基于超像素和最近邻图合并的彩色图像分割

Color Image Segmentation Based on Superpixel and Nearest Neighbor Graph Merging

针对传统的分水岭算法在分割图像时存在过分割的问题,提出了一种基于多尺度形态学梯度重建与最近邻图合并准则的分水岭图像分割方法.该方法首先使用基于标记符控制的多尺度形态学梯度重建进行图像预处理,消除冗余的区域极值和噪声;然后通过构建最近邻图合并准则进一步对分水岭变换产生的超像素区域进行合并,提高了对目标特征的描述能力,使得算法在分割前景目标的同时也能获得背景目标的特征信息.实验结果表明,该方法能够较好地将相似的区域进行合并,与传统分水岭分割方法相比,可以有效弱化过分割问题,大幅提升目标的分割精度.

Aiming at the over-segmentation problem in the existing watershed algorithms,this paper proposed a watershed image segmentation method based on multi-scale morphological gradient reconstruction and the nearest neighbor graph merging criterion.Firstly,multi-scale morphological gradient reconstruction based on marker control was used for image pre-processing to eliminate redundant regional extremum and noise.Secondly,the superpixel regions generated by watershed transformation were merged by constructing the nearest neighbor graph merging criterion,so it improved the description ability of target features and enabled the algorithm to obtain the feature information of background targets while dividing foreground targets.The experimental results show that the proposed method can combine similar regions better,and effectively solve the over-segmentation problem.Compared with the traditional watershed segmentation methods,this proposed method effectively weakens the over-segmentation problem and greatly improves the segmentation accuracy.