利用灰度共生矩阵纹理特征识别空心村损毁建筑物的方法  

Recognition of Damage Buildings in Hollow Village Based on Texture Feature of Gray Level Co-occurrence Matrix

作　　者：谢嘉丽 李永树 李何超[2] 吴玺[2] 

机构地区：[1]西南交通大学地球科学与环境工程学院,四川成都611756 [2]四川省土地统征整理事务中心,四川成都610041

出　　处：《测绘通报》2017年第12期90-93,102,共5页Bulletin of Surveying and Mapping

基　　金：“十二五”国家科技支撑项目(2014BAL01B00)

摘　　要：从无人机影像上快速识别空心村损毁建筑物,不仅能精确掌握损毁建筑物的位置分布,还能为实地调研提供指导材料。本文首先在试验区范围内获取建筑物影像,并对这些建筑物逐一标记。然后以单个建筑物为对象,计算纹理特征参数和光谱特征参数,选取与损毁程度正相关的参数标准化后组成特征参数向量,根据向量的可视化结果获取损毁建筑物的分布、损毁程度等信息。试验结果表明,该方法能实现从无人机影像上识别出损毁建筑物,并能对建筑物的损毁程度得到初步认识,可以有效实现高分辨率影像中空心村损毁建筑物的识别。Rapid recognition of hollow villages from UAV images can not only accurately get location distribution of damaged buildings, but also provide guidance material for field investigation.This paper presents a method of visualizing damage buildings in hollow villages. Firstly,we mark the buildings one by one in experimental area.Then,the texture and spectral feature parameters of each building are calculated.For a single building,parameters that are positively correlated with the level of damage are normalized to form a parameter vector. Finally, the distribution and damage level of damaged buildings can be exhibited according to the visualization results of parameter vectors.Experimental results show that this method can realize recognizing damage buildings in UAV images, and get initial understanding of buildings damage level. In summary, this method can effectively and rapidly realize the recognition of damaged buildings with high resolution images.

关 键 词：灰度共生矩阵 纹理特征 光谱特征 空心村 损毁建筑物 

分 类 号：P237[天文地球—摄影测量与遥感]