基于多纹理特征融合的震后SAR图像倒塌建筑物信息提取

Earthquake Induced Building Damage Assessment on SAR Multi-texture Feature Fusion

合成孔径雷达(SAR)凭借其全天候观测能力以及SAR图像中丰富的纹理信息,在震后建筑物倒塌评估中发挥了重要作用。针对SAR图像中倒塌建筑物纹理特征多样但利用率较低,且特征信息冗余的问题,提出一种基于主成分分析的SAR图像多纹理特征分类方法。该方法基于灰度直方图、灰度共生矩阵、局部二值模式、Gabor滤波器提取了26种纹理特征信息,构建主成分变量进行多维特征优选与降维融合,通过随机森林分类算法提取建筑物的倒塌信息。以2016年日本熊本地震为例验证了该方法的有效性,结果显示其提取精度高达79.85%,倒塌建筑物的识别效率有所提高,分类结果优于单种纹理特征提取方法及多种纹理特征组合提取法,可用于震后建筑物震害信息的快速提取。

Synthetic Aperture Radar(SAR)plays an important role in building collapse assessment after earth⁃quake with its all-weather observation capability and rich texture information in SAR images.In order to solve the problems of multi-texture features of collapsed buildings in SAR images,such as low utilization rate and re⁃dundant feature information,a multi-texture feature classification method based on Principal Component Analy⁃sis(PCA)is proposed.This method extracts 26 kinds of texture feature information based on gray-level histo⁃gram,gray level co-occurrence matrix,Local Binary Pattern(LBP)and Gabor filters,constructs principal component variable for multi-dimensional feature selection and dimension reduction fusion,and extracts col⁃lapse information of buildings through Random Forest classification algorithm.Taking the Kumamoto earth⁃quake in Japan in 2016 as an example to verify the effectiveness of this method,the results show that the extrac⁃tion accuracy is up to 79.85%,the identification efficiency of collapsed buildings is improved,and the classifica⁃tion results are superior to each texture feature extraction method and multi-texture feature combination extrac⁃tion method,which can be used for the rapid extraction of earthquake damage information of buildings.