Application Of Digital Image Restoration Technology In Restoration Of Barley Rock Paintings

Rock paintings are the general name for the paintings on the rocks,cave walls and stone cliffs as well as reliefs.China has a wealth of rock painting resources.However,with the passage of time,rock paintings have been affected by a variety of natural and human factors,many of which have suffered varying degrees of damage.Some rock paintings are covered with cracks,and even some rock paintings have a large surface surface shedding,which makes the rock paintings more seriously damaged.Therefore,in order to protect these precious cultural heritage,we must repair and protect the rock paintings.However,the manual repair of rock paintings is too limited and does not restore the original appearance very well.Now,with the rapid development of digital image processing technology,it is very suitable to use it to protect and repair rock art.This paper provides a set of repair methods for damaged rock paintings of barley land,and through a large number of experiments and comparisons,the proposed repair method can achieve good results in repairing rock paintings.The main work of this paper is as follows:(1)Pretreatment of barley rock paintings.Firstly,the basic methods of image denoising are studied.Finally,based on the median filtering denoising,and the method of extremum median filtering denoising is compared,an improved method of extremum median filtering denoising is proposed.According to the experimental results,it shows that the improved denoising method has better results than the traditional median filtering denoising and extremum median filtering denoising.The experiment shows that the improved denoising method is beneficial to improve the image restoration quality,so that the effective information of the image can be more completely preserved,which lays a foundation for the subsequent restoration work.(2)Extraction of damaged areas of rock paintings.Before the restoration of rock paintings,we first need to extract the damaged areas.In this paper,different extraction methods are selected for structural images and texture images.The SR model algorithm extraction in the saliency detection is taken for the structural image.A method of combining automatic extraction and manual marking for complex textured images.(3)The improved TV model algorithm is used to repair digital images of structural rock paintings.The traditional TV model algorithm has a certain degree of repair effect on the structural image,but there are inevitable defects,such as poor robustness and easy to generate false boundaries.In order to avoid the influence of these defects on the repair quality of rock paintings,this paper optimizes the selection and correlation coefficient of the field to be repaired,and combines the method of detecting the damage area with significant detection to propose a repairing structural image.Set of improved TV model algorithm flow.The experimental comparison shows that the improved algorithm is better than the traditional algorithm.(4)Digital images of textured rock paintings were repaired using the improved Criminisi algorithm.The Criminisi algorithm has excellent repair effects on large-scale damaged images,but its priority cannot effectively solve the best repair order of the edges,and the repair time is longer.In this paper,the improved Criminisi algorithm is used to optimize the priority calculation rules and confidence update.Combined with the extraction method of automatic crack extraction and artificial marking,a set of restoration methods suitable for textured rock painting digital images is formed.Experiments show that the method proposed in this paper can make great progress in quality and time.