基于HSI高光谱数据的水稻光谱特征分析与识别技术研究
|
摘要
水稻是中国的主要粮食作物,实时准确地获取水稻面积等信息,有利于国家粮食安全和贸易安全的保证。传统水稻遥感监测主要集中利用TM、CBERS星的CCD数据以及MODIS数据。HSI数据是国内首个航天高光谱数据,国内对HSI数据在水稻分类识别中的研究几乎空白。以广西玉林市-博白地带2010年10月22日的一景HSI数据为研究区,主要进行了如下研究工作:
(1)HSI数据介绍和预处理研究:对HSI数据的特点、命名规则等进行介绍。针对HSI数据产品特点,对HSI数据进行包括数据格式转换、绝对辐射亮度值变换、FLAASH大气校正、质量差波段去除和几何精校正等预处理工作。
(2)晚稻等地物类光谱特征分析:数据预处理后,根据研究区概况和同时相HJ-1A星CCD多光谱数据情况,获取晚稻等地物的空间信息,并得到不同地物在HSI数据的光谱变化曲线,分析晚稻与其它地物的光谱差异,为波段选择做好铺垫。
(3)波段选择方法研究:通过对信息量和类别可分性波段选择方法进行研究,在此基础上,提出改进的基于信息量和类别可分性的组合波段选择方法,最后选择得分值排序靠前的30个波段作为波段选择结果,为后面的精准分类做了铺垫。
(4) SVM在研究区晚稻识别研究:基于SVM和MLC理论,进行HSI数据的晚稻识别,得出波段选择后,SVM对HSI数据总体分类精度最高,其次为波段选择后的MLC方法,而基于非波段选择数据的分类准确率低于波段选择后的结果,证明波段选择在高光谱遥感分类中的必要性,并说明SVM在高光谱分类中的优越性;在SVM理论下,HSI数据晚稻识别精度和同时相CCD数据对比,得出前者总体精度为89.46,略低于后者3.36个百分点,但其仍具有较高的分类精度,可为进一步的定量遥感奠定基础。
Rice is the main grain crop in china, obtaining punctually and exactly rice area and other information is beneficial for the guarantee of the country’s food supply security and trade security.The traditional remote-sensing monitoring of rice mainly focused use of TM imagery、CCD imagery of CBERS satellite and MODIS data. The HSI imagery of HJ-1A Satellite is not researched in domestic rice identification. the region of yulin city-bobai country in guangxi was used as research area,many works was done as follows:
(1)HSI Imagery Introduction and Proprecessing:the article introduced the character of HSI imagery,etc.some proprecessing work were done,such as data format transformation , absolute radiance value transformation,FLAASH atmospheric correction,poor quality band removal and Geometric Correction,etc.
( 2 ) The Spectral Difference of Rice and other geo-objects : After preprocessing of HSI imagery, the distribution of rice and other geo-objects were obtained according to the summary of the research area and the Multispectral CCD Imagery of HJ-1A Satellite,The analysis of spectral difference was researched based on the spectral curve of different geo-objects,which laid a strong theoretical foundation for Band Selection .
(3)The research of Band Selection methods:Band Selection methods were introduced based on the Information of Band and the Class Separability. the innovative approach of Band Selection,the combination of Band Informationt and Class Separability,was raised by the foundation of the above methods. The top 30 band of HSI imagery were reserved as the result of band selection according to the score of different bands in some two geo-objects,which made good bedding for the following Classification.
(4)The Identification of Rice in research area based on the method of SVM: The Identification of Rice in HSI Imagery were introduced by the methods of SVM and MLC.the collusion was as follows:In classification accuracy,after Band Selection,the result of HSI imagery based on SVM is the best result,then, the result of HSI Imagery based on MLC is the better result,finally, because of not Band Selection,the result of HSI Imagery is lower,but the result of SVM is higher than MLC under the circumstance.So,it is essential to select Band in Hyperspectral Remote Sensing,and the SVM have the advantage of Hyperspectral Imagery Classification. Under the guidance of SVM method,compared the accuracy of Rice in HSI Imagery with the accuracy of Rice in the same temporal CCD Imagery,the overall accuracy of the former which was 89.46% was lower than the latter,but the HSI imagery still had the good classification accuracy.
