摘要
在卫星遥感图像的分析中,提取图像的边缘和纹理特征,对于识别图像的类型,从而完成对目标物体的分析十分重要。根据欧空局提供的Envisat卫星的共享数据,通过C++程序设计转换成dat数据格式并在Matlab环境中显示。依据卫星遥感图像的特点,讨论了几种图像边缘检测和纹理特征提取的方法,如微分算子、模糊C均值聚类、灰度共生矩阵等,通过分析、比较、判断,应用图像实例验证说明了上述分析方法的有效性与特点。
To extract the photo edge and characteristic of texture from the satellite remote sensing image is important to distinguish the image type. According to the share data supplied by Envisat satellite of European Space Agency the writing formula is carried on by C++ in order to transform the data into dat type which is exhibited by Matlab. According to the characteristic of the satellite remote sensing image some methods about the extraction of the texture from image and the examination of imaged edge are discussed. For example, FCM, Gray-level and co-matrix are described by analysis, comparison and decision.
出处
《成都信息工程学院学报》
2006年第z1期36-42,共7页
Journal of Chengdu University of Information Technology
关键词
遥感图像
图像分割
边缘检测
纹理分析
模糊聚类
remote sensing image
image segmentation
examination of image edge
texture analysis
FCM
