周家香, 朱建军, 梅小明, 马慧云. 多维特征自适应MeanShift遥感图像分割方法[J]. 武汉大学学报 ( 信息科学版), 2012, 37(4): 419-422.
引用本文: 周家香, 朱建军, 梅小明, 马慧云. 多维特征自适应MeanShift遥感图像分割方法[J]. 武汉大学学报 ( 信息科学版), 2012, 37(4): 419-422.
ZHOU Jiaxiang, ZHU Jianjun, MEI Xiaoming, MA Huiyun. An Adaptive MeanShift Segmentation Method of Remote Sensing Images Based on Multi-Dimension Features[J]. Geomatics and Information Science of Wuhan University, 2012, 37(4): 419-422.
Citation: ZHOU Jiaxiang, ZHU Jianjun, MEI Xiaoming, MA Huiyun. An Adaptive MeanShift Segmentation Method of Remote Sensing Images Based on Multi-Dimension Features[J]. Geomatics and Information Science of Wuhan University, 2012, 37(4): 419-422.

多维特征自适应MeanShift遥感图像分割方法

An Adaptive MeanShift Segmentation Method of Remote Sensing Images Based on Multi-Dimension Features

  • 摘要: 针对MeanShift算法分割遥感图像的自动化程度和精度不高的问题,提出一种多特征自适应Mean-Shift遥感图像分割方法。3组实验结果表明,本方法相比EDISON软件能得到更好的分割效果,且能在一定程度上提高遥感影像分割的自动化。

     

    Abstract: Due to low segmentation efficiency and low accuracy of Mean-Shift algorithm,this paper puts forward to an adaptive Mean-Shift segmentation method of remote sensing images.Firstly,location features,multi-band spectrum principal components and texture features are extracted to form multi-dimension feature spaces.Then,based on classical Mean Shift clustering algorithm,initial clustering images are got by using less fixed space bandwidths and global optimal spectrum bandwidths that are estimated by plug-in rules.Mean space distance,mean spectrum distance and texture distance are calculated for each region in the initial clustering images,and used for space bandwidths,spectrum bandwidths and texture bandwidths of sequential clustering.Further,multi-dimension feature Mean-Shift Clustering was implemented by using calculated bandwidths.Lastly,the clustered regions are combined to get segmentation images.Three experiment results of remote sensing images show that the proposed method in this paper are better than EDISON software,and to some extent improve the efficiency of segmentation of remote sensing images.

     

/

返回文章
返回