基于高分二号的云南松林遥感影像提取方法研究
|
摘要
基于高空间分辨率的高分二号遥感影像,建立包含影像光谱特征、几何特征、纹理特征在内的分类特征集,使用面向对象的分类算法提取研究区的云南松林。研究表明:光谱特征、几何特征和纹理特征同时参与分类,得到的分类效果比单独使用一种特征或两种特征组合的分类效果更显著;与贝叶斯分类算法、K最邻近分类算法、支持向量机分类算法相比,随机森林算法的分类结果总体精度最高,为93.35%,对研究区云南松林提取的生产者精度为96.91%,用户精度为92.35%;云南松林面积占研究区面积的40.10%,是研究区最为重要的树种。
Based on high-spatial GF-2 imagery,the classification feature set which contained spectral features,geometric features,and texture features was established. Different object-oriented classification algorithms were used to extract Pinus yunnanensis forest. The results showed that( a) When put spectral features,geometric features and texture features into random forest classifier simultaneously,it had the best classification effect than using only one kind of feature or the combination of two kinds of features.( b) By comparing with Bayes classifier,Knearest neighbor classifier and support vector machine( SVM) classifier,the classification results of random forest classifier met the highest overall accuracy which was 93. 35%,the producer accuracy and user accuracy of Pinus yunnanensis was 96. 91% and 92. 35%,respectively.( c) The proportion of Pinus yunnanensis forest is 40. 10%,which is the most significant tree species in the study area.
Based on high-spatial GF-2 imagery,the classification feature set which contained spectral features,geometric features,and texture features was established. Different object-oriented classification algorithms were used to extract Pinus yunnanensis forest. The results showed that( a) When put spectral features,geometric features and texture features into random forest classifier simultaneously,it had the best classification effect than using only one kind of feature or the combination of two kinds of features.( b) By comparing with Bayes classifier,Knearest neighbor classifier and support vector machine( SVM) classifier,the classification results of random forest classifier met the highest overall accuracy which was 93. 35%,the producer accuracy and user accuracy of Pinus yunnanensis was 96. 91% and 92. 35%,respectively.( c) The proportion of Pinus yunnanensis forest is 40. 10%,which is the most significant tree species in the study area.
引文
[1]Nelson R,Krabill W,Tonelli J.Estimating forest biomass and volume using airborne laser data[J].Remote Sensing of Environment,1988,24(2):247-267.
[2]李俊明,邢艳秋,杨超,等.基于环境与灾害监测预报小卫星的树种识别[J].东北林业大学学报,2013,41(11):41-45.
[3]刘旭升,张晓丽.森林植被遥感分类研究进展与对策[J].林业资源管理,2004(1):61-64.
[4]周洪泽,曲智林.连续森林清查应用地理信息估计蓄积量的方法[J].东北林业大学学报,2000,28(5):50-53.
[5]HELLER R C,DOVERSPIKE G E,ALDRICH R C.Identification of tree species on large-scale panchromatic and color aerial photographs[M].Washington D C:USDA Handbook,1964,261:1-17.
[6]Vieira I C G,Almeida A S D,Davidson E A,et al.Classifying successional forests using Landsat spectral properties and ecological characteristics in eastern Amaz8nia[J].Remote Sensing of Environment,2003,87(4):470-481.
[7]张丽云.基于高光谱遥感数据的森林树种分类[D].北京:北京林业大学,2016.
[8]李子艺,王振锡,岳俊,等.基于BP神经网络的高光谱果树树种识别研究[J].江苏农业科学,2016,45(5):410-414.
[9]周艳飞.基于高分辨遥感数据的胡杨、柽柳识别与树冠提取研究[D].乌鲁木齐:新疆农业大学,2014.
[10]尹凌宇,覃先林,孙桂芬,等.基于高分二号多光谱数据的树种识别方法[J].林业资源管理,2016(4):121-127.
[11]魏晶昱,毛学刚,方本煜,等.基于Landsat 8 OLI辅助的亚米级遥感影像树种识别[J].北京林业大学学报,2016,38(11):23-33.
[12]Breiman L.Random Forests[J].Machine Learning,2001,45(1):5-32.
[13]M.Pal.Random forest classifier for remote sensing classification[J].International Journal of Remote Sensing,2005,26(1):217-222.
[14]刘毅,杜培军,郑辉,等.基于随机森林的国产小卫星遥感影像分类研究[J].测绘科学,2012,37(4):198-200.
[15]张晓羽,李凤日,甄贞,等.基于随机森林模型的陆地卫星-8遥感影像森林植被分类[J].东北林业大学学报,2016,44(6):53-57.
[16]Chubey M S.Object-based Analysis of Ikonos-2 Imagery for Extraction of Forest Inventory Parameters[J].Photogrammetric Engineering&Remote Sensing,2006,72(4):págs.383-394.
[17]Yu Q,Gong P,Clinton N,et al.Object-based Detailed Vegetation Classification with Airborne High Spatial Resolution Remote Sensing Imagery[J].Photogrammetric Engineering&Remote Sensing,2006,72(7):799-811.
[18]段晓梅,焦丽丽,王芳芳,等.石林风景名胜保护区景观植被[J].中国城市林业,2007,5(2):54-56.
[19]佟越.石林风景名胜区大石林景区植被研究[D].昆明:西南林业大学,2008.
