Research on target detection algorithm based on improved YOLOv4

Tao Zhang; Yali Qi; Likun Lu; Qingtao Zeng; Wu Dong; Liqin Yu

doi:10.1117/12.2684175

14 June 2023 Research on target detection algorithm based on improved YOLOv4

Tao Zhang, Yali Qi, Likun Lu, Qingtao Zeng, Wu Dong, Liqin Yu

Proceedings Volume 12708, 3rd International Conference on Internet of Things and Smart City (IoTSC 2023）; 127080V (2023) https://doi.org/10.1117/12.2684175
Event: 3rd International Conference on Internet of Things and Smart City (IoTSC 2023）, 2023, Chongqing, China

Abstract

Classification, segmentation, and detection are the most important tasks in computer vision, and target detection as one of them is a hot research topic in the field of computer vision, which is widely used in medical, traffic, surveillance, etc. YOLOv4 and R-CNN have excellent target detection performance, and an improved YOLOv4 target detection algorithm is proposed to improve the real-time detection of small targets for target recognition. A priori frames are designed using the K-means clustering algorithm for adapting to different small and medium sizes; a feature layer is extracted according to the size of small and medium-sized labeled objects and four different feature layers are fused for detection; the Mish activation function is applied to the neck of the detection model to improve the detection performance. The experimental results show that the improved algorithm can effectively improve the detection accuracy.

Citation Download Citation

Tao Zhang, Yali Qi, Likun Lu, Qingtao Zeng, Wu Dong, and Liqin Yu "Research on target detection algorithm based on improved YOLOv4", Proc. SPIE 12708, 3rd International Conference on Internet of Things and Smart City (IoTSC 2023）, 127080V (14 June 2023); https://doi.org/10.1117/12.2684175

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