Improved mine pedestrian detection algorithm based on YOLOv4-Tiny

Fengbo Wu; Wei Liu; Shuqi Wang; Gang Zhang

doi:10.1117/12.2661076

2 February 2023 Improved mine pedestrian detection algorithm based on YOLOv4-Tiny

Fengbo Wu, Wei Liu, Shuqi Wang, Gang Zhang

Proceedings Volume 12462, Third International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022); 124622K (2023) https://doi.org/10.1117/12.2661076
Event: International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022), 2022, Xi'an, China

Abstract

In view of the complex underground structure and harsh environment of coal mines, it is easy to cause problems such as low pedestrian detection accuracy and missed detection. An improved mine pedestrian detection algorithm based on YOLOv4-Tiny was proposed. The algorithm introduced spatial pyramid pooling (SPP) module after 13×13 features, which realized the extraction of global and local features of image information, and finally improved the detection precision of the system. The mAP of the improved algorithm was 91.98%, which was 2.32% higher than the original algorithm, the precision increased by 3.87% and the recall by 0.93%. The experimental results showed that the algorithm improved the detection of the mine pedestrian system effectively and had a better detection effect.

Citation Download Citation

Fengbo Wu, Wei Liu, Shuqi Wang, and Gang Zhang "Improved mine pedestrian detection algorithm based on YOLOv4-Tiny", Proc. SPIE 12462, Third International Symposium on Computer Engineering and Intelligent Communications (ISCEIC 2022), 124622K (2 February 2023); https://doi.org/10.1117/12.2661076

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