[20]杜珂磊,唐斌,李佳.基于e Cognition的玉树结古镇灾情动态监测研究[J].测绘与空间地理信息,2016,39(2):31-35.
[21]刘龙飞,陈云浩,李京.遥感影像纹理分析方法综述与展望[J].遥感技术与应用,2003,18(6):441-447.
[22]王昆,张晓丽,王珊,等.鹫峰地区Quick Bird影像纹理特征与生物量估测关系初探[J].地理与地理信息科学,2013,29(3):52-55.
[23]田绍鸿,张显峰.采用随机森林法的天绘数据干旱区城市土地覆盖分类[J].国土资源遥感,2016,28(1):43-49.
[24]黄健熙,侯矞焯,苏伟,等.基于GF-1 WFV数据的玉米与大豆种植面积提取方法[J].农业工程学报,2017,33(7):164-170.
[25]王栋,岳彩荣,田传召,等.基于随机森林的大姚县TM遥感影像分类研究[J].林业调查规划,2014(2):1-5.
[26]钟文君,兰樟仁.基于高空间分辨率遥感影像的湿地信息提取技术研究[J].云南地理环境研究,2007,19(5):134-139.
[2]李俊明,邢艳秋,杨超,等.基于环境与灾害监测预报小卫星的树种识别[J].东北林业大学学报,2013,41(11):41-45.
[3]刘旭升,张晓丽.森林植被遥感分类研究进展与对策[J].林业资源管理,2004(1):61-64.
[4]周洪泽,曲智林.连续森林清查应用地理信息估计蓄积量的方法[J].东北林业大学学报,2000,28(5):50-53.
[5]HELLER R C,DOVERSPIKE G E,ALDRICH R C.Identification of tree species on large-scale panchromatic and color aerial photographs[M].Washington D C:USDA Handbook,1964,261:1-17.
[6]Vieira I C G,Almeida A S D,Davidson E A,et al.Classifying successional forests using Landsat spectral properties and ecological characteristics in eastern Amaz8nia[J].Remote Sensing of Environment,2003,87(4):470-481.
[7]张丽云.基于高光谱遥感数据的森林树种分类[D].北京:北京林业大学,2016.
[8]李子艺,王振锡,岳俊,等.基于BP神经网络的高光谱果树树种识别研究[J].江苏农业科学,2016,45(5):410-414.
[9]周艳飞.基于高分辨遥感数据的胡杨、柽柳识别与树冠提取研究[D].乌鲁木齐:新疆农业大学,2014.
[10]尹凌宇,覃先林,孙桂芬,等.基于高分二号多光谱数据的树种识别方法[J].林业资源管理,2016(4):121-127.
[11]魏晶昱,毛学刚,方本煜,等.基于Landsat 8 OLI辅助的亚米级遥感影像树种识别[J].北京林业大学学报,2016,38(11):23-33.
[12]Breiman L.Random Forests[J].Machine Learning,2001,45(1):5-32.
[13]M.Pal.Random forest classifier for remote sensing classification[J].International Journal of Remote Sensing,2005,26(1):217-222.
[14]刘毅,杜培军,郑辉,等.基于随机森林的国产小卫星遥感影像分类研究[J].测绘科学,2012,37(4):198-200.
[15]张晓羽,李凤日,甄贞,等.基于随机森林模型的陆地卫星-8遥感影像森林植被分类[J].东北林业大学学报,2016,44(6):53-57.
[16]Chubey M S.Object-based Analysis of Ikonos-2 Imagery for Extraction of Forest Inventory Parameters[J].Photogrammetric Engineering&Remote Sensing,2006,72(4):págs.383-394.
[17]Yu Q,Gong P,Clinton N,et al.Object-based Detailed Vegetation Classification with Airborne High Spatial Resolution Remote Sensing Imagery[J].Photogrammetric Engineering&Remote Sensing,2006,72(7):799-811.
[18]段晓梅,焦丽丽,王芳芳,等.石林风景名胜保护区景观植被[J].中国城市林业,2007,5(2):54-56.
[19]佟越.石林风景名胜区大石林景区植被研究[D].昆明:西南林业大学,2008.
[20]杜珂磊,唐斌,李佳.基于e Cognition的玉树结古镇灾情动态监测研究[J].测绘与空间地理信息,2016,39(2):31-35.
[21]刘龙飞,陈云浩,李京.遥感影像纹理分析方法综述与展望[J].遥感技术与应用,2003,18(6):441-447.
[22]王昆,张晓丽,王珊,等.鹫峰地区Quick Bird影像纹理特征与生物量估测关系初探[J].地理与地理信息科学,2013,29(3):52-55.
[23]田绍鸿,张显峰.采用随机森林法的天绘数据干旱区城市土地覆盖分类[J].国土资源遥感,2016,28(1):43-49.
[24]黄健熙,侯矞焯,苏伟,等.基于GF-1 WFV数据的玉米与大豆种植面积提取方法[J].农业工程学报,2017,33(7):164-170.
[25]王栋,岳彩荣,田传召,等.基于随机森林的大姚县TM遥感影像分类研究[J].林业调查规划,2014(2):1-5.
[26]钟文君,兰樟仁.基于高空间分辨率遥感影像的湿地信息提取技术研究[J].云南地理环境研究,2007,19(5):134-139.