Quick graph cutting method based on multiple deformation resolutions

Abstract

The invention discloses a quick graph cutting method based on multiple deformation resolutions. The method comprises steps: multiple different kinds of scale compression and length-width ratio adjustment are simultaneously carried out on an input graph; Graph Cuts cutting is parallelly completed in spaces of different deformation resolutions, a series of not entirely-determined temporary cutting results are acquired, the temporary cutting results are reversibly deformed to space of an original resolution, and a multi-deformation resolution weight graph is obtained in a weighted voting form; a priori mask graph Trimap is corrected according to the information of the multi-deformation resolution weight graph, a to-be-cut area is determined and narrowed, weighted Gaussian mixture models for foreground and background are trained, a reduced graph model is thus built and quickly solved, and a final cutting result is obtained. A foreground object with a complicated detail area can have good cutting effects, and the algorithm framework has flexible extendibility and can replace different bottom layer cutting algorithm to be applied to different cutting scenes.

Description

technical field [0001] The invention belongs to the technical field of image processing, and more specifically relates to a fast image segmentation method based on multi-morph resolution. Background technique [0002] Image segmentation is one of the basic tasks in the field of image processing. The results of segmentation can provide important description information about foreground targets for higher-level machine vision research, such as visual saliency analysis, scene analysis, and medical image analysis. Image segmentation methods can be divided into two categories: unsupervised and interactive. Among them, unsupervised image segmentation methods have strict requirements on input images or application scenarios, and the segmentation results are often unpredictable and cannot be intervened and corrected. Therefore, the segmentation method that uses low-cost user interaction (such as using a marker box or brush) to assist in the accurate segmentation of objects has a wi...

AI Technical Summary

Problems solved by technology

[0003] However, no matter how it is extended or improved, any segmentation method based on the GraphCuts framework inherently has the problem of "shrinking bias" (shrinking bias) at a certain image resolution, that is, due to the energy functional The smooth term of represents the edge properties in the image, so in the process of using the maximum flow / minimum cut algorithm to solve the graphical model, the solution set will always tend to contain shorter boundaries (shorter boundaries), specifically in the segmentation For some foreground objects with slender areas or sunken areas, the final segmentation result may paranoidly shrink and truncate these areas with too long borders directly, resulting in mis-segmentation of the corresponding areas

Images