(1)HSI数据介绍和预处理研究:对HSI数据的特点、命名规则等进行介绍。针对HSI数据产品特点,对HSI数据进行包括数据格式转换、绝对辐射亮度值变换、FLAASH大气校正、质量差波段去除和几何精校正等预处理工作。
(2)晚稻等地物类光谱特征分析:数据预处理后,根据研究区概况和同时相HJ-1A星CCD多光谱数据情况,获取晚稻等地物的空间信息,并得到不同地物在HSI数据的光谱变化曲线,分析晚稻与其它地物的光谱差异,为波段选择做好铺垫。
(3)波段选择方法研究:通过对信息量和类别可分性波段选择方法进行研究,在此基础上,提出改进的基于信息量和类别可分性的组合波段选择方法,最后选择得分值排序靠前的30个波段作为波段选择结果,为后面的精准分类做了铺垫。
(4) SVM在研究区晚稻识别研究:基于SVM和MLC理论,进行HSI数据的晚稻识别,得出波段选择后,SVM对HSI数据总体分类精度最高,其次为波段选择后的MLC方法,而基于非波段选择数据的分类准确率低于波段选择后的结果,证明波段选择在高光谱遥感分类中的必要性,并说明SVM在高光谱分类中的优越性;在SVM理论下,HSI数据晚稻识别精度和同时相CCD数据对比,得出前者总体精度为89.46,略低于后者3.36个百分点,但其仍具有较高的分类精度,可为进一步的定量遥感奠定基础。
Rice is the main grain crop in china, obtaining punctually and exactly rice area and other information is beneficial for the guarantee of the country’s food supply security and trade security.The traditional remote-sensing monitoring of rice mainly focused use of TM imagery、CCD imagery of CBERS satellite and MODIS data. The HSI imagery of HJ-1A Satellite is not researched in domestic rice identification. the region of yulin city-bobai country in guangxi was used as research area,many works was done as follows:
(1)HSI Imagery Introduction and Proprecessing:the article introduced the character of HSI imagery,etc.some proprecessing work were done,such as data format transformation , absolute radiance value transformation,FLAASH atmospheric correction,poor quality band removal and Geometric Correction,etc.
( 2 ) The Spectral Difference of Rice and other geo-objects : After preprocessing of HSI imagery, the distribution of rice and other geo-objects were obtained according to the summary of the research area and the Multispectral CCD Imagery of HJ-1A Satellite,The analysis of spectral difference was researched based on the spectral curve of different geo-objects,which laid a strong theoretical foundation for Band Selection .
(3)The research of Band Selection methods:Band Selection methods were introduced based on the Information of Band and the Class Separability. the innovative approach of Band Selection,the combination of Band Informationt and Class Separability,was raised by the foundation of the above methods. The top 30 band of HSI imagery were reserved as the result of band selection according to the score of different bands in some two geo-objects,which made good bedding for the following Classification.
(4)The Identification of Rice in research area based on the method of SVM: The Identification of Rice in HSI Imagery were introduced by the methods of SVM and MLC.the collusion was as follows:In classification accuracy,after Band Selection,the result of HSI imagery based on SVM is the best result,then, the result of HSI Imagery based on MLC is the better result,finally, because of not Band Selection,the result of HSI Imagery is lower,but the result of SVM is higher than MLC under the circumstance.So,it is essential to select Band in Hyperspectral Remote Sensing,and the SVM have the advantage of Hyperspectral Imagery Classification. Under the guidance of SVM method,compared the accuracy of Rice in HSI Imagery with the accuracy of Rice in the same temporal CCD Imagery,the overall accuracy of the former which was 89.46% was lower than the latter,but the HSI imagery still had the good classification accuracy.
引文
[1] X Xiao,S Boles,J Liu,et al.Mapping paddy rice agriculture in southern China using multi-temporal MODIS images [J].Remote Sensing of Environment,2005.95(4):480-492.
[2]国家统计局农村社会经济调查司.中国农村统计年鉴[M].北京:中国统计出版社,2010:70-80.
