Complex behavior recognition method

Abstract

The invention discloses a complex behavior recognition method. The complex behavior recognition method comprises: acquiring three-dimensional bone joint point information of a target motion by using asensor; pre-processing the joint point information, and normalizing a coordinate system; extracting a motion trajectory of each joint point, projecting the motion trajectory to three two-dimensionalplanes; extracting a motion vector between each two frames and the length and the direction angle thereof, using the k-means algorithm to perform clustering to acquire motion primitives, and performing counting to acquire histograms; calculating the weight of each joint point is calculated to form a descriptor according to the time pyramid, the time information, and values of the clusters of allthe histograms; performing SVM classification to achieve motion recognition. The invention can perform feature extraction and effective represent on the motion bone joint point information, and can improve the accuracy of the motion recognition; all the motion information is completely retained, and the motion reconstruction can be performed; all motion classes are clustered, and motion features of the human are captured globally; low-level features are used, the computational difficulty is lowered, the motion recognition efficiency is improved, and the the real-time requirements of the systemis satisfied.

Description

technical field [0001] The invention relates to a complex behavior recognition method, which belongs to the technical field of image recognition. Background technique [0002] Action recognition is a research hotspot in the field of machine vision. Action recognition methods have been widely used in human-computer interaction, virtual reality, video retrieval, and security monitoring. With the development of depth cameras, the information of human bone joint points can be directly obtained, and the action recognition method based on bone features has greatly improved the accuracy of action recognition. Despite many related studies and exciting results, the effective description of human motion is still a challenging task. [0003] Many methods extract a variety of high-level features from bone information, and then combine them in a certain form to form a descriptor, but the descriptor constructed by this combination mode is not complete, and there is always a loss of motio...

AI Technical Summary

Problems solved by technology

[0003] Many methods extract a variety of high-level features from bone information, and then combine them in a certain form to form a descriptor, but the descriptor constructed by this combination mode is not complete, and there is always a loss of motion information.

On the other hand, many methods train different action classes separately, which leads to bias in the global features of human actions in the description of each individual class action

At the same time, the method of using high-level features still has the problem of high computational cost

Images