The Research of image edge segmentation optimization problems. Compared to the traditional edge image segmentation method, when the object structure is complex, causing pixels overlap or adhesion, caused image edge the partitioning of the error rate is higher. This paper proposes a method based on MCMC grayscale image segmentation method. Through the establishment of four kinds of image grey model and traversal of Markov space, preventing the traditional method brings segmentation wrong question because of edge points adhesion not clear, finally complete the integrity of the efficient grayscale image segmentation. Experiment shows that the application of this method can not only realizes image segmentation, but also can complete with high efficiency, the segmentation satisfactory results have been obtained.%研究图像边缘分割优化问题,由于图像边缘信息复杂,用传统边缘分割方法,当图像中的物体结构复杂的时候,会造成图像像素重叠、粘连等形成干扰的因素,导致图像边缘分割时错误率较高问题.为解决上述问题,提出一种基于MCMC的灰度图像分割方法.通过建立四种图像灰度模型和遍历的Markov Chains求解空间,能够避免传统方法由于边缘点粘连不清,而带来的分割错误问题,最终完成灰度图像的完整高效分割.实验证明,改进方法不仅能实现图像完整分割,而且能具有较高的分割效率,取得了令人满意的效果.
展开▼
