Image segmentation method based on hierarchical higher order conditional random field

Abstract

An image segmentation method based on a hierarchical higher order conditional random field model is provided, comprising: firstly, extracting a multi-class texture feature for a target image, and constructing a one-variable potential function and a pairing potential function of a pixel level; then, acquiring super pixel fragments of different granularities by using a unsupervised segmentation algorithm; designing a one-variable potential function and a pairing potential function of a super pixel level corresponding to each granularity layer; constructing a hierarchical higher order conditional random field model; learning a parameter of the hierarchical higher order conditional random field model in a supervision manner by using a manual marked sample; and finally, acquiring a final segmentation marking result for a to-be-tested image by means of model reasoning. The hierarchical higher order conditional random field model used in the present invention fuses multi-feature texture information and multi-layer super pixel segmentation information of an image, and can effectively improve boundary segmentation accuracy of a multi-target object in the image.

Description

technical field [0001] The invention relates to the technical field of image segmentation, in particular to an image multi-object segmentation method based on hierarchical high-order conditional random fields. Background technique [0002] Image segmentation is a key problem in the field of computer vision. The quality of image segmentation has an important impact on subsequent applications such as image content analysis and pattern recognition. The current image segmentation algorithms mainly include the following categories: 1) Image segmentation based on threshold. This type of method is suitable for target images where the target and background have different gray scale ranges. 2) Region-based image segmentation. Its idea is based on the pixels with similar characteristics, through the image segmentation technology of region growing and region merging. 3) Segmentation based on deformation model. Such methods need to give the initial closed segmentation curve of the ...

AI Technical Summary

Problems solved by technology

Although this method has the characteristics of fast operation, the conditional random field model based on superpixels [4] usually enforces the consistency of the classification labels of all pixels in a superpixel, resulting in the segmentation results relying heavily on unsupervised segmentation algorithms pros and cons

For example, if the granularity of superpixel segmentation is not appropriate, a superpixel may contain different targets at the same time, and often the final segmentation accuracy is not as good as the pixel-based conditional random field model.

Images