Human body posture recognition method based on video processing and threshold setting of decision tree

Abstract

The invention discloses a human body posture recognition method based on video processing and threshold setting of decision tree, belongs to a human body posture recognition technology, and aims to solve the problems that an existing human body posture recognition algorithm based on video processing has a great error when distinguishing different postures and is poor in recognition effect. The method comprises the steps that judging whether a working mode is established for a human posture database or not, and when the working mode is established for the human posture database, executing the step of establishing the human posture feature threshold interval database based on the decision tree; otherwise, executing an attitude recognition step based on human body attitude characteristic threshold interval matching, determining threshold intervals of different human body attitudes through the decision tree classifier, and judging the same attitude by utilizing a plurality of human body attitude characteristics, so that the defect of fixed threshold setting in traditional attitude recognition is overcome, and the accuracy of attitude recognition is improved.

Description

technical field [0001] The invention belongs to human body gesture recognition technology. Background technique [0002] Human body posture recognition is a research direction with great development potential in the field of computer vision, and it is also a very important research direction in the field of pattern recognition and artificial intelligence. It is widely used in home furnishing, sports training and other fields. [0003] Human body posture recognition is mainly divided into two categories according to the way of raw data: non-visual sensor data collection method and visual sensor data collection method. Based on non-visual methods, wearable sensors or sensors arranged in human activity areas are used to collect motion information data, such as acceleration and angular velocity of human motion, although the data collected by this scheme can be simply filtered and processed. It is used for gesture recognition in the later stage, but this solution requires peopl...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the existing human body posture recognition algorithm based on video processing will have a large error when distinguishing different postures, and the recognition effect is poor, and to provide a human body posture recognition algorithm based on video processing and decision tree setting threshold. gesture recognition method

Images