[3]王人潮,史舟等.农业信息科学与农业信息技术[M].北京:中国农业出版社,2003:3-5.
[4] Roy F. Bartlett.Estimating the Total of a continuous population[J].Journal of Statistical Planning and Inference,1986.13:51-66.
[5]刘海启,金敏毓,龚维鹏.美国农业遥感技术应用状况概述[J].中国农业资源与区划,1999.20(2):56-60.
[6] S C Liew, S P Kam, T P Tuong et al.Application of multitemporal ERS-2 synthetic aperture radar in delineating rice cropping systems in the Mekong River Delta,Vietnam[J].GeosciencesandRemote Sensing,1998.36(5):1412-1420.
[7] P .P .Nageswara Rao.V. R. Rao. Rice crop identification and area estimation using remotely-sensed data from Indian cropping patterns[J].International Journal of Remote Sensing.1987,8(4):639-650.
[8]张峰,吴炳方.泰国水稻种植面积月变化的遥感监测[J].遥感学报,2004(11):664-671.
[9] S. B. Tennakoon. V. V. N. Murty. A. Eiumnoh. Estimation of cropped area and grain yield of rice using remote sensing data [J].International Journal of Remote Sensing.1992.13(3):427-439.
[10] K. Okamoto .Estimation of rice-planted area in the tropical zone using a combination of optical and microwave satellite sensor data[J].International Journal of Remote Sensing,1999.20(5):1045-1048.
[11] Suk-Young Hong, Kyu-Sung Lee,Sang-Kyu Rim, et al. Estimation of rice field area using two-date Landsat TM images in Korea[J]. Geoscience and Remote Sensing,1999.6(2):732-734.
[12]郑长春.水稻种植面积遥感信息提取研究[D].乌鲁木齐:新疆农业大学硕士学位论文,2008:1-4.
[13]孙九林.中国农作物遥感动态监测与估产总论[M].北京:中国科学技术出版社,1996:92-95.
[14]赵锐,王延颐,戴锦芳.中国水稻遥感动态监测与估产[M].北京:中国科学技术出版社,1996:175-189.
[15]张明席,胡成群.用卫星探测资料建立水稻种植面积测算模式研究[J].气象,1992(4):11-18.
[16]程乾,王人潮.数字高程模型和多时相MODIS数据复合的水稻种植面积遥感估算方法研究[J].农业工程学报,2005.21(5):89-92.
[17]阎静,王汶,李湘阁.利用神经网络方法提取水稻种植面积-以湖北省双季早稻为例[J].遥感学报,2001.5(3):227-230.
[18]杨晓华,黄敬峰.概率神经网络的水稻种植面积遥感信息提取研究[J].浙江大学学报:农业与生命科学版,2007.33(6):691-698.
[19]杨晓华.基于神经网络和支持向量机的水稻遥感信息提取研究[D].杭州:浙江大学博士学位论文,2007:4-7.
[20]刘海,邓文胜.基于证据理论的水稻遥感信息提取方法研究[J].测绘科学,2008.33(5):21-24.
[21]周义,阮仁宗.多源信息复合的遥感数据水稻田信息提取方法研究[J].遥感信息,2009(3):30-33.
[22]张微微.基于多时相CBRES CCD图像的水稻种植面积监测[D].西南大学硕士学位论文,2008:4-5.
[23]韩立建,潘耀忠,贾斌,等.基于多时相IRS—P6卫星AWiFS数据的水稻种植面积提取方法[J].农业工程学报,2007, 23(5):137-143.
[24]张远.微波遥感水稻种植面积提取、生物量反演与稻田甲烷排放模拟[D].杭州:浙江大学博士学位论文,2009:4-9.
[25]李杨.基于环境卫星数据的水稻面积空间抽样研究[D].南京:南京林业大学硕士学位论文,2010:4-5.
[26]孙林,柳钦火,陈良富.环境与减灾小卫星高光谱成像仪陆地气溶胶光学厚度反演[J].遥感学报,2006.10(5):770-776.
[27]乐云峰,李云梅,查勇.太湖悬浮物对水体生态环境的影响及其高光谱反演[J].环境科学,2008.28(10):2148-2155.
