Research Article
Fall Behavior Recognition Algorithm in Video Surveillance Based on Feature and Deep Learning
@INPROCEEDINGS{10.1007/978-3-030-94182-6_21, author={Hai-jing Zhou}, title={Fall Behavior Recognition Algorithm in Video Surveillance Based on Feature and Deep Learning}, proceedings={IoT and Big Data Technologies for Health Care. Second EAI International Conference, IoTCare 2021, Virtual Event, October 18-19, 2021, Proceedings, Part II}, proceedings_a={IOTCARE PART 2}, year={2022}, month={6}, keywords={Deep learning Feature learning Behavior recognition Intelligent video}, doi={10.1007/978-3-030-94182-6_21} }- Hai-jing Zhou
Year: 2022
Fall Behavior Recognition Algorithm in Video Surveillance Based on Feature and Deep Learning
IOTCARE PART 2
Springer
DOI: 10.1007/978-3-030-94182-6_21
Abstract
Aiming at the problem of low recognition rate and accuracy rate of human fall behavior recognition algorithm in current video surveillance, the human behavior feature extraction module is relatively backward. To solve this problem, a fall behavior recognition algorithm based on feature and deep learning is designed. The video image preprocessing is completed by dilation and erosion. The covariance matrix of image features is constructed to extract the features of human fall behavior. The standard image database is constructed, and the deep learning algorithm and neural network are used to complete the human fall behavior recognition in video surveillance. So far, the design of human fall behavior recognition algorithm in video surveillance based on feature and deep learning is completed. Through the example test, the application effect of the feature algorithm is better than that of the traditional algorithm.
