Ceramic Decoration Extraction Method Based on Computer Vision and Image Processing

China is one of the first countries to invent pottery in the world. Like India, West Asia, Japan and the central Balkans, our ancestors invented pottery in the Neolithic period of primitive society. As a product of computer technology, the development and application of graphics processing technology has been extended to all sectors of society, especially in the field of ceramic decorative pattern design. The speed and accuracy of traditional intelligent manufacturing detection can not meet the current detection requirements. Machine vision detection technology captures the image through the vision system, uses the Prewitt directional derivative approximation operator to detect the edge of the copied image, and completes the template matching and matching value weighting by transforming the image detection space into the parameter space. After non-maximum suppression, all directions of corner response function are processed, and the right corner output and its direction are obtained. After expanding the corner area, the detection area is defined. Zernike moment discretization is used to process the real part and imaginary part of the image, and the image features are extracted. Combining the coordinates of each corner and sub-pixel coordinates of the image, the fake area and the real area are determined, so as to realize the image tampering detection. In this paper, we focus on the body patterns of ceramic artworks, aiming at high-quality reproduction and design re-creation, deeply study the collection and image correction technology of three-dimensional patterns of ceramic artworks, and deeply analyze the algorithm of feature detection and extraction.