| 基于空间邻域信息的加权FCM图像分割方法 |
| |
| 引用本文： | 彭代强,孙玉秋.基于空间邻域信息的加权FCM图像分割方法[J].长江大学学报,2009(2):69-72. |
| |
| 作者姓名： | 彭代强 孙玉秋 |
| |
| 作者单位： | [1]南京电子技术研究所,江苏南京210013 [2]长江大学信息与数学学院,湖北荆州434023 |
| |
| 摘 要： | 提出了一种新的加权模糊c均值图像分割方法，该方法根据邻域象素间的灰度变化，选取与目标象素灰度特性相似的邻近象素组成高度相似象素区域，建立邻域像素隶属度对目标象素隶属度的空间影响函数，通过空间影响函数来重新估计目标象素的模糊隶属度，以反映不同象素点对目标象素点分类的影响。由于该算法利用象素空间邻域关系对模糊隶属度进行了调整，使得目标分类时考虑了图像象素的邻域信息。试验结果表明，提出的方法抗噪性能强、收敛速度快，能够有效的分割图像。
|
| 关 键 词： | 模糊C均值 图像分割 邻域信息 |
A Novel Weighted Fuzzy C-means Algorithm for Image Segmentation Based on Spatial Information |
| |
| PENG Dai-qiangSUN Yu-qiu.A Novel Weighted Fuzzy C-means Algorithm for Image Segmentation Based on Spatial Information[J].Journal of Yangtze University,2009(2):69-72. |
| |
| Authors: | PENG Dai-qiangSUN Yu-qiu |
| |
| Institution: | PENG Dai-qiang(Nanjing Research Institute of Electronics Technology,Nanjing 210013)SUN Yu-qiu(Yangtze University,Jingzhou 434023) |
| |
| Abstract: | A novel weighted fuzzy c-means algorithm for image segmentation was proposed.The proposed algorithm uses the variation of the pixel intensities of a region to choose strong resembling area.Then,the spatial relationships of the membership function can be constructed to estimate the membership of a pixel according the memberships of its neighborhood pixels.The spatial relationship implies the membership relationship between the pixel and its neighborhood pixels and can adjust the membership value based on the... |
| |
| Keywords: | fuzzy c-mean image segmentation spatial information |
| 本文献已被 CNKI 维普 等数据库收录！ |
