Abstract:Ear recognition is a biometric identification technology based on human ear feature information, which can not only detect the human ear in the picture but also determine whose human ear it is, so human identity can be verified by human ear recognition. In order to improve the real-time performance of the ear recognition algorithm and make it better for practical applications, a lightweight ear recognition method based on YOLOv5s is proposed. This method mainly includes the following steps: First, the MobileNetV3 lightweight network is used as the backbone network of the YOLOv5s ear recognition network. Second, using the idea of the Ghostnet network, the C3 module and Conv module in the YOLOv5s neck network are replaced by the C3Ghost module and GhostConv module, and then the YOLOv5s-MG ear recognition model is constructed. Third, three distinctive human ear datasets, CCU-DE, USTB, and EarVN1.0, are collected. Finally, the proposed lightweight ear recognition method is evaluated by four evaluation indexes: mAP value, model size, computational complexity (GFLOPs), and parameter quantity (params). Compared with the best results of YOLOv5s, YOLOv5s-V3, YOLOv5s-V2, and YOLOv5s-G methods on the CCU-DE, USTB, and EarVN1.0 three ear datasets, the params, GFLOPS, and model size of the proposed method YOLOv5s-MG are increased by 35.29%, 38.24%, and 35.57% respectively. The FPS of the proposed method, YOLOv5s-MG, is superior to the other four methods. The experimental results show that the proposed method has the performance of larger FPS, smaller model, fewer calculations, and fewer parameters under the condition of ensuring the accuracy of ear recognition, which can greatly improve the real-time performance and is feasible and effective.

Human Ear Recognition Using HOG with PCA Dimension Reduction and LBP

Learning a Locality Preserving Subspace for Visual Recognition.

[Study of Biometric Identification of Heart Sound Base on Mel-Frequency Cepstrum Coefficient].

Human Ear Image Recognition Method Using PCA and Fisherface Complementary Double Feature Extraction

3D Ear Recognition Using Local Salience and Principal Manifold

An Efficient Approach of 3D Ear Recognition

A real-time hand gesture recognition method

An HOG-CT Human Detector with Histogram-Based Search.

An Efficient and Lightweight Method for Human Ear Recognition Based on MobileNet

Face Feature Extraction and Recognition Via Local Binary Pattern and Two-Dimensional Locality Preserving Projection

Human ear print recognition based on fusion of difference theoretic texture and gradient direction pattern features

An Empirical Study for Human Behavior Analysis

B3D-EAR: Binarized 3D descriptors for ear-based human recognition

Panoramic face and ear image stitching in multi-modal recognition

Lightweight YOLOv5s Human Ear Recognition Based on MobileNetV3 and Ghostnet

A Study of Two Image Representations for Head Pose Estimation

Method for Human Ear Localization in Controlled and Uncontrolled Environments

Research on Recognition Method Based on Spectral Regression and Back Propagation Neural Network

An Exploration of Face Recognition Methods Based on LBP Algorithm and PCA Analysis

Human Detection Based on Fusion of Histograms of Oriented Gradients and Main Partial Features

Face Recognition Using Cascading of HOG and LBP Feature Extraction