基于双层上下文语义分类网的遥感郊外建筑物检测研究

Remote Sensing Suburb Building Detection Based on Double-Layer Context Semantic Classification Network

本文提出了一种基于双层上下文语义网的遥感郊外建筑物检测方法。该方法主要包括以下两个阶段。第一阶段:郊外大视场遥感图像疑似区快速提取;第二阶段:通过构建双层上下文语义分类网对建筑物疑似区精准鉴别。实验表明,本算法具备检测精度高、虚警率低以及运算效率高等优点。

This paper proposed a remote sensing suburban building detection method based on double-layer context semantic network.This method mainly includes the following two stages.The first stage is the rapid extraction of suspected areas from remote sensing images with large field of view in suburbs;the second stage is the accurate identification of suspected areas of buildings by constructing a two-tier context semantic classification network.Experiments show that the algorithm has the advantages of high detection accuracy,low false alarm rate and high operational efficiency.