面向对象方法的复杂地形区地表覆盖信息提取
|
摘要
探讨在复杂地形区基于面向对象分类方法与中等分辨率的Landsat 8 OLI影像相结合能否获得较好的信息提取精度.结果表明:单波段标准差、多波段相关系数和最佳指数因子是选择波段的较优方案,可有效减少各波段间的信息冗余量.通过对研究区进行地理分区的方式建立每一地理子区的分类层次、设定分割参数及规则,根据地物类别的分布特征引入归一化植被指数、归一化建筑物指数、改进归一化差异水体指数、数字高程模型数据、坡度等与地类相关的专题数据,以提高复杂地形区分类精度;整个流域脑山区总分类精度最高,达88.33%,Kappa系数0.86.
This study was to explore whether a better accuracy for information extraction can be obtained by using the combination of object-oriented classification method and medium-resolution Landsat OLI images in the complex terrain areas. The results showed that the single band standard deviation, multiband correlation coefficient and optimum index factor were the optimal solution of band, which could effectively reduce information redundancy between bands. In order to improve the classification accuracy, the study area was divided into different subareas, and classification hierarchy, segmentation parameters and rules were established respectively for every subarea; at the same time, according to the spatial distribution of feature categories, some related thematic data such as normalization difference vegetation index, normalized difference built-up index, modified normalized difference water index, digital elevation model, slope etc. were introduced to help the classification. The total classifying accuracy of the whole study area was 88.33% and Kappa coefficient was 0.86.
This study was to explore whether a better accuracy for information extraction can be obtained by using the combination of object-oriented classification method and medium-resolution Landsat OLI images in the complex terrain areas. The results showed that the single band standard deviation, multiband correlation coefficient and optimum index factor were the optimal solution of band, which could effectively reduce information redundancy between bands. In order to improve the classification accuracy, the study area was divided into different subareas, and classification hierarchy, segmentation parameters and rules were established respectively for every subarea; at the same time, according to the spatial distribution of feature categories, some related thematic data such as normalization difference vegetation index, normalized difference built-up index, modified normalized difference water index, digital elevation model, slope etc. were introduced to help the classification. The total classifying accuracy of the whole study area was 88.33% and Kappa coefficient was 0.86.
引文
[1] Hay G J, Niernann K O, Goodenough D G. Spatial thresholds, image-objects and upscaling:a multiscale evaluation[J]. Remote Sensing of Environment, 1997,62(1):1-19.
[2]张毅,谭龙,陈冠,等.基于面向对象分类法的高分辨率遥感滑坡信息提取[J].兰州大学学报:自然科学版,2014, 50(5):745-750.
[3] Ruvimbo G, Philippe D M, Morgan D D. Object-oriented change detection for the city of Harare, Zimbabwe[J].Expert Systems with Applications, 2009, 36(1):571-588.
[4]陈云海,穆亚超,颉耀文.基于Landsat遥感影像的黑河干流中游湿地信息提取[J].兰州大学学报:自然科学版,2016,52(5):587-592.
[5]牛增懿,丁建丽,李艳华,等.基于高分一号影像的土壤盐渍化信息提取方法[J].干旱区地理,2016, 39(1):171-181.
[6] Honfmann P. Detecting informal settlements from IKONOS image data using methods of object oriented image analysis:an example from Cape Town(South Africa)[J].Remote Sensing of Urban Areas/Femerkundung in Urbanen Raumen, 2001,16(4):654-665.
[7] Giada S,Groeve T, Ehrlich D, et al. Information extraction from very high resolution satellite imagery over Lukole refugee camp, Tanzania[J]. International Journal of Remote Sensing, 2003, 24(22):4251-4266.
[8]王彩艳,王瑷玲,王介勇,等.基于面向对象的北京市区城市内部用地信息提取[J].自然资源学报,2015,30(4):705-714.
[9]张小平,何伟,方婷.湟水谷地农业生态经济系统的能值分析:以西宁市为例[J].干旱区地理,2011,34(2):344-354.
[10]冯仕超,高小红,顾娟.基于CLUE-S模型的湟水流域土地利用空间分布模拟[J].生态学报,2013, 33(3):985-996.
[11]董旭,张胜邦,张更权.青海湟水流域生态保护与建设发展战略研究[M].北京:中国林业出版社,2008.
[12]贾伟,高小红.河谷型城市热岛空间分布格局演变及对策:以西宁市区为例[J].地球信息科学学报,2014,16(4):592-601.
[13]邱玉利,周建中,马林.湟水流域地表水资源特征[J].水资源与水工程学报,2007, 18(6):98-102.
[14]张艳娇,高小红,李其江,等.气候变化情景下青海湟水流域径流变化的HIMS模拟分析[J].地理科学进展,2013, 32(1):59-67.
[15]青海省农牧厅.青海省黄河、湟水流域特色农牧业发展规划(2008-2010年)[R].西宁:青海省农牧厅,2008.
[16]李金山.复杂地形区土地利用/土地覆被分类研究:以湟水流域为例[D].西宁:青海师范大学生命与地理科学学院,2013.
