Progress and Challenges in Ear Biometrics for Secure Identification

Progress and Challenges in Ear Biometrics for Secure Identification | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 15 April 2025 V1 Latest version Share on Progress and Challenges in Ear Biometrics for Secure Identification Authors : Niraj K. Nagrale 0000-0002-1344-5522 [email protected] and Vishakha Nagrale Authors Info & Affiliations https://doi.org/10.22541/au.174471555.52636866/v1 1528 views 200 downloads Contents Abstract Introduction Ear Biometric Recognition Methods Recognition Algorithms Applications of Ear Biometrics Challenges and Future Directions References Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Ear biometrics is a growing field in biometric recognition, offering a unique and stable alternative for personal identification. Unlike traditional biometric modalities such as fingerprints and facial recognition, ears provide a non-intrusive, stable, and unique feature set that remains relatively unchanged over time. Recent advancements in machine learning and deep learning have significantly improved the accuracy of ear recognition systems. This paper explores various methodologies for feature extraction, including geometric-based, texture-based, and deep learning approaches. Additionally, we discuss recognition algorithms such as Principal Component Analysis (PCA), Support Vector Machines (SVM), and Convolutional Neural Networks (CNNs). The applications of ear biometrics extend to security, forensic science, and healthcare, offering promising potential in identity verification and crime investigation. However, challenges such as dataset limitations, occlusions, and variations due to aging persist. Future research should focus on improving deep learning models, increasing dataset diversity, and exploring multi-modal biometric systems for enhanced security and accuracy. This paper provides an in-depth analysis of the advancements, challenges, and future directions in ear biometrics, paving the way for further innovation and real-world applications. Progress and Challenges in Ear Biometrics for Secure Identification Niraj K. Nagrale Dept. of ETC Engineering, Bharti Vishvavidyalaya, Durg, India and G H Raisoni College of Engineering, Nagpur, India [email protected] Vishakha Nagrale Dept. of CS & Engineering, G H Raisoni College of Engineering And Management Nagpur, India [email protected] Abstract — Ear biometrics is a growing field in biometric recognition, offering a unique and stable alternative for personal identification. Unlike traditional biometric modalities such as fingerprints and facial recognition, ears provide a non-intrusive, stable, and unique feature set that remains relatively unchanged over time. Recent advancements in machine learning and deep learning have significantly improved the accuracy of ear recognition systems. This paper explores various methodologies for feature extraction, including geometric-based, texture-based, and deep learning approaches. Additionally, we discuss recognition algorithms such as Principal Component Analysis (PCA), Support Vector Machines (SVM), and Convolutional Neural Networks (CNNs). The applications of ear biometrics extend to security, forensic science, and healthcare, offering promising potential in identity verification and crime investigation. However, challenges such as dataset limitations, occlusions, and variations due to aging persist. Future research should focus on improving deep learning models, increasing dataset diversity, and exploring multi-modal biometric systems for enhanced security and accuracy. This paper provides an in-depth analysis of the advancements, challenges, and future directions in ear biometrics, paving the way for further innovation and real-world applications. Keywords— Biometric security, Deep learning, Ear biometrics, Feature extraction, Machine learning, Recognition algorithms, Security applications Introduction Biometric authentication has gained significant traction in recent years due to its reliability and security. Traditional biometric modalities such as fingerprint and facial recognition dominate the industry; however, ear biometrics is gaining attention due to its uniqueness, permanence, and non-intrusive nature. Unlike fingerprints, which may wear out, or facial recognition, which can be affected by expression changes, the structure of the human ear remains largely unchanged over time. This makes ear biometrics a reliable modality for authentication and identification purposes [1]. The field of ear biometrics has evolved with the integration of advanced computational techniques, particularly in the areas of machine learning and deep learning. Various feature extraction techniques have been proposed, including geometric-based, texture-based, and deep learning-based approaches, to enhance recognition accuracy. Despite its potential, ear biometrics faces several challenges, including dataset limitations, occlusions, and variations due to aging. Addressing these issues is crucial to the further development and adoption of this technology in real-world applications. This paper explores the advancements, challenges, and applications of ear biometrics. Section 2 discusses various feature extraction techniques and recognition algorithms employed in ear biometrics, covering traditional methods such as Principal Component Analysis (PCA) and Support Vector Machines (SVM), as well as modern deep learning techniques like Convolutional Neural Networks (CNNs). Section 3 explores different applications of ear biometrics, including its use in security and surveillance, forensic science, and healthcare. Section 4 addresses the key challenges faced by ear biometric