A Method for Extracting Human Thin Centerline in Depth Image

Abstract

The invention discloses a method for extracting the thinned centerline of a human body in a depth image. Firstly, an outline image of a human body is obtained, and an image smoothing process is performed on the outline image of the human body, and then an image of a self-occluded part is obtained by adopting a hierarchical segmentation processing method of the depth image. Finally, the image of the body area and the image of the self-occlusion area are respectively thinned by the image thinning algorithm to obtain the thinning center line of the human body. The method of extracting the thinning centerline of the human body in the depth image designed by the present invention cleverly integrates the idea of ​​thinning line acquisition into the extraction of occlusion areas, solves the problem of self-occlusion, and obtains the thinning line containing depth information at the same time, which has the advantages of Real-time, high-efficiency and other advantages.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for extracting a thinned centerline of a human body in a depth image. Background technique [0002] With the rapid development of computer technology, "somatosensory" operation technology has increasingly entered people's lives as a new way of human-computer interaction. Robust human skeleton recognition is widely used in human-computer interaction, games, security, remote monitoring, and even medical treatment. There are a wide range of applications, and the emergence of depth cameras has made human skeletal model recognition cheaper and easier to implement. [0003] The skeleton maintains the original graph topology and is a powerful abstraction method to describe the basic topology of objects in two-dimensional or three-dimensional space. Human skeleton extraction using depth images can well overcome the interference of factors such as illumination changes, ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art, provide a method for extracting the thinning centerline of the human body in the depth image, and solve the problem that the traditional 2D thinning algorithm cannot obtain the thinning line of the self-occluded part, that is, the occluded part in front of the body , is a new method that can adapt to depth data

Images