基于灰色关联分析的医学图像分割

Segmentation of medical image based on the grey relational analysis

医学图像分割处理为临床诊断、病理分析、治疗提供可靠的依据。为了分割出图像中感兴趣的区域,提出了一种基于灰色理论区域生长分割算法。该方法利用灰色关联理论对图像进行关联性分析,根据像素间全局和局部的关联度,指导区域生长的分割和停止。该方法的最大不同之处在利用灰色理论进行关联性分析,不需要大量数据的统计,算法简单,高效。实验证明,该方法能够有效地分割出目标区域,并且具有一定的抗噪能力,是一种较为有效的图像分割新方法。

Segmentation is an important aspect of medical image processing, its purpose is to extract the region of in terest from the original images to provide a reliable basis for clinical diagnosis. A novel segmentation technique is presen ted which is based on grey relational analysis of grey system the region of interest from original images can be extracted. theory. According to global and local inte^pixel correlation, The experimental results show that the algorithms can effectively extract the target area, and has strong anti-noise capability. This new method is simple, fast, and works well for some images.