基于高光谱的抽穗期寒地水稻叶片氮素预测模型
|
摘要
为快速、无损地监测水稻叶片氮素营养状况,开展了基于高光谱成像技术的抽穗期寒地水稻叶片氮素预测模型的研究。以不同施氮水平的寒地水稻叶片为研究对象,采用连续投影算法(successive projections algorithm,SPA)和分段主成分分析(segmented principal components analysis,SPCA)方法选择水稻叶片的高光谱特征波段,SPCA方法降维后结合相关分析(correlation analysis,CA)构建特征光谱参量,并建立基于全波段高光谱数据、SPA特征波段及SPCA特征光谱参量的多种回归分析模型且对模型进行检验和筛选。研究结果表明:在校正集决定系数RC2上,基于多元逐步回归分析(multiple stepwise regression analysis,MSRA)的全波段模型较好,RC2=0.9 6 4,校正集均方根误差RMSEC=0.083;RP2为0.961,RMSEP为0.050。该研究结果为快速检测水稻叶片氮素含量及水稻生长期间精确施肥管理提供了技术支撑和理论依据。
In order to realize the dynamic,non-destructive diagnosis of rice nitrogen nutritional status,we use hyperspectral imaging techniques as an approach for nitrogen content prediction of rice leaves in cold region.The experiments were carried out for two years(2014 and 2015) at Fangzheng country,Heilongjiang province,China.Longdao 23 was chosen as the test cultivar.6 nitrogen fertilization rates were applied in our experiments,i.e.,N0(0 kg/hm~2),N1(60 kg/hm~2),N2(90 kg/hm~2),N3(120 kg/hm~2),N4(150 kg/hm~2),and N5(180 kg/hm~2).The hyperspectral reflectance and nitrogen content of rice leaves under different nitrogen levels at heading stage were separately measured using American Headwall imaging spectrometer and German AA3 analyzer.Several regression analysis(RA) estimate models have been built based on different characteristic spectral parameters using different algorithms which include successive projections algorithm(SPA) and segmented principal components analysis(SPCA) combined with correlation analysis(CA) for testing and screening.The first method,a nitrogen content value estimation model based on multiple stepwise regression analysis(MSRA) in the whole wavelength region of 400 ~ 1000 nm has been built and been predicted.We predicted all the estimate models so that testing its accuracy and stability.The results indicated that,reflectance of rice leaves decreases in the visible region,and increases in the near infrared region,with the raise of nitrogen level.From the calibration performance index,the whole wavelength model based on MSRA is the best with a coefficient of determination(RC2= 0.0964) and root mean square error(MRSEC) of 0.083 and a coefficient of determination(RP2=0.961) and the root mean square error(RMSEP) of 0.050.The study provides technical support and theoretical basis for the rapid detection of nitrogen content in rice leaves and the precise fertilization management during rice growth.
In order to realize the dynamic,non-destructive diagnosis of rice nitrogen nutritional status,we use hyperspectral imaging techniques as an approach for nitrogen content prediction of rice leaves in cold region.The experiments were carried out for two years(2014 and 2015) at Fangzheng country,Heilongjiang province,China.Longdao 23 was chosen as the test cultivar.6 nitrogen fertilization rates were applied in our experiments,i.e.,N0(0 kg/hm~2),N1(60 kg/hm~2),N2(90 kg/hm~2),N3(120 kg/hm~2),N4(150 kg/hm~2),and N5(180 kg/hm~2).The hyperspectral reflectance and nitrogen content of rice leaves under different nitrogen levels at heading stage were separately measured using American Headwall imaging spectrometer and German AA3 analyzer.Several regression analysis(RA) estimate models have been built based on different characteristic spectral parameters using different algorithms which include successive projections algorithm(SPA) and segmented principal components analysis(SPCA) combined with correlation analysis(CA) for testing and screening.The first method,a nitrogen content value estimation model based on multiple stepwise regression analysis(MSRA) in the whole wavelength region of 400 ~ 1000 nm has been built and been predicted.We predicted all the estimate models so that testing its accuracy and stability.The results indicated that,reflectance of rice leaves decreases in the visible region,and increases in the near infrared region,with the raise of nitrogen level.From the calibration performance index,the whole wavelength model based on MSRA is the best with a coefficient of determination(RC2= 0.0964) and root mean square error(MRSEC) of 0.083 and a coefficient of determination(RP2=0.961) and the root mean square error(RMSEP) of 0.050.The study provides technical support and theoretical basis for the rapid detection of nitrogen content in rice leaves and the precise fertilization management during rice growth.
