Study On Spring Maize Area Extraction And Growth Condition Based On Remote Sensing Phenological Parameter

Obtaining planted area information of spring maize quickly and accurately is of great practical significance to national food security and the development of modern information agriculture.Remote sensing technology has played a great role in the identification of crops in large areas,and the combination of phenological and spectral information has unique advantages in remote sensing crop recognition,but it is still in less utilization.In the traditional large-scale crop growth monitoring,vegetation index is often used as a monitoring index,so the monitoring results contain the information about the difference of the crop phenophase itself and the validity of the results is greatly disturbed.Kangping County,Faku County,Xinmin City was selected as the research area,MODIS-NDVI time series images with spatial resolution of 250 m and 46 scales from 2014 to 2015 were used as the main data sources.Based on the Savitzky-Golay filtering method,MODIS-NDVI time series data were reconstructed by smoothing its noise,phenological parameters of crops in growth season and NDVI curve change characteristics of other major objects were extracted from the reconstructed data by TIMESAT,then NDWI and LSWI data during rice transplanting period and the near infrared reflectance data during soybean grain filling period in 2015 were jointly used to construct the decision tree,correspondingly,planting area of spring maize were then extracted by the decision tree.For the problem of mixed pixels,the spring maize endmember spectrum was extracted based on MOD09A1 reflectance image,and the mixed-pixel decomposition based on the linear spectral mixture model was used to obtain the spring naize abundance.Finally,the spring rnaize planting area was accurately extracted according to the classification result of the decision tree and the abundance ratio of spring maize.On the basis of accurate acquisition of spring maize acreage,using reconstructed MODIS-NDVI time series data by Savitzky-Golay filtering method from 2008 to 2015,crop phenological parameters including NDVI maximum,growth season length,growth rate and NDVI active accumulation in growth season are extracted,the relationship between phenological parameters and yield was verified by using Pearson correlation analysis method,and the feasibility of estimating the yield as a monitoring index was analyzed.The phenological parameters were used as the monitoring indicators,the mean values of them were obtained from 2008 to 2014,the spring maize growth condition was graded and analyzed by the difference interannual variation crop growth model.The statistical data was also used to evaluate the spring maize area accuracy respectively by the decision tree method and the decision tree combined mixed-pixel decomposition method,the results showed that the precision was respectively 85.591%and 90.570%;Using the sampling sites to evaluate the error matrix accuracy,the results showed that the overall classification accuracy is 89.031%and the Kappa coefficient is 0.814 by the decision tree classification method:the OLI supervised classification image was used to evaluate the accuracy of the spring maize area,the results showed that the precision of spring maize planting area was 80%.As for phenological parameters,there is a positive correlation between the regional average NDVI active accumulation value?NDVI maximum in growth season and the yield per unit area,And the correlation coefficient was above 0.6,the growth rate in growth season is negatively correlated with the yield per unit area and correlation coefficient was above 0.4,they can well Indicates yield changes as growth condition monitoring indicators.The result indicates that using high-resolution remote sensing time series image data extracts phenological parameters and extracting crop acreage quickly and accurately is feasible,the mixed-pixel decomposition method can further improves the area extraction accuracy,and this will contributes to promote the development of information-based precision agriculture.In the new phenological parameters used in this paper crop growth monitoring indicators,regional average NDVI active accumulation value and NDVI maximum can well indicate the crop growth,and has the practical significance for indicating the change of spring maize yield.