Layered observation vector decomposed hidden Markov model-based method for identifying behaviors

Abstract

The invention relates to a layered observation vector decomposed hidden Markov model-based method for identifying behaviors. The method comprises the following aspects of: target detection, target tracking, characteristic extraction, motion modeling, behavior identification and the like. The method improves the conventional hidden Markov model by aiming at deficiencies of a present method for identifying behaviors of multi-user, separates an individual state and an interactive state in the model to stand out interactive relation among targets, and reduces calculated amount by decomposing observation nodes. Meanwhile, the model allows the number of the targets taking the movement to be changed, so that the method has more flexibility in characteristic selection problem than the traditionalmethod and allows simultaneously using discrete characteristics and continuous characteristics. The method plays an important role in analyzing the multi-user interactive behaviors in the fields of video monitoring, and content-based video retrieval and the like.

Description

technical field [0001] The present invention relates to a human body behavior recognition method in video, in particular to a multi-person behavior recognition method based on a hierarchical observation vector decomposition hidden Markov model, which is applied to pattern recognition, artificial intelligence, computer vision, image processing and other technologies field. Background technique [0002] In recent years, video surveillance technology has attracted the attention of the society, especially since the "911" terrorist attacks, Madrid bombings and London bombings, the research and development of automatic video surveillance products for computers has become more urgent. Now, cameras can be seen everywhere in many communities, campuses, and streets. However, the current video surveillance technology is far from intelligent enough, and the monitoring work is basically done manually. However, due to factors such as human energy, physical strength, and labor costs, the ...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims at the deficiencies of current multi-person behavior recognition methods in video, and provides a behavior recognition method based on Hidden Markov Model of Hierarchical Observation Vector Decomposition, which can realize multi-person behavior recognition and allow the number of targets participating in the movement Changes occur, and continuous features and discrete features appear at the same time to solve the problem of expressing interactive relationships in multi-person behavior recognition

Images