水稻叶片含水量的高光谱遥感估算
|
摘要
针对高光谱估算不同品种水稻叶片含水量模型精度较低和参数复杂的问题,采集了20个品种、4个关键生长期(拔节-孕穗、孕穗-抽穗、抽穗-灌浆、灌浆-成熟)的水稻叶片高光谱和含水量数据,利用15种常见的植被指数反演水稻叶片含水量,建模效果均不够理想;利用叶片含水量敏感波段的反射率、光谱一阶导数构建归一化植被指数、比值植被指数和差值植被指数进一步探究,结果表明,利用一阶导数构建的差值植被指数DDV(R1 833,R2 236)建模精度和预测效果最佳,建模决定系数为0.72,验证决定系数为0.81,相对分析误差为2.29,可以有效估计不同品种水稻4个生长期的叶片含水量。
Hyperspectral estimation models for leaf water content of different varieties of rice had low precision and complicated parameters.This article collected hyperspectral data and water content data of rice leaves which involved twenty varieties and four key growth periods which are jointing-booting,booting-heading,heading-filling,and filling-maturity,used fifteen kinds of common vegetation indexes inversing water content of rice leaf,but the models were not ideal.It used reflectance of sensitive band of leave water content and first derivative of spectrum to construct normalized difference vegetation index,ratio vegetation index and difference vegetation index.By making a further inquiry,the results showed that accuracy and prediction result of difference vegetation index DDV(R1 833,R2 236)constructed by first derivative of spectrum was the best,with the model decision coefficient 0.72,the validation decision coefficient 0.81,and the relative analysis error 2.29,which can effectively estimate the forecast water content of rice leaf in four growth periods.
Hyperspectral estimation models for leaf water content of different varieties of rice had low precision and complicated parameters.This article collected hyperspectral data and water content data of rice leaves which involved twenty varieties and four key growth periods which are jointing-booting,booting-heading,heading-filling,and filling-maturity,used fifteen kinds of common vegetation indexes inversing water content of rice leaf,but the models were not ideal.It used reflectance of sensitive band of leave water content and first derivative of spectrum to construct normalized difference vegetation index,ratio vegetation index and difference vegetation index.By making a further inquiry,the results showed that accuracy and prediction result of difference vegetation index DDV(R1 833,R2 236)constructed by first derivative of spectrum was the best,with the model decision coefficient 0.72,the validation decision coefficient 0.81,and the relative analysis error 2.29,which can effectively estimate the forecast water content of rice leaf in four growth periods.
引文
[1]黄敬峰,杨忠恩,王人潮,等.基于GIS的水稻遥感估产模型研究[J].遥感技术与应用,2002,17(3):125-128.
[2]宫兆宁,林川,赵文吉,等.WorldView-2影像的湿地典型挺水植物群落含水量估算研究:以北京野鸭湖湿地为例[J].红外与毫米波学报,2014,33(5):533-545.
[3]孙俊,毛罕平,羊一清,等.基于冠层光谱特性的水稻叶片含水率模型[J].农业工程学报,2009,25(9):133-136.
[4]陈海波,李就好.基于光谱反射率的作物水分状况研究进展[J].节水灌溉,2010(8):69-72.
[5]林川,宫兆宁,赵文吉.叶冠尺度野鸭湖湿地植物群落含水量的高光谱估算模型[J].生态学报,2011,31(22):6645-6658.
[6] THOMAS J R,NAMKEN L N,OERTHER G F,et al.Estimating leaf water content by reflectance measurement[J].Agronomy Journal,1971(63):845-847.
[7] CARTER G A.Primary and secondary effects of water content on the spectral reflectance of leaves[J].American Journal of Botany,1991,78(7):919-924.
[8]王纪华,赵春江,郭晓维,等.用光谱反射率诊断小麦叶片水分状况的研究[J].中国农业科学,2001,34(1):104-107.
[9]付彦博,范燕敏,盛建东,等.紫花苜蓿冠层反射光谱与叶片含水率关系研究[J].光谱学与光谱分析,2013,33(3):766-769.
[10]刘小军,田永超,姚霞,等.基于高光谱的水稻叶片含水量监测研究[J].中国农业科学,2012,45(3):435-442.
[11]田永超,曹卫星,姜东,等.不同水氮条件下水稻冠层反射光谱与植株含水率的定量关系[J].植物生态学报,2005,29(2):318-313.
[12]梁亮,张连蓬,林卉,等.基于导数光谱的小麦冠层叶片含水量反演[J].中国农业科学,2013,46(1):18-29.
[13]刘小珊.土壤二向反射特性研究与标准光谱库的应用[D].武汉:华中农业大学,2015.
[14]赵钊,李霞,尹业彪,等.荒漠植物含水量的光谱特征分析[J].光谱学与光谱分析,2010,39(9):2500-2503.
