欢迎访问《地震》,

地震 ›› 2013, Vol. 33 ›› Issue (2): 115-122.

• • 上一篇    下一篇

基于数学形态学的高分遥感影像地震重点目标信息提取

张小咏, 买莹, 张凌   

  1. 中国地震应急搜救中心, 北京 100049
  • 收稿日期:2013-01-01 出版日期:2013-04-30 发布日期:2020-09-27
  • 作者简介:张小咏(1976-), 女, 黑龙江绥化人, 副研究员, 主要从事地震遥感灾害评估, 地震灾情信息推演等研究。
  • 基金资助:
    国家十二五科技支撑项目(2012BAK15B05)

Extraction of Earthquake Key Objects from High Spatial Resolution Images Based on Mathematical Morphology

ZHANG Xiao-yong, MAI Ying, ZHANG Ling   

  1. National Earthquake Response Support Service, Beijing 100049, China
  • Received:2013-01-01 Online:2013-04-30 Published:2020-09-27

摘要: 重点目标是抗震救援的关键, 重点目标的破坏大多与目标本身的破坏或破坏前后的形态变化有关。 本文利用数学形态学方法对地物形态特征(包括形状、 大小、 方向等)处理的优势和思路, 构建了重点目标信息提取流程和算法, 并利用海地地震的高分辨率遥感影像, 对震后油罐破损状态的分离和提取、 以及对机场跑道旧损状态的提取。 试验结果表明, 对于形态上能够明显区别于背景的重点目标, 可以利用灰度和结构信息通过数学形态学进行提取。 而且数学形态学算法构造灵活, 处理速度快, 便于硬件实现, 对应急救援的快速处理有一定应用前景。

关键词: 地震救援, 数学形态学, 重点目标, 信息提取

Abstract: Earthquake key objects are densely populated areas and key facilities, which include schools, hospitals, dangerous goods, transportation hub and so on. They are the main goal of the emergency and rescue after earthquakes occurred. This paper makes full use of the processing advantages of morphological characteristics (including shape, size, direction, etc.) based on mathematical morphology, and constructs the extraction algorithm and process of the key object information. And then, the method has been tested using high resolution remote sensing image from the 2010 Haiti earthquake. Separation and extraction of the oil tank damage state is realized. In the same time, the airport runway old damage state is extracted. The test results show that key goals obviously differentiated from the background, gray scale and structure of information can be used for extraction through mathematical morphology. Test results also show that mathematical morphology algorithm has application prospect in the emergency rescue rapid processing.

Key words: Earthquake rescue, Mathematical morphology, Key object, Information abstraction

中图分类号: