摘要
为提高医学图像分割的视觉效果,依据人类视觉感知的分层特性,提出了一种新的复合医学图像分割方法.该方法通过提取医学图像的底层特征,利用Fuzzy-ART神经网络作为像素的分类器,对医学图像进行连续两次分割.实验结果表明,该医学图像分割方法能有效地解决局部信息与整体分布边缘淡化等相关问题,达到良好的分割视觉效果.
To improve the visual effect of the medical image segmentation,an improved medical image segmentation algorithm was presented based on the visual perception delaminated characters.Based the character of medical image,the local region features were described firstly,and then the local texture feature was designed consistently with visual perception of different image region.The Fuzzy-ART neural network as pixels classifier was used to segment the medical image based on the layer feature functions.The experiment results showed that this method effectively solved the local information and distribution of image local features,and achieved a good segmentation and visual effect.
出处
《郑州大学学报（理学版）》
CAS
北大核心
2011年第1期57-61,共5页
Journal of Zhengzhou University:Natural Science Edition
基金
山东省自然科学基金资助项目
编号ZR2009CM085
山东省高等学校科技计划项目
编号J09LF23
山东省中医药科技计划项目
编号2009-027
