不同时相遥感影像的土地利用分类精度分析

土地利用调查对遥感影像的获取有很强的时间性、季节性要求。本次研究选取研究区不同时相的SPOT5遥感影像进行计算机自动分类试验,结合野外实地调绘工作,统计分类精度,拟合分类误差与时相间的回归关系,推导相关回归函数,对遥感影像开展综合分析与评价,提高遥感数据在土地利用调查中的利用率。

Analysis on classified precision of land use using remote sensing images in different temporal

There is an exigent request on remote sensing images in timeliness and season for land use.Three packs of SPOT5 remote sensing images were chosen in different time in this examination of automatic classification by computer.With the annotation on the spot,we counted the relative square error during classification,fit the regression relationship between classification error and temporal as well,and then gained the logarithm-related regression function.Combining the results with phonological characters in the research area,we synthetically analyzed the remote sensing images and present some evaluation of the images which can be used for the investigation of land use.With the result that we can prolong the gaining time of images,thus increase the utilization rate of remote sensing data in the investigation of land use.