[15]沈艳,牛铮,王坟,等.基于导数光谱变量叶片含水量模型的建立[J].地理与地理信息科学,2005,21(4):16-19.
[16]蒋金豹,黄文江,陈云浩.用冠层光谱比值指数反演条锈病胁迫下的小麦含水量[J].光谱学与光谱分析,2010,30(7):1939-1943.
[17]金林雪,李映雪,徐德福,等.小麦叶片水分及绿度特征的光谱法诊断[J].中国农业气象,2012,33(1):124-128.
[18]潘佩芬,杨武年,简季,等.基于光谱指数的植被含水率遥感反演模型研究:以岷江上游毛尔盖地区为例[J].遥感信息,2013,28(3):69-73.
[19]苏毅,王克如,李少昆,等.棉花植株水分含量的高光谱监测模型研究[J].棉花学报,2010,22(6):554-560.
[20]陈小平,王树东,张立福,等.植被叶片含水量反演的精度及敏感性[J].遥感信息,2016,31(1):48-57.
[21]金秀良,徐新刚,王纪华,等.基于灰度关联分析的冬小麦叶片含水量高光谱估测[J].光谱学与光谱分析,2012,32(11):3103-3106.
[22]孙林,程丽娟.植被叶片生化组分的光谱响应特征分析[J].光谱学与光谱分析,2010,30(11):3031-3035.
[2]宫兆宁,林川,赵文吉,等.WorldView-2影像的湿地典型挺水植物群落含水量估算研究:以北京野鸭湖湿地为例[J].红外与毫米波学报,2014,33(5):533-545.
[3]孙俊,毛罕平,羊一清,等.基于冠层光谱特性的水稻叶片含水率模型[J].农业工程学报,2009,25(9):133-136.
[4]陈海波,李就好.基于光谱反射率的作物水分状况研究进展[J].节水灌溉,2010(8):69-72.
[5]林川,宫兆宁,赵文吉.叶冠尺度野鸭湖湿地植物群落含水量的高光谱估算模型[J].生态学报,2011,31(22):6645-6658.
[6] THOMAS J R,NAMKEN L N,OERTHER G F,et al.Estimating leaf water content by reflectance measurement[J].Agronomy Journal,1971(63):845-847.
[7] CARTER G A.Primary and secondary effects of water content on the spectral reflectance of leaves[J].American Journal of Botany,1991,78(7):919-924.
[8]王纪华,赵春江,郭晓维,等.用光谱反射率诊断小麦叶片水分状况的研究[J].中国农业科学,2001,34(1):104-107.
[9]付彦博,范燕敏,盛建东,等.紫花苜蓿冠层反射光谱与叶片含水率关系研究[J].光谱学与光谱分析,2013,33(3):766-769.
[10]刘小军,田永超,姚霞,等.基于高光谱的水稻叶片含水量监测研究[J].中国农业科学,2012,45(3):435-442.
[11]田永超,曹卫星,姜东,等.不同水氮条件下水稻冠层反射光谱与植株含水率的定量关系[J].植物生态学报,2005,29(2):318-313.
[12]梁亮,张连蓬,林卉,等.基于导数光谱的小麦冠层叶片含水量反演[J].中国农业科学,2013,46(1):18-29.
[13]刘小珊.土壤二向反射特性研究与标准光谱库的应用[D].武汉:华中农业大学,2015.
[14]赵钊,李霞,尹业彪,等.荒漠植物含水量的光谱特征分析[J].光谱学与光谱分析,2010,39(9):2500-2503.
[15]沈艳,牛铮,王坟,等.基于导数光谱变量叶片含水量模型的建立[J].地理与地理信息科学,2005,21(4):16-19.
[16]蒋金豹,黄文江,陈云浩.用冠层光谱比值指数反演条锈病胁迫下的小麦含水量[J].光谱学与光谱分析,2010,30(7):1939-1943.
[17]金林雪,李映雪,徐德福,等.小麦叶片水分及绿度特征的光谱法诊断[J].中国农业气象,2012,33(1):124-128.
[18]潘佩芬,杨武年,简季,等.基于光谱指数的植被含水率遥感反演模型研究:以岷江上游毛尔盖地区为例[J].遥感信息,2013,28(3):69-73.
[19]苏毅,王克如,李少昆,等.棉花植株水分含量的高光谱监测模型研究[J].棉花学报,2010,22(6):554-560.
[20]陈小平,王树东,张立福,等.植被叶片含水量反演的精度及敏感性[J].遥感信息,2016,31(1):48-57.
[21]金秀良,徐新刚,王纪华,等.基于灰度关联分析的冬小麦叶片含水量高光谱估测[J].光谱学与光谱分析,2012,32(11):3103-3106.
[22]孙林,程丽娟.植被叶片生化组分的光谱响应特征分析[J].光谱学与光谱分析,2010,30(11):3031-3035.