Field Remote Sensing Model Research In Cotton Yield Based On LAI

Abstract:Through collecting different growing seasons data of field hyperspectral and LAI, and cotton yield in the different variety, waiter, filter, denty measurement, the paper nanalysises that hyperspectral response for cotton canopy in the different LAI and that the relationship between cotton yield and LAI, and establishes the remote sensing model of cotton yield.The results are followed:1) The correlation between LAI and the spectrum is significant negative in the visional wave, and the correlation between LAI and the near infrared is significant positive; there are three high plaint regions, including the blue region (523-531nm), the yellow region (570-576nm) and the red region (700-755nm) respectively, result from the correlation between LAI and the derivative spectrum, and among them the red region is the widest and the correlation is the strongest.2) The red edge has the "double peaks" phenomenon, and the double peaks lie in 718nm and 723nm, and the valley between the double peaks lies in 720nm. The study indicates that the fore peak is higher when the leaf area index is low, and that the late peak is higher when the leaf area index is high.3) The achievement is improved throughout establishing the subsection models, and SDr is used when the leaf index is under 1.0 and SDr/SDb is used when the leaf area index is above 1.0. The checked result indicated that the relative error is under 20% and the RMSE is under 0.35.4) the dynamic variation law of LAI in the different yield of cotton in the whole growing season is analyzed in the project. The results show that the relationship of the yield and The LAI of cotton is significant from flowering-boll stage to boll opening stage, and that exceptionally the full-boll stage is the best time for the cotton yield estimation. The research establishes the model of the cotton yield estimation through connecting the hyperspectral LAI estimation model with the cotton yield Agricultural estimation model , and the relative error is about 5%, and RMSE is about 100kg/hm2, and the determination coefficient of correlation of the estimated value and the measured value of cotton yield arrive at 0.6224**.5)The L-Y model could accomplish the qualitative estimation on the cotton yield with relationship between the leaf area index and the cotton yield and the multi-time LAI obtained through the model established with the field hyperspectral parameters . The estimation relative error is about 5.44%, and RMSE is 116.2 kg-hm-2 , and coefficient of relationship between measured value and estimation value is 0.836 arriving the extreme significant level of 0.99.The research achieved the fllowing and the creavemeant.1) The new hyerspectral parameters could achieve the better LAI quatatative estimation through analysesing the hyperspectrum responsive for the different LAI cotton canopy. The first different value of "three edge" wavelength of high relationship is better of the first different whole value. D723 is better than Dr, and so Dr is not the best of the red edge parameters.2) The relationship is between LAI and cotton yield in stage, and the best time of estimating the...