模糊各向异性扩散人脑MR图像自动分割

目前对人脑磁共振图像的处理由于噪声等因素的影响,往往准确度不高。论文提出一种新的基于模糊各向异性扩散的人脑磁共振图像自动分割方法,该方法首先结合非线性扩散滤波构造混合信息场,在平滑目标内部同时处理强噪声区域,降低强噪声的影响且保留边界信息。然后利用FCM方法、形态学方法和标记法对混合场进行分析与处理,最终得到分割结果。对脑部MR图像分割的实验表明该方法的分割准确性较高,错误率约为1.3%。

Auto Segmentation of Brain MR Image Based on Fuzzy Anisotropic Diffusion

Current Brain magnetic resonance image segmentation and volume measurement methods are complex and not sufficiently accurate for certain applicatlons.This paper introduces a new fuzzy anisotropic diffusion method to diffuse the images and construct new mixed information.The fuzzy anisotropic diffusion method can not only diffuse the areas inside of the object but also diffuse the strong noise and reserve the edge information.After getting the mixed information the FCM method,a new tag method and morphology theory are introduced to analyze the mixed information and get the result.The experiments to segment the brain MR images show that the method of this paper can get better results in an accuracy way.