在彩色IC缺陷图像的检测与识别系统中 ,通常采用分色技术来降低图像识别的难度 .该文利用 3种方法 (HSV彩色模型、HLS彩色模型及明度信息 )对IC缺陷图像进行了分色处理 ,并对其结果做了主客观评价 ,结果表明HSV模型在突出缺陷及与原始图像的相似性方面优于其他两种方法 In the detection and recognition system of color IC defect images, the technique of color separation is usually adopted to reduce the difficulty of image recognition.This paper uses three methods (HSV color model, HLS color model and lightness information) to separate IC defect images Processing, and the results of the subjective and objective evaluation made, the results show that the HSV model in highlighting defects and the original image similarity is superior to the other two methods