联合星载光学与SAR图像的城市大面积建筑物高度快速提取

星载高分辨率光学图像与SAR图像广泛应用于城市建筑物高度提取,但光学图像存在缺少相关卫星参数的情况,而SAR图像则存在散射特征不完整以及提取效率低等缺陷。针对以上问题,本文提出一种联合高分辨率星载光学与SAR图像的城市大面积建筑物高度快速提取方法。首先,结合支持向量机(SVM)和形态学阴影指数(MSI)快速提取光学图像中的阴影并自动测量阴影长度;之后选择多个合适样本,基于模型匹配法从SAR图像中提取高度;最后将高度与阴影长度作线性回归分析,建立数学模型来提取其他建筑物的高度。该方法将不同卫星系统的数据和特征相结合,互相弥补各自缺陷,不仅提高了效率、降低了成本,同时满足精度要求。

Fusion of Spaceborne Optical and SAR Images for Building Height Quick Extraction in Big Urban Areas

The spaceborne high-resolution optical images and synthetic aperture radar (SAR) images are applied to extract building height in urban areas widely.But the lack of optical satellite parameters along with the SAR images' incomplete scattering characteristics and inefficient extraction make the application flawed.To cure the above problems, we investigated the joint use of the spaceborne high-resolution optical images and SAR images to extract building height information quickly in big urban areas.The chain is decomposed into the main following steps: First, the building shadows are extracted by integrating support vector machines (SVM) with morphological shadow index (MSI) and their lengths are measured automatically.Then, the height extraction from SAR images based on a model matching technique for some appropriate samples.Finally, obtain the other heights based on the simple linear regression analysis.This approach which combines the data and feature from different satellite systems to make up the flaws for each other is not only efficient and low-cost, but also satisfy the basic accuracy requirement.