cp\u003eNumerical values generated from the processed images were used to develop classification and regression models based on the level of added water into the pure milk. Multivariate procedures were Performed using MATLAB software (R2020b, PLS Toolbox, Eigenvector). Characteristics of the different multivariate procedures used in the current study are briefly described in Table\u0026nbsp;(2).\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e: Summary of machine learning algorithms used for the classification task\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of machine learning algorithms used for the classification task\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgorithm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDescription\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eK-nearest neighbor (K-NN)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eK-NN-based classification works by identifying the distances between an unknown object and each of the objects of the training set mostly based on the Euclidean distance. A decision is made based on the majority rule after the selection of the k-nearest objects to the unknown sample (Berrueta et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSoft independent modeling of class analogy (SIMCA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSIMCA calculates the geometric distance from the principal component model and determines the class distance. In addition, the modeling and discriminatory powers are determined (Brereton, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2003\u003c/span\u003e).\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSupport vector machine (SVM)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSVM-based classification works by obtaining the \u0026lsquo;optimal\u0026rsquo; boundary of two classes in a vector space independently on the probabilistic distributions of training vectors in the data set (Berrueta et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePartial least square discriminant analysis (PLS-DA)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePLS-based classification works by finding the components in the input matrix (X) that describe the relevant variations at most in the input variables and have a maximal correlation with the target value in Y (Massart et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1998\u003c/span\u003e).\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5. Model performance evaluation\u003c/h2\u003e \u003cp\u003eThe performance of each model was assessed using a total accuracy method which was computed using the True Positive (TP) and True Negative (TN) values obtained from the confusion matrix (Eq.\u0026nbsp;1) (Tang et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Besides, the precision (Eq.\u0026nbsp;2) recall (Eq.\u0026nbsp;3) was calculated based on False Negative (FN) and False Positive (FP) values to support the classification model effectiveness (Lopes et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cem\u003eAccuracy\u0026thinsp;=\u0026thinsp;TP/(TP\u0026thinsp;+\u0026thinsp;FN)\u003c/em\u003e (1)\u003c/p\u003e \u003cp\u003e \u003cem\u003ePrecision\u0026thinsp;=\u0026thinsp;TP/(TP\u0026thinsp;+\u0026thinsp;FP) (2)\u003c/em\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eRecall\u0026thinsp;=\u0026thinsp;TP/(TP\u0026thinsp;+\u0026thinsp;FN) (3)\u003c/em\u003e \u003c/p\u003e \u003c/div\u003e "},{"header":"3. Results","content":" \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Exploratory Analysis\u003c/h2\u003e \u003cp\u003eA total of 25 predictor variables from 900 image data \u003cem\u003e(i.e., 180 x 5 groups)\u003c/em\u003e were inspected visually from the excel file to identify potential outliers. Based on the observation, 29 image data were removed and the remaining 871 image data were used to develop the classification models. Before the analysis, Kenard stone technique was employed to randomly separate 80% of the data into the training set and the remaining 20% into a test set. The effect of variation in feature size was corrected by autoscaling the predictor variables.\u003c/p\u003e \u003cp\u003ePrincipal Component Analysis (PCA) was applied to reduce data dimensionality and new variables that are linear combinations of the original image feature values were generated. The selection of an optimal number of PCs was done based on the lowest prediction error in cross-validation (Venetian blinds).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Multivariate Classification\u003c/h2\u003e \u003cp\u003eThe result table indicating the performance of each classification algorithm is given in Table\u0026nbsp;(3). Of the four classification algorithms, SIMCA provided the worst performance with less than 60% total accuracy in a training dataset. Next to SIMCA, poor classification performance was obtained with the PLSDA algorithm. In contrast to the two classifiers, KNN and SVM achieved fair classification with total accuracy of 79.45 and 89.48 respectively. SVM generally achieved superior results compared to all the classifiers with 89.48% accuracy, 95.10% precision, and 83.24% recall values.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e: Performance measures of different classification algorithms over the training, cross-validation and prediction dataset.\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\u003ePerformance measures of different classification algorithms over the training, cross-validation and prediction dataset.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgorithm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePerformance measures\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCross-validation set\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTesting set\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eKNN\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e79.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e81.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e79.45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e89.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e90.