Method for restoring Thangka image by combining shapes and neighborhood classification of damaged piece

Abstract

The invention discloses a method for restoring a Thangka image by combining the shapes and neighborhood classification of damaged pieces. A proper algorithm is automatically selected to restore a damaged Thangka image according to the shapes and neighborhood information of the damaged pieces, the characteristics of the conventional restoration algorithms and the like. The method mainly comprises the following steps of: segmenting the image to be restored by using a watershed segmentation method, and classifying the damaged pieces in a damaged area into linear damaged pieces and blocky damagedpieces to realize the shape classification of the damaged pieces; extracting the second-order statistical information of neighborhood pieces by adopting a gray level co-occurrence matrix, extracting various characteristics of reaction textures by using the gray level co-occurrence matrix, performing Gaussian normalization on the characteristics, and classifying the neighborhood pieces into texture blocks and non-texture blocks by adopting a conventional K-nearest neighbor method to realize the neighborhood classification of the damaged pieces; and formulating selection rules for the algorithms by combining the characteristics of the restoration algorithms, the shapes of the damaged pieces and the neighborhood classes of the damaged pieces, and automatically restoring the damaged area. Themethod is applied to the restoration of various linear and blocky damaged pieces and various damaged neighborhoods of Thangka digital images, and has high restoration speed and efficiency.

Description

Technical field [0001] A Thangka image restoration method that combines the shape of the damaged block and the classification of the neighborhood of the damaged block belongs to the field of digital image restoration. Background technique [0002] The so-called thangka image restoration method that combines the shape of the damaged block and the neighborhood classification is that the computer uses a digitizer to detect the missing information on the image, and according to reasonable rules, automatically use the information of the undamaged area to damage the original image. A technique to partially fill in to achieve a good visual effect. Thangka image restoration is a practical application problem, which has important practical significance and theoretical value in cultural heritage protection. [0003] Digital image restoration technology is currently a research hotspot in the field of computer vision, and an important application is the virtual restoration of cultural heritag...

AI Technical Summary

Problems solved by technology

Image repair There is no unified method to solve all broken image repair problems

Moreover, the picture and damage of Thangka images are very complicated, and there is no special repair technology and practical repair method for Thangka images at present.

Images