Study on color extraction and processing of Dunhuang nine-color deer image based on K-Means cluster analysis

This research focuses on the color extraction and processing direction of nine-color deer images in Dunhuang. In this paper, K-Means clustering analysis algorithm is used to extract the color of the image and get the proportion of each color, so as to build a scientific, accurate and reasonable color feature model of Dunhuang nine-color deer image extraction. Firstly, image filtering is used for image pre-processing, so that the data generated by the image is more accurate. Secondly, the image segmentation method is used to segment the Dunhuang nine-color deer image, and the key areas are separated and classified, and then the color information of the image is deeply mined by K-Means algorithm, and finally the scientific, accurate and reasonable color feature model and the main color proportion color card of the Dunhuang nine-color deer image are obtained. This research is helpful to the application of computer redesign of Dunhuang art, and provides a new application path for image processing and color extraction research field.