Due to the problem of insufficient dynamic human ear data, the Changchun University dynamic human ear (CCU-DE) database, which is a small sample human ear database, was developed in this study. The database fully considers the various complex situations and posture changes of human ear images, such as translation angle, rotation angle, illumination change, occlusion and interference, etc., making the research of dynamic human ear recognition closer to complex real-life situations, and increasing the applicability of human ear dynamic recognition. In order to test the practicability and effectiveness of the developed CCU-DE small sample database, we designed a dynamic human ear recognition system block diagram based on a deep learning model, which was pre-trained by a migration learning method. Aiming at multi-posture changes under different contrasts, translation and rotation motions, and with or without occlusion, simulation studies were conducted using the CCU-DE small sample database and different deep learning models, such as YOLOv3, YOLOv4, YOLOv5, Faster R-CNN, and SSD. The experimental results showed that the CCU-DE database can be well used for dynamic ear recognition, and it can be tested by using different deep learning models with higher test accuracy.
With the development of deep learning technology, more and more researchers are interested in ear recognition. Human ear recognition is a biometric identification technology based on human ear feature information and it is often used for authentication and intelligent monitoring field, etc. In order to make ear recognition better applied to practical application, real time and accuracy have always been very important and challenging topics. Therefore, focusing on the problem that the mAP@0.5 value of the YOLOv5s-MG method is lower than that of the YOLOv5s method on the EarVN1.0 human ear dataset with low resolution, small target, rotation, brightness change, earrings, glasses and other occlusion, a lightweight ear recognition method is proposed based on an attention mechanism and feature fusion. This method mainly includes the following several steps: First, the CBAM attention mechanism is added to the connection between the backbone network and the neck network of the lightweight human ear recognition method YOLOv5s-MG, and the YOLOv5s-MG-CBAM human ear recognition network is constructed, which can improve the accuracy of the method. Second, the SPPF layer and cross-regional feature fusion are added to construct the YOLOv5s-MG-CBAM-F human ear recognition method, which further improves the accuracy. Three distinctive human ear datasets, namely, CCU-DE, USTB and EarVN1.0, are used to evaluate the proposed method. Through the experimental comparison of seven methods including YOLOv5s-MG-CBAM-F, YOLOv5s-MG-SE-F, YOLOv5s-MG-CA-F, YOLOv5s-MG-ECA-F, YOLOv5s, YOLOv7 and YOLOv5s-MG on the EarVN1.0 human ear dataset, it is found that the human ear recognition rate of YOLOv5s-MG-CBAM-F method is the highest. The mAP@0.5 value of the proposed YOLOv5s-MG-CBAM-F method on the EarVN1.0 ear dataset is 91.9%, which is 6.4% higher than that of the YOLOv5s-MG method and 3.7% higher than that of the YOLOv5s method. The params, GFLOPS, model size and the inference time per image of YOLOv5s-MG-CBAM-F method on the EarVN1.0 human ear dataset are 5.2 M, 8.3 G, 10.9 MB and 16.4 ms, respectively, which are higher than the same parameters of the YOLOv5s-MG method, but less than the same parameters of YOLOv5s method. The quantitative results show that the proposed method can improve the ear recognition rate while satisfying the real-time performance and it is especially suitable for applications where high ear recognition rates are required.
To address the problems of false detection and the lack of accurate target bounding boxes in the process of detecting human ear recognition, a SE-YOLOv3 target detection algorithm with improved YOLOv3 is proposed. Based on the YOLOv3 algorithm, the effects of embedding the SE attention module into different positions of the model on the detection performance of the algorithm are studied and analyzed separately. It is proved that embedding the SE attention module in the parallel ResNet network in the backbone of the YOLOv3 model can effectively improve the detection accuracy of the algorithm. And it is experimentally demonstrated in CCU-DE and USTB human ear datasets.
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.
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.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.