Image collaborative segmentation method based on minimum fuzzy divergence

Abstract

The invention discloses an image collaborative segmentation method based on minimum fuzzy divergence, and belongs to the technical field of image processing and computer vision. According to the method, the segmentation effect is judged through an Intersection over Union (IOU) value, and the method comprises the following steps: 1, acquiring an image segmentation data set, and performing conversion from an RGB space to an LAB space; 2, constructing a fuzzy divergence formula by using a Gamma-type membership function, constructing a new energy function, and performing curve evolution accordingto a minimum fuzzy divergence criterion to achieve a good segmentation effect. The target edge is better processed by using the fuzzy set theory. The color information of one image is introduced intothe energy function of the other image, so that the robustness of initial curve replacement can be enhanced. An optimal segmentation effect is achieved by solving a local minimum value of an energy function by utilizing a region-based active contour model. The established model can reduce the complexity of calculation time, and can be applied to early-stage work of an integrated imaging three-dimensional display system.

Description

technical field [0001] The invention belongs to the technical field of image processing and computer vision, and in particular relates to an image cooperative segmentation method based on minimum fuzzy divergence. Background technique [0002] With the increasing scale of image data in recent years, the demand for image segmentation in many practical applications is no longer satisfied with small-scale single image segmentation. Extracting common objects from a group of images has become an active research topic, namely, image segmentation. Co-segmentation problem. [0003] The idea of ​​collaborative segmentation is to consider an additional foreground similarity constraint on the segmentation method based on a single image, so as to realize the segmentation of common objects. In response to this problem, a variety of collaborative segmentation algorithms have been proposed in the past few years, and have been successfully applied to practical problems, such as the collabo...

AI Technical Summary

Problems solved by technology

In response to this problem, a variety of collaborative segmentation algorithms have been proposed in the past few years, and have been successfully applied to practical problems, such as the collaborative method based on Markov Random Field (MRF), according to the optimal criterion in estimation theory To determine the objective function of the segmentation problem and transform the segmentation problem into an optimization problem, the difficulty lies in the selection of image feature selection and the selection of the foreground similarity measurement method; the collaborative method based on thermal diffusion is to model the segmentation problem as the maximum temperature of thermal diffusion optimization problem, but the effect is not ideal for the foreground (such as the human body) composed of multiple unique regions; the collaborative method based on random walk, the optimization calculation is simpler, but the user needs to mark the foreground region and Background area; in the collaborative method based on active contours, different feature selection methods and consistency measurement methods will have different effects on the model

[0004] Due to the defects of the above algorithm, it is difficult to achieve ideal results in practical applications, so it is necessary to improve

Images