基于HSV综合显著性的彩色图像分割方法  

Color image segmentation based on compositive HSV saliency

作　　者：张辰[1] 杨文柱[1] 刘召海[1] 

机构地区：[1]河北大学数学与计算机学院,河北保定071002

出　　处：《计算机工程与设计》2013年第11期3944-3947,共4页Computer Engineering and Design

基　　金：国际合作专项基金项目(2013DFA11320);国家科技支撑计划基金项目(2013BAK07B04);河北大学人才基金项目(2010-207);河北省教育厅基金项目(Q2012063);河北省科技支撑计划基金项目(12210133)

摘　　要：为解决彩色图像中目标难以准确分割的问题,提出了基于HSV综合显著性的彩色图像分割方法。该方法将原始RGB图像转换为HSV图像,从中分离出H通道、S通道和V通道,将其中的H通道和S通道作为原始的颜色信息来计算图像的颜色信息显著性,把其中的V通道作为原始的亮度信息并由此计算亮度信息显著性;通过加权的颜色信息显著性和亮度信息显著性得到综合显著图,并通过阈值分割的方法得到最终的目标图像。将该方法应用于棉花异性纤维图像的分割,分割结果表明,该方法能够准确地分割出彩色棉花异性纤维图像中的异性纤维目标。To solve the problem that targets in a color image is difficult to separate from the background accurately, a new approach for color image segmentation based on comprehensive HSV saliency is proposed. Firstly, the color image is converted from RGB color space to HSV color space. The converted color image in HSV color space is divided into H, S and V channels. The hue channel H and saturation channel S are treated as the color information of the image, and the value channel V as the bright information of the image. Secondly, color information saliency is calculated from H and S, and bright information saliency is obtained from V. Thirdly, comprehensive saliency map is obtained by weighted color information saliency and bright information saliency, where the weights are determined by the quantity of color information and bright information. At last, the ultimate targets are separated out from the comprehensive saliency map using the Otsu＇s method. The proposed method is applied for the foreign fiber color image segmentation, and the results indicate that it can segment out the foreign fiber objects from the color image accurately.

关 键 词：HSV空间 颜色信息显著性 亮度信息显著性 综合显著图 彩色图像分割 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]