Research On The Color Restoration Technology Based On The Ancient Damaged Murals

Ancient artifacts,especially murals,as a representative of China’s precious historical and cultural heritage,require us to take effective remedial measures to protect these fine artistic material.The application of digital image technology to the restoration of cultural relics,cultural relics in the protection and appreciation will have a very special significance.The paper,we first analyze the characteristics of the disappearance of image color in the ruins of ancient mural artifacts,and solve the problem of texture extension and sample block errors when using Criminisi algorithm to repair damaged image of frescoes with complex color information and geometric structure information Matching the problem,so that the efficiency of image processing and repair quality has not been further improved,this subject gives some improvements and optimization:（1）First of all,the paper prepares the obtained color mural images,and realizes the automatic recognition and marking of the images by different system of defect and color fading,and removes the color murals by denoising and texture background elimination In the unnecessary background information for the follow-up work to lay the foundation.（2）The data item and the confidence term are then added to the characteristics determined by the size of the priority value in the repair order of the boundary point in the Criminisi algorithm.There is a large fluctuation problem in the process of algorithm processing,and the algorithm is optimized by usinge^C（p）instead ofC（p）and introducing the golden weight ratio.And the optimal matching block distance is introduced at the confidence update to perform the parameter selection for the case where the confidence level is rapidly reduced and tends to zero when performing large-scale complex color and texture image restoration,and a new confidence degree is given the improved equation,the improved priority function,effectively improves the limitation when it is repaired.（3）The paper,image color segmentation and color decay curve are used to transform and repair the image color.Based on the combination of image texture and texture,the color of the image is analyzed and compared.Color space in the image segmentation role,and studied in a certain time after the color of the specific situation of fading,but also initially mastered the law of visual color distribution,and finally the use of several colors of the HVC component decay curve to fade recovery,So that the murals to restore the original color state.This paper,the improved image color restoration method is given by using the image recognition technology and the segmentation and matching method in view of the limitation of the traditional image restoration method when the mural with complex color and structure information is repaired.Compared with the traditional color restoration results,this paper adopts the method of image sample texture synthesis and segmentation matching to solve the problem of shortcomings and color unevenness in the traditional method.Mural repair time,but also improve the quality of the fresco,to achieve a more realistic visual effects and connectivity,the peak signal to noise ratio compared to the original algorithm also increased by 1³dB.