Study Of Color Difference Classify For Ceramic Tile With Computer Vision

With the improvement of living standard and the development of the base construction, the consumption quantity and variety of ceramic floor tile becomes more and more large. And the requirement for it's quality also becomes more and more serious. Due to the complexity of the production, the ceramic products have the non-homogeneous problems in the color part. The color aberration on the surface of ceramic tile is an important index when assessing the quality. Excluding the color aberration factor, the homogeneity degree of color distribution of homochromatic tile is also an important factor which influences the tufting effect directly.The image measurement on the base of computer vision is a new technology which developed in the measurement field. It's a synthesis measurement system which combines the technology of optics, mechanical, electronic and computer. The technology is on the base of optics and combines the optical electronics technology, computer technology, laser technology and image-processing technology etc. Aim at the color-detection compatibility problem exists in the ceramic field, a new color-detection method basing on the computer vision color-aberration classify technology is proved in this paper. With the technology, the quality difference in personal factor can be cancelled in the process of manual detection. So this method has the characteristics of high automatic, excellent compatibility, low cost etc.We designed the whole color-detection system from the beginning of color characteristic. We analyzed and determined the color model and hardwarebasing on the foreign and home nation research of ceramic tile. In this paper, we analyzed the image preprocessing algorithm and removed the noise interferences by the median filtering method. Using the technology of image angular point detection, we allocated the image. And according to the minor color-aberration theory, we analyzed and calculated the aberration, and studied the color-classify on the base of fuzzy theory. Finally, the conclusion is given in the paper.