【作者】 刘苗苗； 黎宁； 吴园；

【Author】 LIU Miao-miao LI Ning WU Yuan The college of information science and technology Nanjing University of Aeronautics and Astronautics Nanjing 210016.

【机构】 南京航空航天大学信息科学与技术学院；

【摘要】 本文针对脑部 MR 图像具有弱边缘、对比度低等特点,对 Chunming Li 在[3]中提出的水平集方法(li 方法)进行了改进,并且结合 Kai Zhang 等人在[4]中提出的快速均值漂移算法(Fast Mean Shift)提出了一种新的 MR 图像分割算法。文中首先利用快速均值漂移算法对图像进行过度分割,然后再利用改进的水平集方法进行分割, 最后由轮廓提取算法将感兴趣区域提取出来。实验结果表明,该算法能够快速的得到准确的分割结果。更多还原

【Abstract】 For weak edges and low contrast in medical images,segmentation based on level set method proposed by LI in [3] is improved in the paper.Combined with fast mean shift algorithm presented by Zhang [4],a new segmentation method for MR images based on level set is then proposed.First,fast mean shift algorithm is used to over-segment the image.Second,improved LI-method is used to automatically segment the image.Finally,tissues of brain are then segmented from MR images by extracting object contour.The experiment results show that the processing of the proposed algorithm is fast and segmentation is accurate.更多还原