引文
[1]孙曦.作物营养和施肥[M].北京:中国农业出版社,1990.
[2]王树文,郑博元,张长利.氮素胁迫下水稻高光谱特征研究[J].农机化研究,2015,37(8):162-165.
[3]彭显龙,王海含,刘小纶,等.稻田施尿素后短期内氮素分布和氨挥发损失研究[J].东北农业大学学报,2015,46(12):24-32.
[4]Du Lin,Gong Wei,Shi Shuo,et al.Estimation of rice leaf nitrogen contents based on hyperspectral LIDAR[J].International Journal of Applied Earth Observation and Geoinformation,2016,44:136-143.
[5]王树文,赵珊,张长利,等.基于成像光谱技术的寒地玉米苗期冠层氮含量预测模型[J].农业工程学报,2016,32(13):149-154.
[6]贺婷,李建东,刘桂鹏,等.基于高光谱遥感的玉米全氮含量估测模型[J].沈阳农业大学学报,2016,47(3):257-265.
[7]唐菁,杨承栋,康红梅,等.植物营养诊断方法研究进展[J].世界林业研究,2005,18(6):45-48.
[8]裘正军,宋海燕,何勇,等.应用SPAD和光谱技术研究油菜生长期间的氮素变化规律[J].农业工程学报,2007,23(7):150-154.
[9]张筱蕾,刘飞,聂鹏程,等.高光谱成像技术的油菜叶片氮含量及分布快速检测[J].光谱学与光谱分析,2014,34(9):2513-2518.
[10]王仁红,宋晓宇,李振海,等.基于高光谱的冬小麦氮素营养指数估测[J].农业工程学报,2014,30(19):191-198.
[11]Liu Haiying,Zhu Hongchun.Hyperspectral characteristic analysis for leaf nitrogen content in different growth stages of winter wheat[J].Applied Optics,2016,55(34):151-161.
[12]Song Xiao,Xu Duanyang,He Li,et al.Using multi-angle hyperspectral data to monitor canopy leaf nitrogen content of wheat[J].Precision Agriculture,2016,17(6):721-736.
[13]徐新刚,赵春江,王纪华,等.基于可见光-近红外新光谱特征和最优组合原理的大麦叶片氮含量监测[J].红外与毫米波学报,2013,32(4):351-358.
[14]Fitzgerald G,Rodriguez D,O’Leary G.Measuring and predicting canopy nitrogen nutrition in wheat using a spectral index:The canopy chlorophyll content index(CCCI)[J].Field Crops Research,2010,116(3):318-324.
[15]Jin Liang,Hu Kelin,Tian Mingming,et al.Diagnosis of nitrogen content in upper and lower corn leaves Bbased on hyperspectral data[J].Spectroscopy and Spectral Analysis,2013,33(4):1032-1037.
[16]王巧男,叶旭君,李金梦,等.基于双波段植被指数(TBVI)的柑橘冠层含氮量预测及可视化研究[J].光谱学与光谱分析,2015,35(3):715-718.
[17]唐延林,王人潮,黄敬峰,等.不同供氮水平下水稻高光谱及其红边特征研究[J].遥感学报,2004,8(2):185-192.
[18]Tian Yongchao,Gu Kaijian,Chu Xu,et al.Comparison of different hyperspectral vegetation indices for canopy leaf nitrogen concentration estimation in rice[J].Plant and Soil,2014,376(1-2):193-209.
[19]顾清,邓劲松,陆超,等.基于光谱和形状特征的水稻扫描叶片氮素营养诊断[J].农业机械学报,2012,43(8):170-174.
[20]Onoyama H,Ryu C,Suguri M,et al.Integrate growing temperature to estimate the nitrogen content of rice plants at the heading stage using hyperspectral imagery[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2506-2515.
[21]Chu Xu,Guo Yongjiu,He Jiaoyang,et al.Comparison of different hyperspectral vegetation indices for estimating canopy leaf nitrogen accumulation in rice[J].Agronomy Journal,2014,106(5):1911-1920.
[22]Du Lin,Gong Wei,Shi Shuo,et al.Estimation of rice leaf nitrogen contents based on hyperspectral LIDAR[J].International Journal of Applied Earth Observation and Geoinformation,2016,44:136-143.
