Paper
20 October 2023 Implementation of campus supervision system based on Yolov5 and SlowFast optimization model
Ninghai Zhang, Yanchao Wang
Author Affiliations +
Proceedings Volume 12814, Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023); 1281409 (2023) https://doi.org/10.1117/12.3010350
Event: Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023), 2023, Chongqing, China
Abstract
With the end of the COVID-19 epidemic in China, colleges and universities across the country have restored the original order and implemented offline teaching. However, the turnover rate of campus personnel is also gradually increasing, and the traditional campus supervision system has pain points such as old and complex system and low efficiency. In order to solve the above problems and better assist the monitoring and management of the campus, we propose and make a campus monitoring system based on computer vision technology. We use the Yolov5 model for object detection, the optimized deepsort algorithm for object tracking, and the SlowFast network model open sourced by the Facebook team for behavior prediction. Through verification and actual test, the model has a range accuracy of 0.5 meters to 30 meters, and has strong anti-physical interference. Therefore, we conclude that this model has better efficiency and a wide range of application scenarios.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Ninghai Zhang and Yanchao Wang "Implementation of campus supervision system based on Yolov5 and SlowFast optimization model", Proc. SPIE 12814, Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023), 1281409 (20 October 2023); https://doi.org/10.1117/12.3010350
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Object detection

Education and training

Data modeling

Feature extraction

Video

Mathematical optimization

Visual process modeling

RELATED CONTENT


Back to Top