基于主色的最大类间方差的彩色图像分割
首发时间:2009-12-31
摘要:提出了一种基于主色的最大类间方差的彩色图像分割方法。在量化HSV颜色模型下,为了减小频率出现低的量化颜色值对分割造成的影响,选取图像中出现频率高的量化颜色值作为主颜色,将所有像素归并到和它相似度最大的主颜色,得到图像的主色图;对主色图采用最大类间方差法进行分割。实验结果表明该方法能够较好地将目标从背景中分割出来。
For information in English, please click here
Otsu Algorithm Based on Dominant Color for Color Image Segmentation
Abstract:A method of Otsu algorithm based on dominant color for color image is presented in this paper. In order to reduce the affect by the low frequency colors to color segmentation, the high frequency colors are chosen as dominant colors under the quantized HSV color model. All the pixels are classified to the nearest dominant colors. Dominant color image is obtained. Otsu algorithm is used to segment color image based on dominant color image. Experimental results show the method can extract the object from the background.
Keywords: color segmentation dominant color Otsu algorithm
基金:
论文图表:
引用
No.3837150850112622****
同行评议
共计0人参与
勘误表
基于主色的最大类间方差的彩色图像分割
评论
全部评论0/1000