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e88.46\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e71.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e67.59\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSVM\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e86.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e89.48\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e93.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e95.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e78.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e83.24\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003ePLS-DA\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e66.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e66.97\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e74.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e75.98\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e48.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e46.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e47.90\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e\u003cb\u003eSIMCA\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e41.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e87.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e: Performance measures for class prediction of KNN, PLS-DA and SVM algorithms\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\u003ePerformance measures for class prediction of KNN, PLS-DA and SVM algorithms\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlgorithm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eM: W\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCross-validation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTesting\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eKNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e81.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e79.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e79.38\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e80.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.59\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e77.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e72.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e88.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e89.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003ePLS-DA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e63.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e63.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e68.74\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e66.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e67.57\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e56.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e62.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e56.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e55.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e55.48\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e88.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e87.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e80.35\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e87.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e91.95\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e85.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e88.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e84.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e88.99\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e82.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e86.18\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e40%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e92.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e92.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Estimation of adulteration level\u003c/h2\u003e \u003cp\u003eThe dataset was also used to develop a prediction model for the level of adulteration. The prediction performance of Partial Least Squares Regression (PLSR), Principal Component Regression (PCR), and SVMR algorithms was evaluated. The summary of quantitative prediction performance measures is presented in Table\u0026nbsp;(5).\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e: Performance measures of regression models developed for quantitative adulterant prediction\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance measures of regression models developed for quantitative adulterant prediction\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMethod\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003epreprocess.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLV/PC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRMSEC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRMSECV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e(Cal)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e(CV)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e(P)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePCR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026gt;\u0026gt;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e11.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e11.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePLSR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026gt;\u0026gt;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e9.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e10.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSVMR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026gt;\u0026gt;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e8.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e "},{"header":"4. Discussion","content":" \u003cp\u003eThe exploratory analysis showed that the first three PCs explained more than 75% of the data variance as indicated in a 3-dimensional PCA score-plot obtained from three PCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The change in color intensity can be observed from the score-plot. Increasing the amount of added water could be related to the diminishing color density of the images which is illustrated in reduced scores in PC 1. Since milk color is influenced by the composition, the addition of water to pure milk can affect the intensity. Detecting such minor differences in the intensity of milk color using the human eye could be difficult unless digital technologies are used with the support of numerical software.