Research on the Identification of Hand-Painted Thangka and Printed Thangka Based on Computer Vision Processing Method

Thangka is a unique painting art form in Tibetan culture. As Thangka was listed as the first batch of national intangible cultural heritage, it has received more and more attention. At the same time, many printed thangkas appeared on the market, and some merchants mixed printed thangkas with hand-painted thangkas and sold them at high prices. How to distinguish between hand-painted thangkas and machine-printed thangkas, and to guide the inheritance and protection of hand-painted thangkas is an important issue. Based on the analysis of the characteristics of thangkas, combined with related computer vision processing methods, this paper conducts method research and model construction, uses the LBP algorithm to analyze the texture of the canvas, identifies the canvas of the thangka based on the training results, and uses restricted Boltzmann s light transmittance analysis of thangkas, and the use of clustering method to analyze the gold line of thangkas and other methods combined, proposed a feasible method of identifying hand-painted thangkas and machine-printed thangkas. This method should play a significant role in the protection and inheritance of the country. This specification is set for scientific papers published in “Computer Applications and Software”. The author is requested to read and implement one by one, if it does not meet the requirements, it will affect the publication of the article.