arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Computer Vision and Pattern Recognition

Research on Human Behavior Feature Recognition and Intelligent Early Warning Methods in Safety Supervision Scene Video based on Yolov7

Wanwan Cao, Longyue Li, Shuai Gong, Xiaoling Dong

· Journal of Physics Conference Series

See on the map

Abstract

Abstract With the continuous improvement of infrastructure construction in China, the construction of various supporting power facilities is constantly followed up. A large number of ultra-high voltage or high-voltage power facilities, such as high-voltage transmission towers, transformers, or substations, are gradually emerging in populated areas, farmland, or fishponds where people still pass by sometimes. In such circumstances, people who do not have the qualification to contact power facilities at times approach, contact, or even climb these dangerous high-voltage transmission and distribution facilities illegally for various reasons, which not only poses a huge hidden peril or even threat to the safety of people’s lives and property but also cause a huge negative impact on the smooth operation of power infrastructure. To solve this intractable problem, this topic proposes a pedestrian detection and recognition algorithm, which can accurately distinguish between ordinary pedestrians and power grid maintenance inspectors based on YOLOv7, providing a reliable method for timely detection and early warning of pedestrians who may contact power facilities.

More like this