基于多重模糊属性的彩色图像分割

Color image segmentation based on multiple fuzzy properties.

提出了一种综合考虑多重模糊属性的彩色图像分割算法.该算法定义了像素的邻域模糊距离,并根据同一区域内部消除亮度信息后彩色部分相关性非常高的原则,利用区域生长方法产生准种子区域和模糊像素集.对于这些准种子区域按照面积大小进行区域提纯,从而得到确信度很高的种子区域;对于模糊像素,利用合理的目标函数确定出它们的区域属性.讨论了区域的紧性和相邻区域之间颜色相似度的权重关系,并用于区域合并.实验表明,算法的分割结果与人的主观视觉感知具有良好的一致性.

A new algorithm for color image segmentation based on muhiple fuzzy properties is presented. First, fuzzy distance in the neighborhood of a pixel is defined. And according to the principle that the color components whithout brightness are highly related in a homogeneous region, some quasi-seed regions and fuzzy plxel sets are produced by region growing. In terms of area, some seed regions can be accurately extracted from all the quasi-seed regions; and for the fuzzy pixels, which regions they belong to can be determined by the reasonable objective function. Finally, the weight between region compactness and color similarity among adjacent regions is discussed and used to merge regions. The experimental results show that the color image segmentation results of the proposed algorithm hold favorable consistency in terms of human perception.