测绘通报 ›› 2017, Vol. 0 ›› Issue (12): 90-93,102.doi: 10.13474/j.cnki.11-2246.2017.0386

• 技术交流 • 上一篇    下一篇

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

谢嘉丽1, 李永树1, 李何超2, 吴玺2   

  1. 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
  • 收稿日期:2017-03-15 修回日期:2017-05-12 出版日期:2017-12-25 发布日期:2018-01-05
  • 作者简介:谢嘉丽(1992-),女,博士生,主要从事遥感影像处理及识别方面的研究工作。E-mail:254782451@qq.com
  • 基金资助:
    “十二五”国家科技支撑项目(2014BAL01B00)

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

XIE Jiali1, LI Yongshu1, LI Hechao2, WU Xi2   

  1. 1. Faculty of Geosciences and Environmental Engineering, Southwest Jiaotong University, Chengdu 611756, China;
    2. Center of Land Acquisition and Consolidation in Sichuan Province, Chengdu 610041, China
  • Received:2017-03-15 Revised:2017-05-12 Online:2017-12-25 Published:2018-01-05

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。

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

Abstract: 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.

Key words: gray-level co-occurrence matrix, textural feature, spectral feature, hollow village, damage building

中图分类号: