The unavoidable nature of Ulva prolifera mixed pixel in low-resolution remote sensing images would result in rough boundary of U.prolifera patches,omission of tiny patches,and overestimation of coverage area.The decom...The unavoidable nature of Ulva prolifera mixed pixel in low-resolution remote sensing images would result in rough boundary of U.prolifera patches,omission of tiny patches,and overestimation of coverage area.The decomposition of U.prolifera mixed pixel addresses the issue of coverage area overestimation,and the remaining problems can be alleviated by subpixel mapping(SPM).Due to the drift and dissipation of U.prolifera,a suitable SPM method is the single image-based unsupervised method.However,the method has difficulties in detail reconstruction,insufficient learning of spectral information,and SPM error introduced by abundance deviation.Therefore,we proposed a multiple-feature decision fusion SPM(MFDFSPM)method.It involves three branches to obtain the spatial,abundance,and spectral features of U.prolifera while considers multi-feature information using the fusion strategy.Experiments on the Geostationary Ocean Color Imager images in the Yellow Sea of China indicate that the MFDFSPM overperforms several typical U.prolifera SPM methods in higher accuracy and stronger robustness in both SPM and abundance calculation,which produced subpixel map with more detailed spatial information and less noise.展开更多
MODIS time-series imagery is promising for generating regional and global land cover products.For Brazil,however,accurate fractional cropland covers(FCC)information is difficult to obtain due to frequent cloud coverag...MODIS time-series imagery is promising for generating regional and global land cover products.For Brazil,however,accurate fractional cropland covers(FCC)information is difficult to obtain due to frequent cloud coverage and the mixing-pixel problem.To address these problems,this study developed an innovative approach to mapping the FCC of the Mato Grosso State,Brazil through integrating Linear Spectral Mixture Analysis(LSMA)and Seasonal Dynamic Index(SDI)models.With MOD13Q1 time-series EVI imagery,a SDI was developed to represent the phenology of croplands.Furthermore,fractional land covers(e.g.,vegetation,soil,and low albedo components)were derived with the LSMA algorithms.A stepwise regression model was established to estimate the FCC at the regional scale.Finally,ground truth cropland cover information was extracted from Landsat TM imagery using a hybrid method.Results indicated that the combination of multiple feature variables produced better results when compared with individual variables.Through cross-validation and comparative analysis,the coefficient of determination(R^(2))between the reference and estimated FCCs reached 0.84 with a Root Mean Square Error(RMSE)of 0.13.This indicates that the proposed method effectively improved the accuracy of fractional cropland mapping.When compared to the traditional per-pixel“hard”classification,the sub-pixel level maps illustrated detailed cropland spatial distribution patterns.展开更多
基金Supported by the Shandong Provincial Natural Science Foundation of China(No.ZR2019MD023)the National Natural Science Foundation of China(No.41776182)。
文摘The unavoidable nature of Ulva prolifera mixed pixel in low-resolution remote sensing images would result in rough boundary of U.prolifera patches,omission of tiny patches,and overestimation of coverage area.The decomposition of U.prolifera mixed pixel addresses the issue of coverage area overestimation,and the remaining problems can be alleviated by subpixel mapping(SPM).Due to the drift and dissipation of U.prolifera,a suitable SPM method is the single image-based unsupervised method.However,the method has difficulties in detail reconstruction,insufficient learning of spectral information,and SPM error introduced by abundance deviation.Therefore,we proposed a multiple-feature decision fusion SPM(MFDFSPM)method.It involves three branches to obtain the spatial,abundance,and spectral features of U.prolifera while considers multi-feature information using the fusion strategy.Experiments on the Geostationary Ocean Color Imager images in the Yellow Sea of China indicate that the MFDFSPM overperforms several typical U.prolifera SPM methods in higher accuracy and stronger robustness in both SPM and abundance calculation,which produced subpixel map with more detailed spatial information and less noise.
基金The study is funded by the Strategic Priority Research Program of Chinese Academy of Sciences(Grant No.XDA20020101)the National R&D Program of China(Granted No.2017YFB0504201)the Natural Science Foundation of China(Grant No.61473286 and 41201460).
文摘MODIS time-series imagery is promising for generating regional and global land cover products.For Brazil,however,accurate fractional cropland covers(FCC)information is difficult to obtain due to frequent cloud coverage and the mixing-pixel problem.To address these problems,this study developed an innovative approach to mapping the FCC of the Mato Grosso State,Brazil through integrating Linear Spectral Mixture Analysis(LSMA)and Seasonal Dynamic Index(SDI)models.With MOD13Q1 time-series EVI imagery,a SDI was developed to represent the phenology of croplands.Furthermore,fractional land covers(e.g.,vegetation,soil,and low albedo components)were derived with the LSMA algorithms.A stepwise regression model was established to estimate the FCC at the regional scale.Finally,ground truth cropland cover information was extracted from Landsat TM imagery using a hybrid method.Results indicated that the combination of multiple feature variables produced better results when compared with individual variables.Through cross-validation and comparative analysis,the coefficient of determination(R^(2))between the reference and estimated FCCs reached 0.84 with a Root Mean Square Error(RMSE)of 0.13.This indicates that the proposed method effectively improved the accuracy of fractional cropland mapping.When compared to the traditional per-pixel“hard”classification,the sub-pixel level maps illustrated detailed cropland spatial distribution patterns.
