SAR图像与可见光图像融合的建筑物提取算法

A Building Extraction Algorithm Based on the Fusion of SAR and Optical Images

提出了一种基于SAR图像和可见光图像融合的建筑物提取算法,由SAR图像提供建筑物的可能方位信息,由可见光图像提供建筑物提取所需的视觉特征。该算法充分利用建筑物在2幅已配准的SAR图像和可见光图像中所表现的共性和互补特征,采用边缘匹配和感知组合的方法提取建筑物的轮廓,在可见光图像中采用屋顶灰度均匀性特征进行建筑物验证。实验结果表明,本文算法具有较好的检测效果。

A building extraction algorithm based on the fusion of SAR and optical images is proposed. The possible position information of building is provided by SAR image,while the visual features required by outline extraction are provided by the optical image. The algorithm utilizes the common and complementary features of building shown in two aligned SAR and optical images,and extracts building outline through edge matching and perceptual grouping. Finally,the homogeneity of roof region in optical image is used to verify the existence of building. The experiment results demonstrate the good performance of the proposed algorithm.