Abstract
Violent behavior recognition is an important direction of behavior recognition research. For traditional violent behavior recognition algorithms, there is too much background information when processing video information, which will cause greater interference in feature extraction, so the recognition accuracy is not high. Improved on the basis of effective recurrent convolutional network, a long-term recurrent convolutional network with attention mechanism is proposed. In the video preprocessing stage, a variety of attention mechanisms are introduced. In the feature extraction stage, the lightweight end-to-side neural network architecture GhostNet and convLSTM are selected to build a long-term recurrent convolutional network. The global average pooling and fully connected layer are used in the classification process. The combined approach realizes the classification of behaviours. The final results show that in the Hockey dataset, the algorithm in this paper has increased by 0.4% compared to LRCN, in the RWF-2000 dataset with more samples, it has increased by 10.5% compared to LRCN, and has increased by 1.75% compared to I3D, indicating that the algorithm in this paper can effectively suppress the background information. Interference, improve the performance of the algorithm.
Highlights
Violent behavior recognition has broad application prospects in the field of intelligent security and is a key field of behavior recognition research
In order to effectively extract the key information of the human body in the video and reduce the impact of complex background information on the model, this paper uses a variety of attention mechanisms for data preprocessing on the basis of the time series model long-term recurrent convolutional neural network, and uses lightweight convolution Network GhostNet[10] extracts the spatial information of the video, uses ConvLSTM to extract the timing information, and uses global average pooling (GAP) and fully connected layer for classification
Before GhostNet, the spatial and channel squeeze & excitation (scSE), iscSE, and spatial attention mechanism (SAM) modules are added for experiments
Summary
Violent behavior recognition has broad application prospects in the field of intelligent security and is a key field of behavior recognition research. The framework of deep learning currently commonly used feature extraction methods mainly include Two-stream CNN model, spatio-temporal model and time series model. The long-term recurrent CNN of time series model is mixed with a large amount of background information when extracting spatiotemporal information, which leads to low accuracy of behavior recognition. In order to effectively extract the key information of the human body in the video and reduce the impact of complex background information on the model, this paper uses a variety of attention mechanisms for data preprocessing on the basis of the time series model long-term recurrent convolutional neural network, and uses lightweight convolution Network GhostNet[10] extracts the spatial information of the video, uses ConvLSTM to extract the timing information, and uses global average pooling (GAP) and fully connected layer for classification. The results show that the algorithm in this paper can reduce the interference of background information and improve the accuracy of the network
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