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFurther analysis on the model\u0026rsquo;s prediction performance for each class of samples exhibited efficient classification performance of SVM algorithms in extreme classes Table\u0026nbsp;(4). This means milk samples with no adulteration and milk samples that have 40 % added water were identified with better classification performance compared to other samples. Correct identification of pure milk sample was achieved using the same algorithm with an accuracy of 91.95% in prediction set samples. Also, SVM achieved the highest classification accuracy (92.04) in milk samples adulterated with 40% water.\u003c/p\u003e \u003cp\u003eThis result outperformed the previously developed procedure (Kobek, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), who found total classification accuracy of 81.66 using an Artificial Neural Network (ANN) based classification model. In another research, SIMCA and KNN classification algorithms were applied to distinguish milk adulterated with water from pure milk, and total accuracy of 82 and 92% respectively for SIMCA and KNN were found (Dos Santos \u0026amp; Pereira-Filho, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). However, indicator chemicals were used to bring the desired color change in the two findings. Given these facts, our finding verified the possibility of using digital images to determine milk adulteration with water without the necessity of adding indicator chemicals.\u003c/p\u003e \u003cp\u003eExcept for the SVMR algorithm, inadequate prediction performance was found in predicting the level of extraneous water with prediction R\u003csup\u003e2\u003c/sup\u003e of 0.16, 0.44 and 0.52 in PCR, PLSR and MLR respectively. Interestingly, SVMR achieved better performance in predicting the amount of adulterated water in the milk samples with R\u003csup\u003e2\u003c/sup\u003e(CV) and R\u003csup\u003e2\u003c/sup\u003e(P) of 0.65 and 0.71 respectively.\u003c/p\u003e "},{"header":"5. Conclusion","content":" \u003cp\u003eThe change in color of milk due to dilution by water has proved to be useful to detect adulteration through the use of processed images coupled with machine learning algorithms. SVM classification model discriminated milk samples based on the level of added water with accuracy and precision of 89.48 % and 95.10%, respectively. The performance of the proposed technique is satisfactory to apply for screening of raw milk samples at dairy processing industry. The proposed technique can be used for the rapid determination of extraneous water in raw milk without the necessity of any additional reagent.\u003c/p\u003e "},{"header":"Abbreviations","content":"\u003cp\u003eSVM: Support Vector Machine;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSVMR: Support Vector Machine Regression;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCV: Cross validation;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eP: Prediction;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHIS: Hue, Intensity, Saturation;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTP: True positive;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTN: True negative;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFN: False negative\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFP: False positive\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePCA: Principal Component analysis\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSIMCA: Soft Independent Modeling of Class Analogies\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eKNN: K- nearest neighbors\u003c/p\u003e\n\u003cp\u003ePLSDA: Partial Least Squares Discriminant Analysis\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePCR: Principal Component Regression\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePLS: Partial Least Squares\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePLSR: Partial Least Square Regression\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEIAR: Ethiopian Institute of agricultural Research\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthical approval and consent to participate\u003c/h2\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003ch2\u003eConsent for publication\u003c/h2\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003ch2\u003eAvailability of data and materials\u003c/h2\u003e\n\u003cp\u003eData can be shared upon request\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no known competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026rsquo; contributions\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eBezuayehu G. Asefa\u003c/strong\u003e: Conceptualization, Methodology, Data analysis, Writing - original draft. \u0026nbsp;\u003cstrong\u003eLegesse Hagos\u003c/strong\u003e: Data acquisition, Writing - review \u0026amp; editing. \u003cstrong\u003eTamirat Kore\u003c/strong\u003e: Data acquisition, Writing - review \u0026amp; editing. \u003cstrong\u003eShimelish A. Emiru\u003c/strong\u003e: Methodology, Writing - review \u0026amp; editing, supervision.\u003c/p\u003e\n\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eThe authors wish to thank the Ethiopian Institute of Agricultural Research for providing the necessary facilities for carrying out the study.\u003c/p\u003e\n\u003ch2\u003eAuthor\u0026rsquo;s information\u003c/h2\u003e\n\u003cp\u003e\u003csup\u003e1\u003c/sup\u003eFood Science and Nutrition Research, National Fishery and Aquatic Life Research Center, Ethiopian Institute of Agricultural Research, P. O. Box 64, Sebeta, Ethiopia; \u003csup\u003e2\u003c/sup\u003eFood Science and Nutrition Research, Debre Zeit Agricultural Research Center, Ethiopian Institute of Agricultural Research, Debre Zeit, Ethiopia; \u003csup\u003e3\u003c/sup\u003eDepartment of Food Engineering, School of Chemical and Bioengineering, Addis Ababa University, P.O. Box 33381, Addis Ababa, Ethiopia.