systems, such as variability in ear images, limited datasets, and integration with other biometric modalities. Finally, Section 5 presents the conclusion and highlights future directions for research in ear biometrics. By examining these aspects, this paper aims to contribute to the growing body of knowledge on ear biometrics, emphasizing its potential, current limitations, and possible future improvements. Ear Biometric Recognition Methods Feature Extraction Techniques Feature extraction is a crucial step in ear biometric recognition, as it determines the distinguishing characteristics necessary for accurate classification. Geometric-based approaches focus on the structural features of the ear, such as the contours, curves, and distances between key points. These methods are computationally efficient and provide stable features, but their effectiveness can be affected by variations in pose and lighting conditions [2]. Texture-based methods, on the other hand, utilize spatial and frequency domain techniques to extract surface details, such as wavelet transforms and Local Binary Patterns (LBP). These methods have been shown to improve recognition accuracy, particularly when combined with machine learning classifiers [3]. Feature Extraction, Ear Recognition Process and Ear Anatomy With the rise of deep learning, convolutional neural networks (CNNs) have revolutionized ear biometric recognition. CNNs automatically learn hierarchical features from ear images, making them highly effective for classification tasks. Recent studies have demonstrated that deep learning approaches, particularly CNN architectures such as VGGNet, ResNet, and MobileNet, outperform traditional feature extraction methods in terms of accuracy and robustness [4]. Transfer learning has also been explored to fine-tune pre-trained deep learning models on ear biometric datasets, reducing the need for extensive labeled data while maintaining high performance [5]. Recognition Algorithms Once features are extracted, recognition algorithms play a key role in classifying and verifying identities. Traditional machine learning methods, such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA), have been widely used for dimensionality reduction and classification. These techniques transform high-dimensional feature spaces into lower-dimensional representations while preserving discriminative information [6]. Support Vector Machines (SVM) and k-Nearest Neighbors (kNN) have also been applied to ear biometric recognition, offering competitive performance, especially when paired with well-engineered features [7]. Deep learning-based recognition methods have further enhanced classification accuracy. CNN-based approaches achieve state-of-the-art performance by directly learning discriminative representations from raw ear images. Hybrid deep learning models, combining CNNs with Recurrent Neural Networks (RNNs) or transformers, have been proposed to capture spatial dependencies and contextual information, further improving recognition rates [8]. Recent research has also explored multi-modal biometric systems that integrate ear biometrics with other modalities, such as face and fingerprint recognition, to enhance security and reliability. Fusion techniques, including score-level and feature-level fusion, have been investigated to leverage complementary biometric information, resulting in improved identification accuracy [9]. Applications of Ear Biometrics Ear biometrics has been widely adopted in various domains, including security, forensic science, and healthcare. The unique shape and structure of the human ear make it an effective biometric trait for reliable identification and verification systems. One of the most significant applications of ear biometrics is in security and surveillance. Many high-security environments, such as airports, government facilities, and corporate offices, employ biometric-based access control systems. Unlike facial recognition, which can be affected by facial expressions or disguises, ear biometrics offers a stable and non-intrusive alternative. Studies have shown that ear biometrics can improve identification accuracy in security applications, particularly when combined with deep learning techniques for feature extraction and classification [10]. Recent advancements in deep learning-based ear recognition techniques have been integrated into real-time surveillance systems for automatic suspect identification and tracking [11]. In forensic science, ear biometrics plays a critical role in criminal investigations. Ear prints left at crime scenes can be used as evidence, similar to fingerprints. Researchers have developed sophisticated image processing techniques to analyze ear prints and match them against biometric databases for suspect identification [12]. Compared to other biometric traits, ears are less likely to be deliberately altered, making them valuable forensic evidence. Some studies have explored the role of ear biometrics in sex estimation for forensic investigations using machine learning models [13]. Law enforcement agencies have started incorporating ear biometrics into their forensic analysis toolkit, benefiting from high accuracy rates in ear-based identification systems [14]. The healthcare industry has also recognized the potential of ear biometrics for patient identification and secure medical record access. Traditional authentication methods, such as passwords and ID cards, pose security risks and are prone to loss or theft. Ear biometrics offers a seamless and secure alternative for verifying patient identity, ensuring that medical records remain protected. Some hospitals have begun integrating ear biometric authentication in electronic health