Open AccessResearch on safety helmet wearing YOLO-V3 detection technology improvement in mine environment
Author(s) -
Fuchuan,
Rongxin Wang
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1345/4/042045
Subject(s) - residual , object detection , convolution (computer science) , computer science , artificial intelligence , process (computing) , feature (linguistics) , safety monitoring , scale (ratio) , deep learning , training (meteorology) , computer vision , pattern recognition (psychology) , algorithm , artificial neural network , geography , cartography , linguistics , philosophy , microbiology and biotechnology , meteorology , biology , operating system
The existing AI object detection technology cannot meet the demand of safety helmet wearing detection accuracy in mine environment. This paper studies YOLO correlation algorithm, establishing an optimal model based on YOLO-V3, combines the deep residual network technology with the multi-scale convolution feature based on the YOLO-V3 detection algorithm, combines the multi-scale detection training and adjusts the loss function in the training process. The experimental results show that with satisfying the detection speed, safety helmet wearing detection accuracy in mine environment is significantly improved.