Full-connection bootstrap digital image filtering method and system

Abstract

The invention discloses a full-connection bootstrap digital image filtering method and system. The method at least comprises the steps of extracting the minimum spanning tree of a bootstrap image; determining the spatial similarity weight on the minimum spanning tree; according to the spatial similarity weight, calculating a linear representing coefficient on an image domain; based on the linear representing coefficient and the spatial similarity weight, conducting the weighted average treatment on pixels one by one. According to the method, based on the four-adjacency relation among the pixels of a bootstrap image, the formed bootstrap image is specifically represented in such a manner that the pixels of the bootstrap image are regarded as nodes. Based on the four-adjacency relation among the pixels, a connection relationship among edges between the nodes can be established. After that, the spatial similarity weight between two adjacent pixels is measured and the weight value is assigned to an edge formed through connecting the corresponding nodes of the above two pixels. Therefore, the graph representation of the bootstrap image is formed. Based on the method and the system, the technical effect of improving the edge keep smoothing effect is at least partially realized.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of computer graphics processing, and in particular, to a fully connected and guided digital image filtering method and system. Background technique [0002] In computer graphics applications, edge-preserving filtering is an essential component because it removes minor details (smoothing) while maintaining the main structure of the image (edge-preserving). A common way to deal with this problem is to change it to an explicit support region Ω around pixel i i weighted average within Among them, W ij Indicates the weighted kernel coefficient. The coefficient is calculated by the kernel function and can be understood as q i Calculate p in the weighted average i the weight of. q i and p i Denote the color values ​​of the input image and output image at point i, respectively. [0003] Existing edge-preserving image filtering methods usually explicitly select the support domain of th...

AI Technical Summary

Problems solved by technology

However, these algorithms still cannot select all relevant points

This defect limits the ability of these algorithms to represent narrow and long structures and complex geometric structures, and thus makes it impossible to generate ideal edge-preserving smoothing filtering results in practical applications.

However, the explicit support domain cannot be easily removed for the following reasons: 1. Some existing filtering methods do not consider spatial similarity, and the useless connections between long-distance pixels can only be removed by local support domains; 2. The computational complexity of some filtering methods depends on the size of the explicit support domain, so a small window must be selected to reduce the computational complexity

Images