record (EHR) systems, reducing medical identity fraud and improving patient safety [15]. Moreover, smartphone-based ear biometric applications have been proposed for patient identification in field settings, making healthcare access more efficient and secure [16]. In recent years, researchers have explored the integration of multi-modal biometric systems, where ear biometrics is combined with other biometric traits such as fingerprints, iris, and voice recognition. These hybrid systems have demonstrated increased recognition accuracy, making them suitable for high-security applications such as national ID programs and border control. Score-level and feature-level fusion techniques have been applied to combine biometric modalities, resulting in improved robustness against spoofing attacks and environmental variations [17]. To summarize the progress in ear biometric applications, Table 1 provides an overview of different technologies, their applications, and associated limitations. Overview of Technologies, Limitations, and Year of Application in Ear Biometrics 2014 Geometric-based feature extraction [2] Security & Access Control Affected by pose variations and lighting 2015 Smartphone-based ear biometrics [16] Patient identification Requires controlled imaging conditions 2019 Ear-based forensic analysis [13] Forensic Science Accuracy affected by occlusions and noise 2021 Deep learning (CNN) for security [11] Surveillance & Identification Requires large labeled training datasets 2023 Machine learning for forensic sex estimation [13] Forensic Science Variability in dataset representation 2024 Secure wearable ear biometrics [17] Multi-modal biometric authentication High computational cost Ear biometrics continues to gain traction across multiple domains, with ongoing research aimed at improving recognition accuracy, robustness, and real-world applicability. Challenges and Future Directions Traditional Biometrics and Their Limitations Biometric authentication methods have evolved over time, with fingerprints, facial recognition, iris scanning, and voice recognition being the most commonly used modalities. Fingerprint recognition has been one of the most widely adopted biometric methods due to its high uniqueness and ease of implementation. However, fingerprint-based systems face challenges related to hygiene, physical wear and tear, and spoofing attacks. For instance, during the COVID-19 pandemic, fingerprint scanners posed a significant public health risk as users had to physically touch sensors, increasing the chance of virus transmission [1]. Facial recognition has gained popularity, especially with the advancement of deep learning models, but it has limitations. Variations in lighting conditions, facial expressions, aging, and occlusions due to masks can significantly reduce recognition accuracy. The pandemic highlighted a major drawback of facial recognition, as mask-wearing obstructed a large portion of the face, leading to higher false rejection rates [2]. Iris recognition is considered one of the most reliable biometric modalities due to the high stability and uniqueness of the iris pattern. However, iris scanners require close-range imaging, and users often find them intrusive and uncomfortable. Additionally, high costs associated with iris recognition hardware limit its widespread deployment in public security applications [3]. Voice recognition, although useful for remote authentication, is highly susceptible to background noise, speech impairments, and replay attacks. Changes in an individual’s voice due to illness, aging, or environmental factors reduce the effectiveness of voice-based authentication systems [4]. Given these limitations, ear biometrics emerges as a promising alternative. Unlike fingerprints and facial recognition, ear features remain stable over time, unaffected by aging or facial expressions. Ear biometrics do not require physical contact, making them hygienic and suitable for pandemic-resistant authentication. Additionally, ear images can be captured from a distance, making it a non-intrusive and practical biometric method for surveillance, healthcare, and high-security applications [5]. Challenges in Ear Biometrics After the text edit has been completed, the paper is ready for the template. Duplicate the template file by using the Save As command, and use the naming convention prescribed by your conference for the name of your paper. In this newly created file, highlight all of the contents and import your prepared text file. You are now ready to style your paper; use the scroll down window on the left of the MS Word Formatting toolbar. Pose and Illumination Variability Variations in head position and lighting conditions significantly impact ear recognition accuracy. Side angles or slight head tilts can change the ear’s appearance, making it difficult for traditional feature-based methods to extract consistent patterns [1]. Additionally, shadows and uneven lighting introduce noise in the captured images, affecting classification performance. Deep learning models have shown improvements in robustness against pose variations, but extreme angles and poor lighting still pose challenges [2]. Occlusions and Accessories A major limitation of ear biometrics is occlusion due to hair, headphones, earrings, or hats. In real-world scenarios, individuals may have their ears partially or fully covered, making feature extraction difficult. Studies show that occluded ear images can significantly decrease recognition accuracy in biometric systems [3]. Removing obstructions manually is not feasible, and automatic occlusion detection and reconstruction methods are still under development. Dataset Limitations and Diversity Issues