Automatic ear detection and feature extraction using Geometric Morphometrics and convolutional neural networks

Cintas, Celia; Quinto‐Sanchez, Mirsha; Acuña, Víctor Hugo; Paschetta, Carolina; Azevedo, Soledad de; Cerqueira, Caio César Silva de; Virginia, R.; Gallo, Carla; Poletti, Giovanni; Bortolini, María Cátira; Canizales‐Quinteros, Samuel; Rothhammer, Francisco; Bedoya, Gabriel; Ruiz-Linares, Andrés; González‐José, Rolando; Delrieux, Claudio

doi:10.1049/iet-bmt.2016.0002

Cited by 50 publications

(36 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The technique has used 1000 uncropped images (10 images per subject, captured in an unconstrained environment) of the Annotated Web Ears (AWE), in which 750 images are used for training and 250 images are used for testing. Cintas et al introduced a technique for ear detection and description. The method uses Geometric Morphometrics and Deep Learning for automatic ear detection, where the training instances manually land marked points are given as input to the deep learning network.…”

Section: Related Workmentioning

confidence: 99%

Unconstrained ear detection using ensemble‐based convolutional neural network model

Ganapathi

Prakash

Dave

et al. 2019

Concurrency and Computation

View full text Add to dashboard Cite

This paper presents a technique for ear detection from 2D profile face images that is capable of significantly reducing the false positives. In an ear biometrics system, recognition performance highly depends on the performance of the ear detection module. The trade-off between the complexity and the false positive detection is one of the essential component, where the complexity of a system increases proportionally to achieve a zero false positive rate detection.In literature, available ear detection techniques based on handcrafted features face challenges with low-quality acquired images affected by illumination, occlusion, and pose variations. We propose an ear detection technique using ensemble of convolutional neural network (CNN). The first part of the technique trains three models of CNN on a given dataset, whereas in later part, weighted average of the outputs of trained models is utilized to detect the ear regions. The used ensemble models show better performance as compared to the case when each individual model is used standalone. The proposed technique is being evaluated on two databases, viz, IIT Indore-Collection A (IIT-Col A) database and annotated web ear (AWE) database. Experimental results of ear detection demonstrate the superior performance of the proposed technique over other state-of-the-art techniques in handling illumination, occlusion, and pose variations. KEYWORDS deep learning, ear detection, ensemble model, unconstrained environment INTRODUCTIONIn recent years, ear biometrics has been considered a reliable biometrics and has got attention over face mainly due to its invariant behavior to the age of a person. Biometrics systems based on face have to consider the changes occurring in the face with a change in age or expression.The ear of a person remains consistent in its shape and it is fixed in its position at the center of the profile image. To obtain reliable performance, face images are required to be captured in the controlled setting, while in case of ear, it can be captured with a foreseeable background due to its fixed side face background. An ear biometric system mainly consists of two processes, ie, Ear detection and Ear recognition. Majority of the ear biometrics techniques in the literature use manually cropped ear images to perform ear recognition process. Furthermore, the ear biometric techniques that crops ear automatically are able to detect ear only when the ear has small variations in the side face image. In cases of pose or scale variations, these techniques are found to be inefficient. Additionally, in a real-time environment, these techniques lack in being fully automatic approaches and speed.Performance of an ear detection technique based on handcrafted feature learning methods depends on the dataset used for learning the features. 1 These methods are prone to produce erroneous results when used with the images captured in unconstrained settings like different illumination, background, and pose. Tasks like extraction of suitable features along with the discriminati...

show abstract

Section: Related Workmentioning

confidence: 99%

Unconstrained ear detection using ensemble‐based convolutional neural network model

Ganapathi

Prakash

Dave

et al. 2019

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…The approach published in [9] was based on geometric morphometrics and deep learning. It was proposed for automatic ear detection and feature extraction in the form of landmarks.…”

