针对煤粒表面图像的表征问题,采集了7个密度级共70幅煤粒图像,研究了煤粒图像的两个颜色空间,分析了煤粒图像RGB和HSV的6个颜色特征的一、二:、三阶矩,并对其进行筛选。结果表明煤粒表面呈现深灰色并且色彩信息很少,在进行煤粒密度级判别时,可以选取亮度一阶、亮度二阶、亮度三阶、饱和度一阶和饱和度三阶这5个颜色特征,参数表征不同密度级煤粒的颜色信息,为实现机器视觉识别煤质指标提供了参考。
For the characterization of the coal particle surface image , a total of 70 coal parti- cle images of the 7 density levels were collected, the two coal image color space (RGB color space and HSV color space) were researched, the first , the second and the third moment of six colour features were analyzed and selected . The results showed that the coal particle surface presents dark gray and have very little color information. During the recognition of the coal particle densi- ty, 5 color features (the first moment of the value , the second moment of the value, the third moment of the value , the first moment of the saturation and the third moment of the saturation) can be selected to characterize the color information of the different density level coal particles surface image, which provide a reference for the machine vision to identifv coal aualitv index_