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e \u003cli\u003e\u003cspan\u003eBerrueta LA, Alonso-Salces RM, H\u0026eacute;berger K (2007) Supervised pattern recognition in food analysis. J Chromatogr A 1158(1\u0026ndash;2):196\u0026ndash;214\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrereton RG (2003) Chemometrics: Data analysis for the laboratory and chemical plant. John Wiley \u0026amp; Sons\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDas S, Goswami B, Biswas K (2016) Milk Adulteration and Detection: A Review. Sensor Letters 14(1):4\u0026ndash;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1166/sl.2016.3580\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDos Santos PM, Pereira-Filho ER (2013) Digital image analysis\u0026ndash;an alternative tool for monitoring milk authenticity. Anal Methods 5(15):3669\u0026ndash;3674\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaraz A, Lateef M, Mustafa MI, Akhtar P, Yaqoob M, Rehman S (2013) Detection of adulteration, chemical composition and hygienic status of milk supplied to various canteens of educational institutes and public places in Faisalabad. JAPS Journal of Animal Plant Sciences 23(1 Supplement):119\u0026ndash;124\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHandford CE, Campbell K, Elliott CT (2016) Impacts of Milk Fraud on Food Safety and Nutrition with Special Emphasis on Developing Countries. Comprehensive Reviews in Food Science Food Safety 15(1):130\u0026ndash;142. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1541-4337.12181\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKandpal SD, Srivastava AK, Negi KS (2012) ESTIMATION OF QUALITY OF RAW MILK (OPEN \u0026amp; BRANDED) BY MILK ADULTERATION TESTING KIT. Indian Journal of Community Health 24(3):188\u0026ndash;192\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKobek JA (2017) Vision based model for identification of adulterants in milk. Strathmore University\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLopes JF, Ludwig L, Barbin DF, Grossmann MVE, Barbon S (2019) Computer vision classification of barley flour based on spatial pyramid partition ensemble. Sensors 19(13):2953\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMassart DL, Vandeginste BG, Buydens LM, Lewi PJ, Smeyers-Verbeke J, Jong SD (1998) Handbook of chemometrics and qualimetrics. Elsevier Science Inc\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoore JC, Spink J, Lipp M (2012) Development and Application of a Database of Food Ingredient Fraud and Economically Motivated Adulteration from 1980 to 2010. J Food Sci 77(4):R118\u0026ndash;R126. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1750-3841.2012.02657.x\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNascimento CF, Santos PM, Pereira-Filho ER, Rocha FR (2017) Recent advances on determination of milk adulterants. Food Chem 221:1232\u0026ndash;1244\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark YW, Haenlein GF, Ag DS (2013) Milk and dairy products in human nutrition. \u003cem\u003eWilet-Blackwell. A John Wiley \u0026amp; Sons, Ltd., Publication\u003c/em\u003e, \u003cem\u003e700\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoonia A, Jha A, Sharma R, Singh HB, Rai AK, Sharma N (2017) Detection of adulteration in milk: A review. Int J Dairy Technol 70(1):23\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1471-0307.12274\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShabir Barham G (2014) Detection and Extent of Extraneous Water and Adulteration in Milk Consumed at Hyderabad, Pakistan. Journal of Food Nutrition Sciences 2(2):47. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.11648/j.jfns.20140202.15\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoomro AA, Khaskheli M, Memon MA, Barham GS, Haq IU, Fazlani SN, Khan IA, Lochi GM, Soomro RN (2014) Study on adulteration and composition of milk sold at Badin. Intl J Res Appl Nat Social Sci 2(9):57\u0026ndash;70\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang J, Alelyani S, Liu H (2014) Data classification: Algorithms and applications. \u003cem\u003eData Mining and Knowledge Discovery Series, CRC Press (2014)\u003c/em\u003e, 37\u0026ndash;64\u003c/span\u003e\u003c/li\u003e\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":"Milk adulteration, Multivariate classification, Support vector machine, Validation","lastPublishedDoi":"10.21203/rs.3.rs-625039/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-625039/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eA rapid method based on digital image analysis and machine learning technique is proposed for the detection of milk adulteration with water. Several machine learning algorithms were compared, and SVM performed best with 89.48 % of total accuracy and 95.10 % precision. An increase in the classification performance was observed in extreme classes. Better quantitative determination of the extraneous water was achieved using SVMR with R\u003csup\u003e2\u003c/sup\u003e(CV) and R\u003csup\u003e2\u003c/sup\u003e(P) of 0.65 and 0.71 respectively. The proposed technique can be used to screen raw milk based on the level of added extraneous water without the necessity of any additional reagent.\u003c/p\u003e","manuscriptTitle":"Computer Vision Based Detection and Quantification of Extraneous Water in Raw Milk","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-06-29 14:30:16","doi":"10.21203/rs.3.rs-625039/v1","editorialEvents":[{"type":"communityComments","content":2}],"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":"9631a764-a084-4dc3-8e40-5f110c06acf3","owner":[],"postedDate":"June 29th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":5339229,"name":"Food Science \u0026 Technology"}],"tags":[],"updatedAt":"2021-07-27T19:02:33+00:00","versionOfRecord":[],"versionCreatedAt":"2021-06-29 14:30:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-625039","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-625039","identity":"rs-625039","version":["v1"]},"buildId":"-HB7Z8yhvgn0wM9Nzuekk","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}