论文部分内容阅读
在对试验区 IKONOS卫星高分辨率影像预处理的基础上 ,采用 HIS变换法对 1m分辨率全色影像和 4 m分辨率多光谱影像进行了融合处理 ;通过建筑物阴影提取了建筑物高程 ,采用正交小波变换方法提取了图像的纹理信息 ;采用了四种不同的特征图像组合方式 ,对试验区高、中、低层建筑物进行了分类。研究表明 ,由融合图像、高程信息和纹理信息参与的分类结果精度最高 ,对高层建筑物分类的精度可达 80 %。
Based on the preprocessing of IKONOS satellite high resolution image in the experimental area, the HIS transform method was used to fuse 1m resolution panchromatic image and 4m multiresolution image; the elevation of building was extracted from the shadow of building, Orthogonal wavelet transform is used to extract the texture information of the image. Four different feature image combinations are used to classify the high, middle and low buildings in the test area. The research shows that the accuracy of classification results with the fusion images, elevation information and texture information is the highest, and the classification accuracy of high-rise buildings can reach 80%.