论文部分内容阅读
卫星图像线状地物影像宽度 ,由于受图像几何分辨率的影响而产生宽度夸张现象 ,其夸张范围在 3a (a为像元宽度 )以内 ,它对我省少林地区的农田林网资源数据影响很大。为了消除影像夸张误差 ,通过l2 5组实例数据 ,拟合回归方程 ,对其进行修正。这种办法不仅解决了目视判读林带宽度夸张问题 ,也为计算机自动分类奠定基础。
The width of linear imagery of satellite images, due to the influence of image geometric resolution, has an exaggerated width, which is within 3a (a pixel width), which affects the data of farmland shelterbelt resources in Shaolin area in our province Great. In order to eliminate the image exaggerated error, through l2 5 sets of example data, fitting the regression equation, to be amended. This approach not only solved the problem of visual interpretation of the exaggerated forest width, but also laid the foundation for the automatic classification of computers.