Application of lightweight YOLOv4 in vehicle detection

Feng Tian; Lichen Wu; Weibo Fu; Xiaojun Huang

doi:10.1117/12.2640131

1 August 2022 Application of lightweight YOLOv4 in vehicle detection

Feng Tian, Lichen Wu, Weibo Fu, Xiaojun Huang

Proceedings Volume 12257, 4th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2022); 1225719 (2022) https://doi.org/10.1117/12.2640131
Event: 4th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2022), 2022, Guangzhou, China

Abstract

Aiming at the low detection speed caused by the complexity of YOLOv4 target detection algorithm,a lightweight YOLOv4 algorithm is proposed.Firstly,the backbone of YOLOv4 is replaced by the lightweight network MobilenetV3 to reduce the number of network parameters;The 3×3 standard convolution in PANet is replaced by the depthwise separable convolution to reduce computational load;K-means is used to acquire the anchor box.And improved YOLOv4 is tested by using Udacity dataset.According to the result,we can find that under the premise of ensuring the detection accuracy, FPS is increased by 74% and the speed of the improved YOLOv4 is improved.

Citation Download Citation

Feng Tian, Lichen Wu, Weibo Fu, and Xiaojun Huang "Application of lightweight YOLOv4 in vehicle detection", Proc. SPIE 12257, 4th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2022), 1225719 (1 August 2022); https://doi.org/10.1117/12.2640131

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available