[23]王树文,赵越,王丽凤,等.基于高光谱的寒地水稻叶片氮素含量预测[J].农业工程学报,2016,32(20):187-194.
[2]王树文,郑博元,张长利.氮素胁迫下水稻高光谱特征研究[J].农机化研究,2015,37(8):162-165.
[3]彭显龙,王海含,刘小纶,等.稻田施尿素后短期内氮素分布和氨挥发损失研究[J].东北农业大学学报,2015,46(12):24-32.
[4]Du Lin,Gong Wei,Shi Shuo,et al.Estimation of rice leaf nitrogen contents based on hyperspectral LIDAR[J].International Journal of Applied Earth Observation and Geoinformation,2016,44:136-143.
[5]王树文,赵珊,张长利,等.基于成像光谱技术的寒地玉米苗期冠层氮含量预测模型[J].农业工程学报,2016,32(13):149-154.
[6]贺婷,李建东,刘桂鹏,等.基于高光谱遥感的玉米全氮含量估测模型[J].沈阳农业大学学报,2016,47(3):257-265.
[7]唐菁,杨承栋,康红梅,等.植物营养诊断方法研究进展[J].世界林业研究,2005,18(6):45-48.
[8]裘正军,宋海燕,何勇,等.应用SPAD和光谱技术研究油菜生长期间的氮素变化规律[J].农业工程学报,2007,23(7):150-154.
[9]张筱蕾,刘飞,聂鹏程,等.高光谱成像技术的油菜叶片氮含量及分布快速检测[J].光谱学与光谱分析,2014,34(9):2513-2518.
[10]王仁红,宋晓宇,李振海,等.基于高光谱的冬小麦氮素营养指数估测[J].农业工程学报,2014,30(19):191-198.
[11]Liu Haiying,Zhu Hongchun.Hyperspectral characteristic analysis for leaf nitrogen content in different growth stages of winter wheat[J].Applied Optics,2016,55(34):151-161.
[12]Song Xiao,Xu Duanyang,He Li,et al.Using multi-angle hyperspectral data to monitor canopy leaf nitrogen content of wheat[J].Precision Agriculture,2016,17(6):721-736.
[13]徐新刚,赵春江,王纪华,等.基于可见光-近红外新光谱特征和最优组合原理的大麦叶片氮含量监测[J].红外与毫米波学报,2013,32(4):351-358.
[14]Fitzgerald G,Rodriguez D,O’Leary G.Measuring and predicting canopy nitrogen nutrition in wheat using a spectral index:The canopy chlorophyll content index(CCCI)[J].Field Crops Research,2010,116(3):318-324.
[15]Jin Liang,Hu Kelin,Tian Mingming,et al.Diagnosis of nitrogen content in upper and lower corn leaves Bbased on hyperspectral data[J].Spectroscopy and Spectral Analysis,2013,33(4):1032-1037.
[16]王巧男,叶旭君,李金梦,等.基于双波段植被指数(TBVI)的柑橘冠层含氮量预测及可视化研究[J].光谱学与光谱分析,2015,35(3):715-718.
[17]唐延林,王人潮,黄敬峰,等.不同供氮水平下水稻高光谱及其红边特征研究[J].遥感学报,2004,8(2):185-192.
[18]Tian Yongchao,Gu Kaijian,Chu Xu,et al.Comparison of different hyperspectral vegetation indices for canopy leaf nitrogen concentration estimation in rice[J].Plant and Soil,2014,376(1-2):193-209.
[19]顾清,邓劲松,陆超,等.基于光谱和形状特征的水稻扫描叶片氮素营养诊断[J].农业机械学报,2012,43(8):170-174.
[20]Onoyama H,Ryu C,Suguri M,et al.Integrate growing temperature to estimate the nitrogen content of rice plants at the heading stage using hyperspectral imagery[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2506-2515.
[21]Chu Xu,Guo Yongjiu,He Jiaoyang,et al.Comparison of different hyperspectral vegetation indices for estimating canopy leaf nitrogen accumulation in rice[J].Agronomy Journal,2014,106(5):1911-1920.
[22]Du Lin,Gong Wei,Shi Shuo,et al.Estimation of rice leaf nitrogen contents based on hyperspectral LIDAR[J].International Journal of Applied Earth Observation and Geoinformation,2016,44:136-143.
[23]王树文,赵越,王丽凤,等.基于高光谱的寒地水稻叶片氮素含量预测[J].农业工程学报,2016,32(20):187-194.