摘要
为获得适用于高山峡谷地区遥感图像地形校正的方法,以滇西高山峡谷地区为研究区,在6景Landsat-5TM图像和SRTM-1数据的基础上,采用C校正、Teillet-回归、SCS+C校正、Minnaert校正、b校正和VECA校正进行对比实验。从目视效果看,C校正、Teillet-回归、b-II校正和VECA校正结果的亮度与校正前基本一致;在地形阴影严重地区,SCS+C校正出现过校正,b-I校正出现校正不足,Minnaert校正结果较暗且像元值为空,C校正和VECA校正出现彩色合成异常。从标准差看,Teillet-回归、b-I校正和b-II校正结果的标准差小于校正前;Minnaert校正结果的标准差大于校正前;C校正和VECA校正在TM1-3和高太阳高度角TM4上校正结果的标准差小于校正前,在TM5、TM7和低太阳高度角TM4上校正结果的标准差大于校正前;SCS+C校正在TM1-3、TM4和高太阳高度角TM7上校正结果的标准差小于校正前,在TM5和低太阳高度角TM7上校正结果的标准差大于校正前。从分散指数看,b-II校正在31个波段上的分散指数最小,占波段总数的86.11%。结果表明:C校正、SCS+C校正、Minnaert校正、b-I校正和VECA校正不宜用于高山峡谷地区遥感图像的地形校正;Teillet-回归和bII校正可用于高山峡谷地区遥感图像的地形校正,相比之下,b-II校正的效果优于Teillet-回归。
In order to obtain suitable method to correct topographic effect in remote sensing image at high mountain and gorge regions, comparison tests were conducted by using C correction, Teillet-regression, SCS + C correction, Minnaert correction, b correction and VECA correction, based on 6 scenes Landsat - 5 TM images and SRTM - 1 data from high mountain and gorge regions in western Yunnan. From the perspective of visual effect, the brightness rection caused under-correction; the Minnaert correction created dark pixels with the null value; C correction and VECA correction caused color composite anomaly. From the perspective of standard deviation, the standard deviations of Teillet-regression, b - I correction and b - II correction images were less than that of original images. The standard deviations of Minnaert correction images were greater than that of original images. After corrected by C correction and VECA correction, the standard deviations of TM1 - 3 and the high sun elevation angle TM4 were less than that of original images, the standard deviations of TM5, TM7 and the low sun elevation angle TM4 were greater than that of original images. After corrected by SCS + C correction, the standard deviations of TM1 - 3, TM4 and the high sun elevation angle TM7 were less than original images, and the standard deviations of TM5 and low sun elevation angle TM7 were greater than original images. From the perspective of dispersion indices, the dispersion indices of b - II correction done images were the smallest in 31 bands, results show that C correction, SCS + C correction, Minnaert correction, accounting for 86.11% of total band. The b- I correction and VECA correction are not suitable for correction topographic effect in remote sensing image at high mountain and gorge regions. Teillet-re- gression and b -II correction can be used to correct topographic effect in remote sensing image of high mountain and gorge regions. In contrast, the effect of b - II correction is better than Teillet-regression.
出处
《山地学报》
CSCD
北大核心
2016年第5期623-631,共9页
Mountain Research
基金
云南省森林灾害预警与控制重点实验室开放基金项目(ZK150001)
西南林业大学科学研究基金项目(111131)
国家自然科学基金项目(31400493
31260156)~~
关键词
高山峡谷地区
地形效应
地形校正
high mountain and gorge regions
topographic effect
topographic correction