知识与财富的链接
针对随机纹理图像在实际生产环境下的色差检测问题,在研究、分析国内外研究成果的基础上提出了基于CIELAB均匀颜色空间的色差检测算法。文中对算法进行了详细的描述。首先在HSV颜色空间下计算聚类数目,在CIELAB颜色空间下确定初始聚类中心,然后进行模糊C均值聚类,最后进行颜色特征的提取。以雨花石墙地砖为例,分别在光照、粉尘、振动三种典型干扰环境下进行了实验。实验结果表明本文提出的色差检测算法检测结果受粉尘,振动干扰影响较小,色差值小于1.5,认为无色差,而受光照干扰影响较大,色差值大于1.5,认为有色差。这一结果与人眼的视觉判断相一致。建议在实际应用中对光照条件、光源稳定性进行合理的控制。本算法对其它随机纹理产品的颜色检测也有实用价值。
In order to solve the problem that detecting color-difference of the image with random texture in the actual production environment, based on the research and analysis results both at home and abroad, a color-difference detection algorithm based on the CIELAB color space is proposed in this paper. The algorithm is described in this paper. First, the number of cluster is calculated in the HSV color space. Then the initial cluster center is determined in the CIELAB color space. Finally, the color-difference is detected with fuzzy C-Means clustering. Take the pebble tiles for an example, some experiments in the environment of light, dust and vibration are done in this paper. The experimental results show that color-difference detection algorithm is consistent to the human eye and has anti-interference capability. The test result of color-difference detection algorithm isn’t affected by the dust and vibration interference. The color-difference value is less than 1.5. It is affected by the light interference. The color-difference value is greater than 1.5. So the stability of light power should be reasonable control in the practical application. The algorithm can be used to detect surface color of other random texture.
