摘要
图像分割是图像分析和模式识别需要解决的首要问题和基本问题,也是图像处理的经典难题。分水岭变换则是一种适用于图像分割的强有力的形态学工具,然而其不足之处在于它的过分割结果。提出一种基于ISODATA聚类和标记分水岭的分割方案,该方案首先通过中值滤波来消除部分噪声;然后用ISODATA方法进行聚类,获得更明显的特征差异;接着采用Sobel算子进行梯度重建,得到具有边缘信息的简化图像;在此基础上再进行基于标记的分水岭变换。实验结果表明,该方法分割精度达到80%以上,能够较好地抑制过分割。
Image segmentation is a chief and basic issue in the field of image analysis as well as mode identification.Meanwhile,it is also the classical puzzle in image processing.And the watershed transform is a powerful morphological tool for image segmentation.But its shortcoming is to cause over-segmentation.Therefore,a method for remote sensing images based on ISODATA and labeling watershed algorithm has been presented.Firstly,a median filtering is applied to smooth the original image,so it can reduce part of noise. Secondly, use ISODATA to gain more obvious differences of characteristics. Then, the gradient image, which has the edge information, is obtained through the reconstruction of gradient by using Sobel operator. Finally, segmentation result is obtained by using an improved method of labeling watershed algorithm. The result shows that the method can reduce oversegmentation more efficiently,with a precision of more than 80 %.
出处
《计算机技术与发展》
2010年第1期39-42,共4页
Computer Technology and Development
基金
2008年江苏省创新训练计划立项项目(08CX0009)
2007年度南京信息工程大学校基金项目(20070066)
关键词
遥感图像
图像分割
分水岭变换
聚类
remote sensing image
image segmentation
watershed transform
clustering
