Review of Ear Biometrics

Wang, Zhaobin; Yang, Jing; Zhu, Ying

doi:10.1007/s11831-019-09376-2

Cited by 34 publications

(18 citation statements)

References 118 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the unique structure of the ear shape, ear images can provide a rich source of biometric information for constructing successful recognition systems. In addition, there are several desirable characteristics of the human ears which include: ease of capture from a distance, stability over time, ability to identify identical twins [1], and being insensitive to emotions and facial expressions [2], [3]. Given these appealing features we can build and develop reliable recognition systems on numerous devices in a non-intrusive and non-distracting manner [4], [5], [6].…”

Section: Introductionmentioning

confidence: 99%

Deep Convolutional Neural Networks for Unconstrained Ear Recognition

et al. 2020

View full text Add to dashboard Cite

This paper employs state-of-the-art Deep Convolutional Neural Networks (CNNs), namely AlexNet, VGGNet, Inception, ResNet and ResNeXt in a first experimental study of ear recognition on the unconstrained EarVN1.0 dataset. As the dataset size is still insufficient to train deep CNNs from scratch, we utilize transfer learning and propose different domain adaptation strategies. The experiments show that our networks, which are fine-tuned using custom-sized inputs determined specifically for each CNN architecture, obtain state-of-the-art recognition performance where a single ResNeXt101 model achieves a rank-1 recognition accuracy of 93.45%. Moreover, we achieve the best rank-1 recognition accuracy of 95.85% using an ensemble of fine-tuned ResNeXt101 models. In order to explain the performance differences between models and make our results more interpretable, we employ the t-SNE algorithm to explore and visualize the learned features. Feature visualizations show well-separated clusters representing ear images of the different subjects. This indicates that discriminative and ear-specific features are learned when applying our proposed learning strategies.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Convolutional Neural Networks for Unconstrained Ear Recognition

et al. 2020

View full text Add to dashboard Cite

show abstract

“…We refer to approaches that base on ear images acquired under constrained and unconstrained imaging conditions and highlight their achieved recognition performances. For a comprehensive review and analysis of existing work of ear recognition refer to [5], [11], [12].…”

Section: Related Workmentioning

confidence: 99%

Towards Explainable Ear Recognition Systems Using Deep Residual Networks

Alshazly¹,

Linse

Barth

et al. 2021

IEEE Access

View full text Add to dashboard Cite

This paper presents ear recognition models constructed with Deep Residual Networks (ResNet) of various depths. Due to relatively limited amounts of ear images we propose three different transfer learning strategies to address the ear recognition problem. This is done either through utilizing the ResNet architectures as feature extractors or through employing end-to-end system designs. First, we use pretrained models trained on specific visual recognition tasks, inititalize the network weights and train the fully-connected layer on the ear recognition task. Second, we fine-tune entire pretrained models on the training part of each ear dataset. Third, we utilize the output of the penultimate layer of the fine-tuned ResNet models as feature extractors to feed SVM classifiers. Finally, we build ensembles of networks with various depths to enhance the overall system performance. Extensive experiments are conducted to evaluate the obtained models using ear images acquired under constrained and unconstrained imaging conditions from the AMI, AMIC, WPUT and AWE ear databases. The best performance is obtained by averaging ensembles of fine-tuned networks achieving recognition accuracy of 99.64%, 98.57%, 81.89%, and 67.25% on the AMI, AMIC, WPUT, and AWE databases, respectively. In order to facilitate the interpretation of the obtained results and explain the performance differences on each ear dataset we apply the powerful Guided Grad-CAM technique, which provides visual explanations to unravel the black-box nature of deep models. The provided visualizations highlight the most relevant and discriminative ear regions exploited by the models to differentiate between individuals. Based on our analysis of the localization maps and visualizations we argue that our models make correct prediction when considering the geometrical structure of the ear shape as a discriminative region even with a mild degree of head rotations and the presence of hair occlusion and accessories. However, severe head movements and low contrast images have a negative impact of the recognition performance.

show abstract

“…As there is an increasing need to automatically recognize individuals, many methods are used for personal identification, and it has been an efficient field of research over the last decade [1], [2]. Passwords, and ID cards represented traditional methods for personal identification, but they can be purloin, forge, or forgotten, while biometric method has many characteristics, such as: universal, unique, perpetual, and could be measured [3]- [5].…”

Section: Introductionmentioning

confidence: 99%

An ear recognition system based on local wavelet subband energy distribution

Fadhil¹,

George

2021

IJEECS

View full text Add to dashboard Cite

The outer ear features have been used for many years in forensic science of recognition. Human ear is a valuable information provenance of data for individual identification/authentication. Ear meets biometric characteristic (universality, distinctiveness, permanence and collectability). Biometric system depending on ear image facing two major challenges; the first one is the localization of human ear area in given profile face image, and the second one is the selection of proper features to distinguish between individuals. In this work, we propose an alogorithm for ear recognition based on the local spatial energy distribution of wavelet sub-bands, because of wavelet transform has the ability to analyze the local feature of 2-D image by determining where the low frequency and high frequency areas are and it provides full description of the spatial distribution of the ear image. Nearest classifier are used to make a recognition decision in matching stage. The system was tested over a public database consist of 493 images. The attained recognition rate was (95.28%) and the achieved minimum equal error rate (EER) is 0.02%.

show abstract

Review of Ear Biometrics

Cited by 34 publications

References 118 publications

Deep Convolutional Neural Networks for Unconstrained Ear Recognition

Deep Convolutional Neural Networks for Unconstrained Ear Recognition

Towards Explainable Ear Recognition Systems Using Deep Residual Networks

An ear recognition system based on local wavelet subband energy distribution

Contact Info

Product

Resources

About