Human body action recognition method based on a TP-STG framework

Abstract

The invention discloses a human body action recognition method based on a TP-STG framework, which comprises the following steps: taking video information as input, adding priori knowledge into an SVMclassifier, and providing a posteriori discrimination criterion to remove a non-personnel target; segmenting a personnel target through a target positioning and detection algorithm, outputting the personnel target in a target frame and coordinate information mode, and providing input data for human body key point detection; utilizing an improved posture recognition algorithm to carry out body partpositioning and correlation degree analysis so as to extract all human body key point information and form a key point sequence; a space-time graph is constructed on a key point sequence through an action recognition algorithm, the space-time graph is applied to multi-layer space-time graph convolution operation, action classification is carried out through a Softmax classifier, and human body action recognition in a complex scene is achieved. According to the method, the actual scene of the ocean platform is combined for the first time, and the provided TP-STG framework tries to identify worker activities on the offshore drilling platform for the first time by using methods of target detection, posture identification and space-time diagram convolution.

Description

technical field [0001] The invention belongs to the field of computer vision and image processing, and relates to a human action recognition method based on a TP-STG framework. Background technique [0002] With the popularization and wide application of surveillance cameras, massive video data has brought enormous pressure to manual recognition. Using artificial mode to analyze and judge video data, the bottleneck of manpower, experience and analysis capabilities of surveillance personnel restricts the application of intelligent behavior discrimination. overall effectiveness. In recent years, with the continuous advancement of research, the research on human action recognition has made some progress. The traditional methods are template matching, three-dimensional analysis and time series, which are the most common three methods, but the amount of calculation is relatively large, it is susceptible to noise interference, lacks robustness, and considers the integrity of acti...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a human action recognition method based on the TP-STG framework, which solves the problem of poor human action recognition effect and large error in complex scenes in the prior art, and can be used for human action in complex scenes Recognition, target positioning and pose recognition and estimation of people in dynamic scenes, to achieve accurate recognition of people's target actions in images

Images