Section: Ear Detectionmentioning

confidence: 99%

Ear Detection Using Convolutional Neural Network on Graphs with Filter Rotation

Tomczyk

Szczepaniak

2019

Sensors

View full text Add to dashboard Cite

Geometric deep learning (GDL) generalizes convolutional neural networks (CNNs) to non-Euclidean domains. In this work, a GDL technique, allowing the application of CNN on graphs, is examined. It defines convolutional filters with the use of the Gaussian mixture model (GMM). As those filters are defined in continuous space, they can be easily rotated without the need for some additional interpolation. This, in turn, allows constructing systems having rotation equivariance property. The characteristic of the proposed approach is illustrated with the problem of ear detection, which is of great importance in biometric systems enabling image based, discrete human identification. The analyzed graphs were constructed taking into account superpixels representing image content. This kind of representation has several advantages. On the one hand, it significantly reduces the amount of processed data, allowing building simpler and more effective models. On the other hand, it seems to be closer to the conscious process of human image understanding as it does not operate on millions of pixels. The contributions of the paper lie both in GDL application area extension (semantic segmentation of the images) and in the novel concept of trained filter transformations. We show that even significantly reduced information about image content and a relatively simple, in comparison with classic CNN, model (smaller number of parameters and significantly faster processing) allows obtaining detection results on the quality level similar to those reported in the literature on the UBEAR dataset. Moreover, we show experimentally that the proposed approach possesses in fact the rotation equivariance property allowing detecting rotated structures without the need for labor consuming training on all rotated and non-rotated images.In this work, ear images are considered [2,3]. There are several factors that make the research and application of ear recognition important and attractive: people can be identified on that basis; this biometric does not change in time; and its gathering does not create a great deal of controversy. Furthermore, technology enables acquisition of ear images from a distance, which may be of great importance for police investigations and in security systems. It has, however, at least two consequences. Firstly, before the individual can be identified [4,5], the localization of the ear must be precisely detected. Secondly, since the acquisition process is not controlled, the orientation of the head can significantly vary. As a result, the need for detection methods arises, which will be able to cope with ear transformations.Convolutional neural networks (CNNs) became a state-of-the-art solution of many image analysis problems [6,7]. Their main component is convolutional layers designed to apply the same, trainable filters (represented by a rectangular mask) locally to every part of the image. It allows extracting the spatial distribution of image characteristic features (so-called feature maps). These kinds of layers together...

show abstract

“…Learning-based detection methods have been proposed owing to their accuracy and robustness advantages for ear detection, e.g., the AdaBoost algorithm and its modified version [24,25] and ear detection involving faster region-based convolutional neural network (Faster R-CNN) frameworks [26], a modified multiple scale faster region-based convolutional neural network [14], geometrics morphometrics and deep learning [27] and convolutional encoder-decoder networks for ear detection [28]. Automatic ear detection methods have achieved good performance under no occlusion or noise conditions; for example, the accuracy of the method in [12] is 93.34% for 700 images, and that in [19] is 97.57% for 267 images.…”

Section: Automatic 2d Ear Detectionmentioning

confidence: 99%

Local and Holistic Feature Fusion for Occlusion-Robust 3D Ear Recognition

Zhu

Zhang

2018

Symmetry

View full text Add to dashboard Cite

Occlusion over ear surfaces results in performance degradation of ear registration and recognition systems. In this paper, we propose an occlusion-resistant three-dimensional (3D) ear recognition system consisting of four primary components: (1) an ear detection component, (2) a local feature extraction and matching component, (3) a holistic matching component, and (4) a decision-level fusion algorithm. The ear detection component is implemented based on faster region-based convolutional neural networks. In the local feature extraction and matching component, a symmetric space-centered 3D shape descriptor based on the surface patch histogram of indexed shapes (SPHIS) is used to generate a set of keypoints and a feature vector for each keypoint. Then, a two-step noncooperative game theory (NGT)-based method is proposed. The proposed symmetric game-based method is effectively applied to determine a set of keypoints that satisfy the rigid constraints from initial keypoint correspondences. In the holistic matching component, a proposed variant of breed surface voxelization is used to calculate the holistic registration error. Finally, the decision-level fusion algorithm is applied to generate the final match scores. Evaluation results from experiments conducted show that the proposed method produces competitive results for partial occlusion on a dataset consisting of natural and random occlusion.

show abstract

Automatic ear detection and feature extraction using Geometric Morphometrics and convolutional neural networks

Cited by 50 publications

References 46 publications

Unconstrained ear detection using ensemble‐based convolutional neural network model

Unconstrained ear detection using ensemble‐based convolutional neural network model

Ear Detection Using Convolutional Neural Network on Graphs with Filter Rotation

Local and Holistic Feature Fusion for Occlusion-Robust 3D Ear Recognition

Contact Info

Product

Resources

About