Image restoration method and system based on angle perception block matching

Abstract

The invention discloses an image restoration method and system based on angle perception block matching, and the method comprises the steps: dividing a to-be-restored image into a target to-be-restored area and a source area, and carrying out the curved surface fitting; then, determining a to-be-filled target block with the highest priority; rotating a matching block most similar to the target block in the source region, and obtaining an optimal matching block by using angle perception; and finally, filling a to-be-repaired area in the target block, and updating the confidence value of the filled pixel point; and repeating the above steps until the to-be-repaired area is completely filled. According to the method, the optimal spatial position is sensed by using the angular rotation of the matching block, so that the image content including the color, the texture and the geometric shape of the defect area is recovered to the maximum extent, and the visual requirement of human beings can be well met. The method can be successfully applied to image restoration of various large-area damaged areas with complex textures and geometrical structure characteristics, and can also be effectively applied to the aspects of shielding object removal, target object moving and the like.

Description

technical field [0001] The invention relates to the field of computer image processing, in particular to an image restoration method and system based on angle-aware block matching. Background technique [0002] The statements in this section merely mention the background technology related to the present invention and do not necessarily constitute the prior art. [0003] In the process of realizing the present invention, the inventor finds that there are following technical problems in the prior art: [0004] Digital image restoration refers to a technology that uses the known pixel information in the image to automatically fill the damaged area, so that the repaired image can meet the consistency requirements of human vision. Image restoration is a research hotspot in image restoration, which is widely used in the fields of protection of ancient cultural relics, image editing (removing unwanted text or occluded objects, etc.), medicine and military affairs. [0005] Gener...

AI Technical Summary

Problems solved by technology

[0008] (1) Each time only a single matching block is selected from the source area to fill the target block, the angle perception characteristics of the block are not fully considered, the transitivity between blocks is ignored, and the global structure of the image cannot be fully utilized, making the repaired area There is an inconsistency with the surrounding area;

[0009] (2) This method does not consider the influence of the content of the area to be repaired, and only fills it with the pixel value of the corresponding position of the best matching block, which may introduce inconsistent content;

[0010] (3) When looking for the best matching block, the best matching block is only judged according to the minimum value of the distance of the pixel in the color Euclidean space, ignoring the gradient information of the block, which will easily lead to wrong matching. Errors will be continuously propagated, resulting in unsatisfactory final filling results

Images