基于时空混沌的图像分割算法

Image segmentation algorithm based on spatio-temporal chaos

为了实现图像的自适应多级分割,利用混沌动力学系统对初值的敏感性和时空混沌中全局耦合映像混沌同步的特点,提出了一种新的基于时空混沌的图像分割算法。新算法对每个灰度等级构建全局耦合映像,建立灰度等级间关系矩阵和灰度等级更新模型,通过全局耦合映像混沌同步过程完成灰度等级的自适应聚类。实验结果表明:算法对中间类灰度等级聚类具有很好的鲁棒性,分割过程不需要人工干预,并且具有较强的灵活性,优于阈值分割和边缘检测的图像分割方法。

In order to realize adaptive multithreshold image segmentation, based on sensitive dependence on initial conditions of chaotic dynamic system and chaotic synchronization feature of globally coupled map in spatio-temporal chaos, a new image segmentation algorithm based on spatio-temporal chaos was put forward. The algorithm constructed the globally coupled map for each gray level, the correlation matrix among gray levels and the updated model of each gray level were set up and the adaptive clustering of gray level through chaotic synchronization process of globally coupled map was accomplished. The experimental results show that the algorithm is robust enough to cluster ambiguous elements, needn＇t to be intervened in segmenting process and its flexibleness is better than the threshold segmenting method and the edge detection method.