基于面向对象方法的汶川地震土地覆被快速提取
|
摘要
"5.12"汶川8.0级地震后,如何高精度快速提取灾害背景特征参数成为遥感技术在地震应急工作中应用研究的关键.随着高分辨率遥感影像的发展,传统基于像元的分类技术已不能满足需求,引入面向对象的信息提取技术,充分挖掘影像对象的纹理、形状和相互关系等信息,能够有效的提高震害的分类速度与精度.本文通过获取北川县城及附近震前震后的福卫二号影像,对北川县城震前震后的土地覆被变化进行面向对象方法的快速提取.研究表明,面向对象分类方法精度高,Kappa系数远大于传统的监督分类方法,提取速度快,能满足提取速度8小时以内的要求.
After the Wenchuan MS8.0 earthquake occurred on May 12,how to extracting disaster background characteristic parameters with high precision and speed has become the key to remote sensing technology applied in earthquake emergency work.With the development of high-resolution images,the traditional pixel-based classification is not able to meet the requirement.In order to fully excavate the texture,shape and relationship information of remote sensing data,we introduced an object-oriented method and way to extracting the land cover.This paper collects Beichuan FORMOSAT-2 images before and after the earthquake,uses object-oriented method to extracting of land cover rapidly in the pilot area before and after the earthquake.Compared to pixel-based classification,the results showed the object-oriented classification has high classification precision,its Kappa coefficient is much larger than traditional Supervised Classification method,extration speed is fast,it can completely meet the requirements of extracting land cover after earthquake in 8 hours.
After the Wenchuan MS8.0 earthquake occurred on May 12,how to extracting disaster background characteristic parameters with high precision and speed has become the key to remote sensing technology applied in earthquake emergency work.With the development of high-resolution images,the traditional pixel-based classification is not able to meet the requirement.In order to fully excavate the texture,shape and relationship information of remote sensing data,we introduced an object-oriented method and way to extracting the land cover.This paper collects Beichuan FORMOSAT-2 images before and after the earthquake,uses object-oriented method to extracting of land cover rapidly in the pilot area before and after the earthquake.Compared to pixel-based classification,the results showed the object-oriented classification has high classification precision,its Kappa coefficient is much larger than traditional Supervised Classification method,extration speed is fast,it can completely meet the requirements of extracting land cover after earthquake in 8 hours.
引文
[1]孙晓霞,张继贤,刘正军.利用面向对象的分类方法从IKONOS全色影像中提取河流和道路[J].测绘科学,2006,31(1):62-63.
[2]蔡银桥,毛政元.基于多特征对象的高分辨率遥感影像分类方法及其应用[J].国土资源遥感,2007(1):77-81.
[3]胡进刚,张晓东,沈欣,等.一种面向对象的高分辨率影像道路提取方法[J].遥感技术与应用,2006,21(3):184-188.
[4]田新光,张继贤,张永红.面向对象的红树林信息提取[J].海洋测绘,2007,27(2):41-44.
[5]MITRI G H,GITAS I Z.Fire type mapping using object-based classification of Ikonos imagery[J].International Journal of WildlandFire,2006,15:457-462.
[6]ZHOU W,TROY A,GROVE M.Object-based land cover classification and change analysis in the baltimore metropolitan area usingmultitemporal high resolution remote sensing data[J].Sensors,2008:1 613-1 636.
[7]王文宇,李博.基于eCogniton的高分辨率遥感图像的自动识别分类技术[J].北京建筑工程学院学报,2006,22(4):26-29.
[8]杜凤兰,田庆久,夏学齐,等.面向对象的地物分类法分析与评价[J].遥感技术与应用,2004,19(1):20-23.
[9]王晓青,窦爱霞,丁香,等.“地震遥感信息分析与处理系统”技术报告[R].北京:中国地震局地震预测研究所,2007:136.
[10]唐荣余.2008年汶川8.0级地震科学考察图集[M].北京:地震出版社,2009.
[2]蔡银桥,毛政元.基于多特征对象的高分辨率遥感影像分类方法及其应用[J].国土资源遥感,2007(1):77-81.
[3]胡进刚,张晓东,沈欣,等.一种面向对象的高分辨率影像道路提取方法[J].遥感技术与应用,2006,21(3):184-188.
[4]田新光,张继贤,张永红.面向对象的红树林信息提取[J].海洋测绘,2007,27(2):41-44.
[5]MITRI G H,GITAS I Z.Fire type mapping using object-based classification of Ikonos imagery[J].International Journal of WildlandFire,2006,15:457-462.
[6]ZHOU W,TROY A,GROVE M.Object-based land cover classification and change analysis in the baltimore metropolitan area usingmultitemporal high resolution remote sensing data[J].Sensors,2008:1 613-1 636.
[7]王文宇,李博.基于eCogniton的高分辨率遥感图像的自动识别分类技术[J].北京建筑工程学院学报,2006,22(4):26-29.
[8]杜凤兰,田庆久,夏学齐,等.面向对象的地物分类法分析与评价[J].遥感技术与应用,2004,19(1):20-23.
[9]王晓青,窦爱霞,丁香,等.“地震遥感信息分析与处理系统”技术报告[R].北京:中国地震局地震预测研究所,2007:136.
[10]唐荣余.2008年汶川8.0级地震科学考察图集[M].北京:地震出版社,2009.
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