[28]杨煜,李云梅,王桥.基于环境一号卫星高光谱遥感数据的巢湖水体叶绿素a浓度反演[J].湖泊科学,2010(4):495-503.
[29]杨贵军,黄文江,刘三超.环境减灾卫星高光谱数据大气校正模型及验证[J].北京大学学报(自然科学版),2010.46(5): 821-828.
[30]马娜,胡云峰,庄大方,等.基于最佳波段指数和J-M距离可分性的高光谱数据最佳波段组合选取研究—以环境小卫星高光谱数据在东莞市的应用为例[J].遥感技术与应用,2010.25(3):358-365.
[31]张川,基于环境减灾卫星高光谱数据的我国北方农业干旱遥感监测技术研究[D],北京:中国地质大学(北京)硕士学位论文,2010:5-6.
[32]韩瑞梅,环境星HSI数据处理关键技术的研究[D].长沙:中南大学硕士学位论文. 2010.05:7-9.
[33]童庆禧,张兵,郑兰芬.高光谱遥感-原理、技术与应用[M].北京:高等教育出版社.2006:38-46.
[34]王桥,吴传庆,厉青.环境一号卫星及其在环境监测中的应用[J].遥感学报,2010.14(1):113-121.
[35]卫星环境应用中心.数据产品的分类[EB/OL].环境保护部卫星环境应用中心,2010-04-02.
[36]卫星环境应用中心.环境卫星2级产品简介与使用说明[EB/OL].环境保护部卫星环境应用中心,2010-09-21.
[37]贾德伟,钟仕全,陈燕丽,等.HJ-1A高光谱数据预处理方法研究[J].河北遥感.2010(2):13-15.
[38]国家航天局.超光谱成像仪探测原理[EB/OL].环境与灾害监测预报小卫星,2008-08-28.
[39]互动百科.物候期[EB/OL].互动百科.
[40]陈晓玲,赵红梅,田礼乔等.环境遥感模型与应用[M].武汉:武汉大学出版社,2008:16-33.
[49]郝建亭,杨武年,李玉霞,等.基于FLAASH的多光谱数据大气校正应用研究[J].遥感信息,2008,1:78-81.
[41]支晶晶.高光谱图像条带噪声去除方法研究与应用[D].开封:河南大学硕士学位论文,2010:4-5.
[42]韩玲.关于遥感数据几何校正中纠正变换方法的探讨[J].西安地质学院学报,1997.19 (4) :86-90.
[43]孙家炳.遥感原理与应用[M].武汉:武汉大学出版社,2003:205-207.
[44]赵春晖,刘春红.超谱遥感图像降维方法研究现状与分析[J].中国空间科学技术,2004(5):28-46.
[45]姜小光,王长耀,王成.成像光谱数据的光谱信息特点及最佳波段选择-以北京顺义区为例[J].干旱区地理,2000,24(4):214-220.
[46]李行.植被高光谱遥感数据特征波段的选择方法研究[D].泰安:山东科技大学硕士学位论文,2006:9-13.
[47]章孝灿,黄智才,赵元洪.遥感数字图像处理[M].杭州:浙江大学出版社,2003:174-175.
[48] Chavez P.S.J,Berlin G.L.,Sowers L.B.. Statistical Methods for Selecting Landsat MSS Ratios[J].Journal of Applied Photographic Engineering,1982.8(1):24-40.
[49]刘春红,赵春晖,张凌雁.一种新的高光谱遥感图像降维方法[J].中国图象图形学报,2005,10(2):218-222.
[50]谷延峰,张晔.基于自动子空间划分的高光谱数据特征提取[J].遥感技术与应用,2003.18(6):384-387.
[51]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):35-45.
[52]乔蕾.基于支持向量机的高光谱图像分类[D].哈尔滨:哈尔滨工程大学硕士学位论文,2008:21-30.
[53]范明,孟小峰译.数据挖掘概念与设计[M].北京:机械工业出版社,2007:219-224.
[54]杜培军.遥感原理与应用[M].徐州:中国矿业大学出版社,2006:186-187.
[2]国家统计局农村社会经济调查司.中国农村统计年鉴[M].北京:中国统计出版社,2010:70-80.
