GrabCut automatic segmentation algorithm in combination with RGBD data

Abstract

The invention discloses a GrabCut automatic segmentation algorithm in combination with RGBD data. Firstly, using data such as depth and color collected by a depth camera Kinect V2 for preliminarily segmenting a foreground containing a person through a coordinate system of a depth map corresponding to pixel points of a color map, and the foreground containing the person is used as a GrabCut mask frame; and then taking the depth data as a fourth channel of the Gaussian mixture model to improve an energy equation of a GrabCut algorithm, thereby realizing automatic segmentation of the figure contour. The improved Grabcut algorithm solves the problems that the segmentation is incomplete when the clothing color difference of the upper and lower bodies of the figure is large, the segmentation isinaccurate when the colors of the foreground and the background are similar and user interaction is needed in the original algorithm under a complex static background, and has the characteristics of automatic figure contour segmentation, strong real-time performance, high segmentation accuracy and the like. And drawing a figure contour by further utilizing polygon fitting to realize contour vectorization.

Description

technical field [0001] This technology focuses on image segmentation related technologies under the Kinect camera and real-time labeling and fitting of character outlines on RGBD images, which belongs to the field of image segmentation. In the video sequence collected by the Kinect camera, specific technologies and algorithms are used to remove background and other interference information, and only key data about the human body (such as body size, posture, dress color, etc.) A series of studies (such as behavior analysis, target tracking, template matching, etc.) provide effective data processing and lay a solid foundation. Background technique [0002] The accuracy of human contour extraction and segmentation speed determine the final effect of human behavior processing analysis and subsequent processing operations, and the fitting operation can retain the key features of the contour with the fewest points and reduce the amount of data. The focus of human contour recognit...

AI Technical Summary

Problems solved by technology

In recent years, some researchers have applied the GrabCut algorithm to the contour segmentation of people and achieved certain results. However, it needs to manually select the foreground area during its use, which is not suitable for real-time contour extraction of people.

Images