Image Extraction of Mural Line Drawing Based on Color Image Segmentation Algorithm

Abstract

Murals, also known as art on walls, are treasures of ancient Chinese culture and one of the earliest forms of painting in human history. The color image segmentation algorithm is the basis of image recognition, and the quality of the segmentation results directly affects the accuracy of the recognition. Color images are difficult to distribute well due to their complex base and abundance of large numbers of colors. The purpose of this paper is to study the derivation of linear images of murals based on color image algorithms. This paper uses optical predictive modeling and GrabCut to improve color image algorithms and plan color image algorithms based on line drawing. In view of the rich local features of mural images and many cracks, the idea of replacing the whole with parts is introduced, and the semi-automatic crack extraction algorithm is used. The transformation matrix of the partial image is obtained by image matching, and the transformation matrix is used as the transformation matrix of the complete image, and finally, the extraction of the mural image is realized. Experiments have shown that the framework optimized in this paper can save 80% of the time when extracting 100 mural line drawings, which meets the needs of the research. Therefore, the color image segmentation algorithm is one of the important technologies for the extraction of mural line drawing images in the future.