[3]王人潮,史舟等.农业信息科学与农业信息技术[M].北京:中国农业出版社,2003:3-5.
[4] Roy F. Bartlett.Estimating the Total of a continuous population[J].Journal of Statistical Planning and Inference,1986.13:51-66.
[5]刘海启,金敏毓,龚维鹏.美国农业遥感技术应用状况概述[J].中国农业资源与区划,1999.20(2):56-60.
[6] S C Liew, S P Kam, T P Tuong et al.Application of multitemporal ERS-2 synthetic aperture radar in delineating rice cropping systems in the Mekong River Delta,Vietnam[J].GeosciencesandRemote Sensing,1998.36(5):1412-1420.
[7] P .P .Nageswara Rao.V. R. Rao. Rice crop identification and area estimation using remotely-sensed data from Indian cropping patterns[J].International Journal of Remote Sensing.1987,8(4):639-650.
[8]张峰,吴炳方.泰国水稻种植面积月变化的遥感监测[J].遥感学报,2004(11):664-671.
[9] S. B. Tennakoon. V. V. N. Murty. A. Eiumnoh. Estimation of cropped area and grain yield of rice using remote sensing data [J].International Journal of Remote Sensing.1992.13(3):427-439.
[10] K. Okamoto .Estimation of rice-planted area in the tropical zone using a combination of optical and microwave satellite sensor data[J].International Journal of Remote Sensing,1999.20(5):1045-1048.
[11] Suk-Young Hong, Kyu-Sung Lee,Sang-Kyu Rim, et al. Estimation of rice field area using two-date Landsat TM images in Korea[J]. Geoscience and Remote Sensing,1999.6(2):732-734.
[12]郑长春.水稻种植面积遥感信息提取研究[D].乌鲁木齐:新疆农业大学硕士学位论文,2008:1-4.
[13]孙九林.中国农作物遥感动态监测与估产总论[M].北京:中国科学技术出版社,1996:92-95.
[14]赵锐,王延颐,戴锦芳.中国水稻遥感动态监测与估产[M].北京:中国科学技术出版社,1996:175-189.
[15]张明席,胡成群.用卫星探测资料建立水稻种植面积测算模式研究[J].气象,1992(4):11-18.
[16]程乾,王人潮.数字高程模型和多时相MODIS数据复合的水稻种植面积遥感估算方法研究[J].农业工程学报,2005.21(5):89-92.
[17]阎静,王汶,李湘阁.利用神经网络方法提取水稻种植面积-以湖北省双季早稻为例[J].遥感学报,2001.5(3):227-230.
[18]杨晓华,黄敬峰.概率神经网络的水稻种植面积遥感信息提取研究[J].浙江大学学报:农业与生命科学版,2007.33(6):691-698.
[19]杨晓华.基于神经网络和支持向量机的水稻遥感信息提取研究[D].杭州:浙江大学博士学位论文,2007:4-7.
[20]刘海,邓文胜.基于证据理论的水稻遥感信息提取方法研究[J].测绘科学,2008.33(5):21-24.
[21]周义,阮仁宗.多源信息复合的遥感数据水稻田信息提取方法研究[J].遥感信息,2009(3):30-33.
[22]张微微.基于多时相CBRES CCD图像的水稻种植面积监测[D].西南大学硕士学位论文,2008:4-5.
[23]韩立建,潘耀忠,贾斌,等.基于多时相IRS—P6卫星AWiFS数据的水稻种植面积提取方法[J].农业工程学报,2007, 23(5):137-143.
[24]张远.微波遥感水稻种植面积提取、生物量反演与稻田甲烷排放模拟[D].杭州:浙江大学博士学位论文,2009:4-9.
[25]李杨.基于环境卫星数据的水稻面积空间抽样研究[D].南京:南京林业大学硕士学位论文,2010:4-5.
[26]孙林,柳钦火,陈良富.环境与减灾小卫星高光谱成像仪陆地气溶胶光学厚度反演[J].遥感学报,2006.10(5):770-776.
