Smoking Behavior Recognition Method Based on Self-Attention Convolutional Neural Network
Convolutional Neural Network has become the mainstream method in the current visual behavior recognition task because of its powerful feature extraction ability.In order to effectively monitor and warn smoking behavior in public places,this paper proposes a smoking behavior recognition method based on Self-Attention Convolutional Neural Network.By analyzing the key features of smoking behavior in images and videos,an efficient Convolutional Neural Network model is designed.This model can accurately and efficiently extract the key features of images by introducing Self-Attention mechanism to achieve accurate recognition of smoking behavior.The experimental results show that the proposed method has good recognition effect and robustness in different scenarios,and has high practical value.
国家科技期刊平台
NSTL
万方数据