[17]曾永年,勒文凭,何丽丽,等.青海高原东部农业区土地利用遥感分类制图[J].农业工程学报,2012, 28(16):225-231.
[18]贾伟.面向对象的复杂地形区土地利用信息提取研究:以湟水流域为例[D].西宁:青海师范大学生命与地理科学学院,2015.
[19]刘纪远.中国资源环境遥感宏观调查与动态分析[M].北京:中国科学技术出版社,1996.
[20]安琼,杨邦杰,郭琳.面向对象的多光谱图像特征遗传选优方法[J].农业工程学报,2008, 24(4):181-185.
[21] Price J C. An approach for analysis of reflectance spec-tra[J]. Remote Sensing of Environment, 1998, 64(3):316-330.
[22]高恒振,万建伟,朱珍珍,等.基于波段子集特征提取的最小二乘支持向量机高光谱图像分类技术[J].光谱学与光谱分析,2011,31(5):1314-1317.
[23]崔志杰,郭鹏,熊木齐,等.基于ASTER遥感影像武都区松散堆积物研究[J].兰州大学学报:自然科学版,2014, 50(5):739-744.
[24]李金山,高小红.西宁市城镇用地信息提取方法研究[J].测绘科学,2014, 39(3):48-51.
[2]张毅,谭龙,陈冠,等.基于面向对象分类法的高分辨率遥感滑坡信息提取[J].兰州大学学报:自然科学版,2014, 50(5):745-750.
[3] Ruvimbo G, Philippe D M, Morgan D D. Object-oriented change detection for the city of Harare, Zimbabwe[J].Expert Systems with Applications, 2009, 36(1):571-588.
[4]陈云海,穆亚超,颉耀文.基于Landsat遥感影像的黑河干流中游湿地信息提取[J].兰州大学学报:自然科学版,2016,52(5):587-592.
[5]牛增懿,丁建丽,李艳华,等.基于高分一号影像的土壤盐渍化信息提取方法[J].干旱区地理,2016, 39(1):171-181.
[6] Honfmann P. Detecting informal settlements from IKONOS image data using methods of object oriented image analysis:an example from Cape Town(South Africa)[J].Remote Sensing of Urban Areas/Femerkundung in Urbanen Raumen, 2001,16(4):654-665.
[7] Giada S,Groeve T, Ehrlich D, et al. Information extraction from very high resolution satellite imagery over Lukole refugee camp, Tanzania[J]. International Journal of Remote Sensing, 2003, 24(22):4251-4266.
[8]王彩艳,王瑷玲,王介勇,等.基于面向对象的北京市区城市内部用地信息提取[J].自然资源学报,2015,30(4):705-714.
[9]张小平,何伟,方婷.湟水谷地农业生态经济系统的能值分析:以西宁市为例[J].干旱区地理,2011,34(2):344-354.
[10]冯仕超,高小红,顾娟.基于CLUE-S模型的湟水流域土地利用空间分布模拟[J].生态学报,2013, 33(3):985-996.
[11]董旭,张胜邦,张更权.青海湟水流域生态保护与建设发展战略研究[M].北京:中国林业出版社,2008.
[12]贾伟,高小红.河谷型城市热岛空间分布格局演变及对策:以西宁市区为例[J].地球信息科学学报,2014,16(4):592-601.
[13]邱玉利,周建中,马林.湟水流域地表水资源特征[J].水资源与水工程学报,2007, 18(6):98-102.
[14]张艳娇,高小红,李其江,等.气候变化情景下青海湟水流域径流变化的HIMS模拟分析[J].地理科学进展,2013, 32(1):59-67.
[15]青海省农牧厅.青海省黄河、湟水流域特色农牧业发展规划(2008-2010年)[R].西宁:青海省农牧厅,2008.
[16]李金山.复杂地形区土地利用/土地覆被分类研究:以湟水流域为例[D].西宁:青海师范大学生命与地理科学学院,2013.
[17]曾永年,勒文凭,何丽丽,等.青海高原东部农业区土地利用遥感分类制图[J].农业工程学报,2012, 28(16):225-231.
[18]贾伟.面向对象的复杂地形区土地利用信息提取研究:以湟水流域为例[D].西宁:青海师范大学生命与地理科学学院,2015.
[19]刘纪远.中国资源环境遥感宏观调查与动态分析[M].北京:中国科学技术出版社,1996.
[20]安琼,杨邦杰,郭琳.面向对象的多光谱图像特征遗传选优方法[J].农业工程学报,2008, 24(4):181-185.
[21] Price J C. An approach for analysis of reflectance spec-tra[J]. Remote Sensing of Environment, 1998, 64(3):316-330.
[22]高恒振,万建伟,朱珍珍,等.基于波段子集特征提取的最小二乘支持向量机高光谱图像分类技术[J].光谱学与光谱分析,2011,31(5):1314-1317.
[23]崔志杰,郭鹏,熊木齐,等.基于ASTER遥感影像武都区松散堆积物研究[J].兰州大学学报:自然科学版,2014, 50(5):739-744.
[24]李金山,高小红.西宁市城镇用地信息提取方法研究[J].测绘科学,2014, 39(3):48-51.