Deep learning-based human body motion recognition method of multi-channel image feature fusion

Abstract

The invention discloses a deep learning-based human body motion recognition method of multi-channel image feature fusion. The method comprises: (1) extracting original RGB pictures from videos, and calculating dynamic graphs and optical flow graphs of the segmented videos through the RGB pictures; (2) carrying out cropping operations on the input pictures to expand a training data set; (3) constructing a three-channel convolutional-neural-network, and respectively inputting lastly obtained video segments into the three-channel convolutional-neural-network to carry out training to obtain a corresponding network model; and (4) for a to-be-recognized video segment, extracting original RGB pictures, calculating dynamic graphs and optical flow graphs corresponding thereto, and obtaining a recognition result of a final motion category. According to the method, the three-channel convolutional-neural-network is utilized for learning essential features of data for original input of different morphologies, multi-channel dense fusion operations are carried out on the data of the three morphologies in the middle of the network, expression ability of the features is improved, and purposes of multi-channel information sharing and a high accuracy degree are achieved.

Description

technical field [0001] The present invention relates to the technical field of image processing and analysis, and more specifically, relates to a human action recognition method based on multi-channel image feature fusion based on deep learning. Background technique [0002] Human action recognition in video refers to a technology for human action recognition and classification by analyzing and processing visual feature information in video. This technology is widely used in intelligent video surveillance, behavior analysis, video retrieval and so on. Traditional human action recognition is based on manually designed feature training classifiers for action classification. At present, the strategy with the best effect of the traditional method is to extract features based on the improved dense trajectory (improved Dense Trajectory, iDT), combined with Fisher Vector (Fisher Vector, FV) modeling to identify human body work. In recent years, with the rapid development of deep ...

AI Technical Summary

Problems solved by technology

[0003] Although the traditional method based on manual features and the method based on deep learning have achieved good classification performance in human action recognition, due to the complexity of human action, the interference of background factors in the video, and the large intra-class difference of individual actions, the inter-class difference is very large. For small reasons, the current recognition algorithms have certain deficiencies, and cannot achieve high-precision human action recognition.

Images