Key points are not available for this paper at this time.
Mural painting refers to art on the wall, which has high aesthetic value. However, due to its vulnerability to damage, digital image restoration and stitching techniques are now used to permanently preserve mural images. The purpose of our study is to select a more suitable specific calculation method by studying machine learning (ML) methods and conduct in-depth research on mural digital image restoration technology and stitching evaluation, so that it can better serve the restoration and splicing of the current mural digital images. Based on the experiments, it can be seen that the respondents in hall A had a high degree of recognition for digital image restoration and stitching technology. There were 176 respondents, and only 29 people believed that digital image restoration was not important; respondents in hall C did not agree with the digital image restoration technology as much as hall A, only 132 people thought it was important, and 64 people thought it was not important. It can be seen that the model establishment of mural digital image restoration technology and splicing evaluation is deeply affected by the knowledge level of the evaluator. The experimental results of our study showed that the research process of mural digital image restoration technology and stitching evaluation based on ML method was more effective than other methods of analyzing experimental data.
Wang et al. (Thu,) studied this question.