Unlike face and fingerprint databases, ear biometric datasets are relatively small and lack diversity. Large-scale datasets with varying conditions (age, ethnicity, gender, and environmental factors) are crucial for training robust models. The availability of high-quality, labeled ear images is limited, affecting deep learning model generalization [4]. Additionally, real-world datasets often contain imbalanced samples, where some demographic groups are underrepresented, leading to biased recognition models. Ethical and Privacy Concerns As with other biometric modalities, ear biometrics raises privacy and ethical concerns, particularly in public surveillance applications. Unauthorized ear data collection and tracking pose serious privacy risks. There is also a lack of clear legal and ethical frameworks governing the use of ear biometrics for law enforcement and security [5]. To mitigate privacy issues, secure data storage, encryption, and federated learning techniques should be incorporated into biometric systems. Real-Time Recognition and Computational Challenges Real-time ear recognition in dynamic environments (e.g., surveillance cameras or mobile authentication) remains a challenge. Factors such as motion blur, low-resolution images, and varying camera angles can degrade performance. High-accuracy models, such as deep CNNs and transformer-based architectures, require substantial computational resources, making them unsuitable for low-power edge devices [6]. Researchers are exploring lightweight deep learning models and optimization techniques to enhance real-time efficiency without sacrificing accuracy. Standardization and Interoperability Unlike fingerprints and facial recognition, ear biometrics lacks well-established international standards for feature extraction, classification, and database storage. The absence of a universal protocol for ear recognition hinders system interoperability and large-scale deployment [7]. Standardization efforts should focus on creating uniform feature representation methods and cross-platform compatibility to improve global adoption. Multi-Modal Integration Challenges Although multi-modal biometric systems that combine ear, face, iris, and fingerprint recognition improve accuracy and security, fusion techniques remain an ongoing challenge. Feature-level and score-level fusion require advanced algorithms to integrate different biometric traits effectively. Computational complexity, feature redundancy, and alignment issues must be addressed for successful multi-modal systems [8]. Future Directions The future of ear biometrics lies in overcoming existing challenges by leveraging advancements in deep learning, dataset expansion, privacy-preserving techniques, and multi-modal biometric integration. One of the key areas for improvement is the enhancement of deep learning models. Current convolutional neural networks (CNNs) have shown promising results in ear recognition, but further optimization is needed to handle pose variations, occlusions, and varying illumination conditions. The adoption of transformer-based architectures and hybrid deep learning models integrating recurrent neural networks (RNNs) or attention mechanisms could significantly improve robustness. Additionally, few-shot learning and self-supervised learning approaches may help in training models with limited labeled data, addressing the dataset scarcity problem [7]. Another crucial area of focus is expanding and diversifying ear biometric datasets. Most existing datasets are small and lack demographic diversity, leading to biased models that struggle with generalization. Researchers should work towards creating large-scale, high-resolution datasets that include individuals of various ethnicities, age groups, and environmental conditions. Additionally, synthetic data generation techniques, such as Generative Adversarial Networks (GANs), can be utilized to augment datasets and improve model training [4]. The establishment of publicly available benchmark datasets will also facilitate research collaboration and standardization efforts in ear biometric recognition. Privacy and security concerns remain a major challenge in biometric authentication, and future research must address these issues by developing privacy-preserving techniques. Implementing differential privacy, homomorphic encryption, and federated learning can enhance security while ensuring that user data remains protected from unauthorized access. Federated learning, in particular, allows machine learning models to be trained across decentralized devices without exposing raw biometric data, making it an effective approach for secure and privacy-conscious authentication systems [5]. Furthermore, real-time optimization and computational efficiency must be prioritized for practical applications. While high-accuracy deep learning models are powerful, they require substantial computational resources, limiting their deployment on mobile and edge devices. Future work should focus on designing lightweight deep learning models, such as MobileNet and EfficientNet, which offer high recognition accuracy with lower computational requirements. Additionally, advancements in hardware acceleration, such as Tensor Processing Units (TPUs) and optimized neural network chips, will further enable the deployment of real-time ear biometric authentication in various security and surveillance applications [6]. “‘latex Another promising direction is the integration of ear biometrics into multi-modal biometric systems. While ear recognition offers stability and uniqueness, combining it with other biometric traits such as face, iris, gait, and fingerprint recognition can further enhance accuracy and robustness. Feature