Abstract: Nowadays, color image is main form of digital images, and its segmentation problem becomes a hot and difficult topic in image processing. Aimed to improve the accuracy and efficiency of color image segmentation, a method based on a color space is presented. The effect of the proposed color space is mainly to increase difference among the three tunnels, thus the color information of images can be fully used. In order to give segmentation results in a high efficiency, the proposed segmentation method takes vector image as the target, follows the background removing idea and applies split Bregman to solve the corresponding minimization problem. Experimental results demonstrate that the proposed method can segment all target parts in image effectively and integrality, which has a highly accuracy.

Key words: color image, color space, image segmentation, variational model, split Bregman method

摘要： 彩色图像是当今数字图像的主要表现形式之一，彩色图像分割是图像处理的热点与难点问题之一。针对彩色图像分割的精度与速度问题，提出了一种基于色彩空间变换的彩色图像分割方法。该色彩空间重在突出图像三分量间的差异，从而充分利用彩色信息。同时，分割方法以向量化图像为对象，从背景去除的思想出发，采用分裂Bregman快速求解算法，从而达到高效分割的目的。数值实验表明，该方法能有效分割出图像的所有目标区域，分割区域较为完整，结果具有较高的精度。

关键词: 彩色图像, 色彩空间, 图像分割, 变分模型, 分裂Bregman方法