RADARSAT数字图像积雪覆盖下地物解译方法

采用加拿大RADARSAT雷达数字图像，以新疆阿勒泰地区阿勒泰市、布尔津、吉木乃、哈巴河、福海5个县(市)为典型试验区，研究了积雪覆盖下雷达遥感图像地物解译的基本原理和几何纠正方法，比较分析了标准模式的地理参考高分辨率合成孔径雷达图像与经过亮度增强、边缘增强、图像增强、斑点压缩和纹理分析处理的图像上地物目视解译的特点，建立了积雪覆盖下雷达图像上耕地、水系、交通道路、居民地等与抗灾救灾密切相关的地物的解译标志。结果表明，可弥补可见光及近红外遥感积雪监测的缺陷，为雪灾发生区的救援决策提供科学依据。

Interpretation of Snow-Covered Ground Objects Using RADARSAT Image

In this paper, all 5 counties of Aletai, Buerjin, Jimunai, Habahe and Fuhai in Aletai region, Northern Xinjiang are selected as typical study area. By use of Canada RADARSAT data, the geometric correction and fundamental principles of radar remote sensing are studied. And the image interpretation is comparatively analyzed between the rectified geo-referenced radar image and images processed by brightness adjustment, edge enhancement, image enhancement, speckle suppressionand texture analysis methods. The interpretation indicators of radar image are established for the important objects closelyrelated to activities of relief snow disaster, such as cropland, river, traffic, residentialareas and etc. Results showed that microwave radar remote sensing can avoid the difficulties in monitoring of snow disaster using visible and near-infrared remote sensing data, which can provides scientific informationfor decision making in snow disaster areas.