Video image splitting method based on restrain spectral clustering and markov random field

Abstract

The invention belongs to the technical field of computer vision, image treatment and pattern recognition, and particularly relates to a video image splitting method based on restrain spectral clustering and a markov random field. According to the method, a similarity matrix between two pixels is calculated based on the static characteristics of images, and the motion characteristic is added in a spectral clustering framework to be used as a restrain. In comparison with the conventional simple weighting method, two pieces of information with different reliability degrees are utilized well in the treatment method. Besides, as motion information coding is used as a restrain, only sparse point motion trains are required, so that reliable long-time motion information can be utilized to obtain an accurate and dense splitting result. Further, space smooth information of each pixel is coded into corresponding restrains by constructing a markov random field model, so that the video image splitting effect is relatively precise.

Description

technical field [0001] The invention relates to the technical fields of computer vision, image processing, pattern recognition and the like, in particular to a video image segmentation method based on constrained spectral clustering and Markov random field. Background technique [0002] Video segmentation refers to a technique that divides a video sequence into some non-overlapping spatio-temporal regions according to its characteristics. Video segmentation is a key technology of computer vision and the basis of many applications such as video surveillance, human-computer interaction and video editing. [0003] The basis of video segmentation is still image segmentation. Most still image segmentation techniques are performed in a bottom-up manner. They achieve the purpose of segmentation by detecting boundaries or clustering pixels based on features such as color and texture. However, due to the huge semantic gap between the underlying features of the image and the object ...

AI Technical Summary

Problems solved by technology

[0006] Traditional video segmentation methods based on spectral clustering often only use motion information or simply weight motion information and static information for segmentation, and cannot analyze both static features (such as color and texture) and motion features of video images. Information of different reliability is well utilized, and accurate and dense segmentation results cannot be obtained.

Images