fusion techniques, including score-level and decision-level fusion, should be explored to develop more secure and efficient multi-modal authentication systems. Additionally, the use of deep learning-based fusion techniques will enable adaptive biometric systems that can dynamically select the most reliable trait based on environmental conditions [8]. “‘latex Lastly, global standardization and interoperability are essential for the widespread adoption of ear biometrics. Unlike fingerprints and facial recognition, ear biometric systems lack well-established international standards, which poses challenges for cross-platform compatibility and large-scale deployment. Future efforts should focus on defining standardized feature extraction methods, data storage formats, and matching algorithms to facilitate global adoption and interoperability across different biometric authentication systems [9]. “‘latex By addressing these key areas, ear biometrics has the potential to emerge as a leading biometric authentication technology, providing secure, contactless, and reliable identity verification across various domains, including security, healthcare, law enforcement, and mobile authentication. As research continues to refine and improve ear biometric systems, the integration of advanced machine learning techniques, enhanced datasets, and privacy-centric approaches will pave the way for its practical and large-scale deployment in the near future. References 1. A. Pflug and C. Busch, “Ear biometrics: A survey of detection, feature extraction and recognition methods,” IET Biometrics, vol. 1, no. 3, pp. 114–126, 2012. Google Scholar 2. S. Singh and S.K. Singla, “A review on biometrics and ear recognition techniques,” International Journal of Advanced Research in Computer Science, vol. 3, no. 6, 2013. Google Scholar 3. A. Abaza et al., “A survey on ear biometrics,” ACM Computing Surveys, vol. 45, no. 2, pp. 22-34, 2013. Google Scholar 4. A. Booysens and S. Viriri, “Ear biometrics using deep learning: a survey,” Journal of Artificial Intelligence Research, vol. 22, pp. 123-140, 2022. Google Scholar 5. O.G. Oyebiyi et al., “A systematic literature review on human ear biometrics: Approaches, algorithms, and trend in the last decade,” MDPI Information, vol. 14, no. 3, p. 192, 2023. Google Scholar 6. R. Ahila Priyadharshini et al., “A deep learning approach for person identification using ear biometrics,” Applied Intelligence, vol. 51, no. 2, pp. 325-342, 2021. Google Scholar 7. K. Masaoud et al., “A review paper on ear recognition techniques: models, algorithms and methods,” International Journal of Biometrics, vol. 5, no. 4, pp. 411-421, 2013. Google Scholar 8. Z. Wang et al., “Review of ear biometrics,” Springer Engineering Applications of Artificial Intelligence, 2021. Google Scholar 9. M. Alva and A. Srinivasaraghavan, “A review on techniques for ear biometrics,” IEEE Xplore, 2019. Google Scholar 10. M. Hernandez-de-Menendez, ”Biometric applications in education,” The International Journal of Advanced Manufacturing Technology, vol. 112, no. 7, pp. 2245–2259, 2021. Available: Springer Google Scholar 11. S. Jahan, M. Chowdhury, and R. Islam, ”Robust user authentication model for securing electronic healthcare system using fingerprint biometrics,” Journal of Information Security and Applications, vol. 45, pp. 101–118, 2019. Available: Taylor & Francis Google Scholar 12. A. K. Jain and A. Ross, ”Bridging the gap: from biometrics to forensics,” Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 370, no. 1674, 2015. Available: Royal Society Publishing Google Scholar 13. S. Ö. Akyol and N. Sezgin, ”Investigation of the effect of ear measurements on sex estimation in forensic sciences using machine learning techniques,” Forensic Science International, vol. 321, pp. 112–124, 2023. Available: ResearchGate Google Scholar 14. G. Amirthalingam and G. Radhamani, ”A multimodal approach for face and ear biometric system,” International Journal of Computer Science Issues, vol. 10, no. 3, pp. 154–163, 2013. Available: ResearchGate Google Scholar 15. S. A. Bargal, A. Welles, C. R. Chan, and S. Howes, ”Image-based ear biometric smartphone app for patient identification in field settings,” Proceedings of the International Conference on Computer Vision Theory and Applications, 2015. Available: Scitepress Google Scholar 16. Z. Wang, Y. Wang, and J. Yang, ”Earslide: a secure ear wearables biometric authentication based on acoustic fingerprint,” Proceedings of the ACM International Conference on Mobile Systems, Applications, and Services, 2024. Available: ACM Google Scholar 17. M. Burge and W. Burger, ”Current and future biometrics: technology and applications,” Taylor & Francis, 2022. Available: Taylor & Francis Google Scholar Information & Authors Information Version history V1 Version 1 15 April 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords deep learning ear biometrics machine learning security applications Authors Affiliations Niraj K. Nagrale 0000-0002-1344-5522 [email protected] Bharti College of Engineering and Technology View all articles by this author Vishakha Nagrale G H Raisoni Institute of Engineering and Technology View all articles by this author Metrics & Citations Metrics Article Usage 1528 views 200 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Niraj K. Nagrale, Vishakha Nagrale. Progress and Challenges in Ear Biometrics for Secure Identification. Authorea . 15 April 2025. DOI: https://doi.org/10.22541/au.174471555.52636866/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174471555.52636866/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ff514d03b4ee2c5',t:'MTc3OTM4MjIzOA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();