基于Sentinel-1A数据的南京市水体信息提取

Water Information Extraction of Nanjing City Based on Sentinel-1A Data

基于先进的Sentinel-1A SAR数据,对南京地区的水体覆盖信息进行提取,首先对SAR图像进行滤波、辐射校正以及几何校正等预处理;再利用灰度共生矩阵提取影像中的纹理信息,并结合散射强度信息,利用SVM算法进行初分类;然后利用地形信息提取山体;最后从初分类结果中剔除山体阴影,得到水体提取结果。通过对比实验发现,该方法可有效去除山体阴影以及淹水期水田的影响,减少对水体信息的混淆,使结果与真实地表更加接近。研究结果能为南京市的水资源管理部门提供相应的理论支撑。

Based on the advanced Sentinel-1 A SAR data,we extracted the water information of Nanjing City.Firstly,we preprocessed the SAR images by filtering,radiation correction and geometric correction.And then,we used the gray level co-occurrence matrix to extract the texture information,and combining with the scattering intensity information,implemented the initial classification by SVM algorithm.Finally,we used terrain information to extract mountains,then removed the mountain shadow from the initial classification to obtain the final water extraction result.Through comparison experiments,it is found that the method can effectively remove the influence of the mountain shadow and the paddy field during the flooding period,and make the result closer to the real surface.This study can provide a corresponding theoretical support for the water resource management department of Nanjing City.