The application of Deep Learning Techniques in biometrics has grown significantly during the last decade. The use of deep learning models in ear biometrics is restricted due to the lack of large ear datasets. Researchers employ transfer learning based on several pretrained models to overcome the limitations. For the unconstrained AWE ear dataset, traditional Machine Learning (ML) techniques and hand-crafted features fall short of providing a good recognition accuracy. This paper evaluates the influence of separating left and right ears and the effect of occlusion on the recognition accuracy in AWE dataset. The left and right ear of a person need not be identical. A study by separating the left and right ear into two different datasets is carried out with the pretrained ResNet50 based model. There is a remarkable increase in accuracy when the left and right ear images are independently considered. A new data augmentation technique, incorporating occlusion, is also proposed and experimented with the ResNet50 based model.
Biometric is one of the growing fields used in security, forensic and surveillance applications. Various types of physiological and behavioral biometrics are available today. Human ear is a passive physiological biometric. Ear is an important biometric trait due to many advantages over other biometric modalities. Because of its complex structure, face image detection is very challenging. Detection deals with finding or localizing the position of ear in the given profile face image. Various methods like manual, semiautomatic and automatic techniques are used for ear detection. Automatic ear localization is a complex process compared to manual ear cropping. This paper presents an empirical study and evaluation of four different existing ear detection techniques with our proposed method based on banana wavelets and circular Hough transform. A comparative analysis of the five algorithms in terms of detection accuracy is presented. The detection accuracy was calculated by means of manual as well as automatic verification.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.