[27]乐云峰,李云梅,查勇.太湖悬浮物对水体生态环境的影响及其高光谱反演[J].环境科学,2008.28(10):2148-2155.
[28]杨煜,李云梅,王桥.基于环境一号卫星高光谱遥感数据的巢湖水体叶绿素a浓度反演[J].湖泊科学,2010(4):495-503.
[29]杨贵军,黄文江,刘三超.环境减灾卫星高光谱数据大气校正模型及验证[J].北京大学学报(自然科学版),2010.46(5): 821-828.
[30]马娜,胡云峰,庄大方,等.基于最佳波段指数和J-M距离可分性的高光谱数据最佳波段组合选取研究—以环境小卫星高光谱数据在东莞市的应用为例[J].遥感技术与应用,2010.25(3):358-365.
[31]张川,基于环境减灾卫星高光谱数据的我国北方农业干旱遥感监测技术研究[D],北京:中国地质大学(北京)硕士学位论文,2010:5-6.
[32]韩瑞梅,环境星HSI数据处理关键技术的研究[D].长沙:中南大学硕士学位论文. 2010.05:7-9.
[33]童庆禧,张兵,郑兰芬.高光谱遥感-原理、技术与应用[M].北京:高等教育出版社.2006:38-46.
[34]王桥,吴传庆,厉青.环境一号卫星及其在环境监测中的应用[J].遥感学报,2010.14(1):113-121.
[35]卫星环境应用中心.数据产品的分类[EB/OL].环境保护部卫星环境应用中心,2010-04-02.
[36]卫星环境应用中心.环境卫星2级产品简介与使用说明[EB/OL].环境保护部卫星环境应用中心,2010-09-21.
[37]贾德伟,钟仕全,陈燕丽,等.HJ-1A高光谱数据预处理方法研究[J].河北遥感.2010(2):13-15.
[38]国家航天局.超光谱成像仪探测原理[EB/OL].环境与灾害监测预报小卫星,2008-08-28.
[39]互动百科.物候期[EB/OL].互动百科.
[40]陈晓玲,赵红梅,田礼乔等.环境遥感模型与应用[M].武汉:武汉大学出版社,2008:16-33.
[49]郝建亭,杨武年,李玉霞,等.基于FLAASH的多光谱数据大气校正应用研究[J].遥感信息,2008,1:78-81.
[41]支晶晶.高光谱图像条带噪声去除方法研究与应用[D].开封:河南大学硕士学位论文,2010:4-5.
[42]韩玲.关于遥感数据几何校正中纠正变换方法的探讨[J].西安地质学院学报,1997.19 (4) :86-90.
[43]孙家炳.遥感原理与应用[M].武汉:武汉大学出版社,2003:205-207.
[44]赵春晖,刘春红.超谱遥感图像降维方法研究现状与分析[J].中国空间科学技术,2004(5):28-46.
[45]姜小光,王长耀,王成.成像光谱数据的光谱信息特点及最佳波段选择-以北京顺义区为例[J].干旱区地理,2000,24(4):214-220.
[46]李行.植被高光谱遥感数据特征波段的选择方法研究[D].泰安:山东科技大学硕士学位论文,2006:9-13.
[47]章孝灿,黄智才,赵元洪.遥感数字图像处理[M].杭州:浙江大学出版社,2003:174-175.
[48] Chavez P.S.J,Berlin G.L.,Sowers L.B.. Statistical Methods for Selecting Landsat MSS Ratios[J].Journal of Applied Photographic Engineering,1982.8(1):24-40.
[49]刘春红,赵春晖,张凌雁.一种新的高光谱遥感图像降维方法[J].中国图象图形学报,2005,10(2):218-222.
[50]谷延峰,张晔.基于自动子空间划分的高光谱数据特征提取[J].遥感技术与应用,2003.18(6):384-387.
[51]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):35-45.
[52]乔蕾.基于支持向量机的高光谱图像分类[D].哈尔滨:哈尔滨工程大学硕士学位论文,2008:21-30.
[53]范明,孟小峰译.数据挖掘概念与设计[M].北京:机械工业出版社,2007:219-224.
[54]杜培军.遥感原理与应用[M].徐州:中国矿业大学出版社,2006:186-187.