面向对象遥感图像处理方法在建筑物震害评估中的应用研究
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
以汶川地震中的都江堰地区震后遥感图像为例,研究分析了图像分割技术及模糊分类技术,在此基础上,用分类精度更高的单尺度、多特征的面向对象图像处理方法进行建筑物震害评估。重点探索了混淆矩阵在震害评估中的应用,通过Kappa统计值进行震害分类结果的比对与精度评价,并验证该分类方法的稳定性。研究结果表明:改进后的面向对象图像处理方法简单、快速且精度较高,具有良好的适用性和稳定性。
This paper took the Dujiangyan as an example, discussed the image segmentation and fuzzy-logic classification. And then use the method of single scale and multiple features for the damage assessment. Focus on exploring the application of the Confusion Matrix in the damage assessment. Make a comparison and accuracy evaluation of the classification results with Kappa statistics. Then verify the stability of the classification method. The results show that the improved object-oriented image processing method is simple, fast and highly accurate. This method takes on good applicability and stability.
引文
[1]赵福军.遥感影像震害信息提取技术研究[D].北京:中国地震局工程力学研究所,2010.
    [2]Matsuyama.K.Current Account Dynamics in a Finite Horizon Model[J].Journal of International Economics,1987:299-313.
    [3]赵宇鸾,林爱文.基于面向对象和多尺度影像分割技术的城市用地分类研究-以武汉市城市中心区为例[J].国土资源科技管理,2008,25(5):90-95.
    [4]沈阳.基于形态学的图像边缘检测技术研究[D].成都:电子科技大学,2008.
    [5]关元秀.高分辨率遥感卫星处理指南[M].北京:科学出版社,2008.
    [6]曾志远.遥感图像大气校正方法综述[J].遥感信息,2005,(4):66-70.
    [7]赵玉涛.景观异质性研究评述[J].应用生态学报,2002,13(4):495-500.
    [8]尹京苑,柳稼航,单新建.基于图像结构信息的城市房屋震害特征自动提取技术[J].遥感信息,2004,12(1):27-30.
    [9]Congalton,R.G.A Review of Assessing the Accuracy of Classifications of Remotely Sensed Data[J].Remote Sensing Environ,1991,37:35-46.
    [10]Stehman.S.V.Selecting and Interpreting Measures of the Matic Classification Accuracy[J].Remote Sensing Environ,1997,62:77-89.
    [11]Richards.J.A.Classifier Performance and Map Accuracy[J].Remote Sensing Environ,1996,(57):161-166.

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心