摘要
提出将LiDAR数据对水体的敏感性与航空影像的高分辨率特征相结合的水体自动提取方法。利用SIFT算法对LiDAR强度图像和航空影像进行配准,在LiDAR高程图像上提取无回波信号的黑色区域,构建几何约束条件,排除由遮挡产生的无效区域;将水体初始位置映射到航空影像上,结合边缘信息进行区域生长,并对生长区域进行数学形态学运算,最终获取水体区域。实验结果表明,该方法可以获得很好的水体提取效果。
A new approach of automatic water body extraction based on LiDAR data and aerial images is proposed.The LiDAR intensity image and aerial images are co-registrated by SIFT operator.Transformation parameters from LiDAR image to aerial image can be computed by the matched conjugated points.Black regions that have no reflection on height image generated by LiDAR data are automatically extracted.Geometric constraint conditions are used to remove non-water body areas.Edge information is used for region growing with the projected inilial position as initial value.Finally,mathematical morphology techniques are used to get more precise water body areas.Experimental results show that the proposed approach can achieve very good water body information.
出处
《武汉大学学报（信息科学版）》
EI
CSCD
北大核心
2010年第8期936-940,共5页
Geomatics and Information Science of Wuhan University
基金
国家自然科学基金资助项目(40671157)
关键词
水体提取
LIDAR
航空影像
区域生长
数学形态学
water body extraction
LiDAR
aerial image
region growing
mathematical morphology
