Abstract: A method for extracting a black-odorous water body based on a CART classification model includes: selecting a research region, designing sampling points within the region; monitoring relevant chemical indicators of the water body at various sampling points, extracting remote sensing reflectance data of the water body, determining a type of the water body according to a classification standard of relevant chemical indicators for an urban black-odorous water body; comparing and analyzing the remote sensing reflectance data to obtain spectral change features of the black-odorous water body and a general water body; constructing each node of a decision tree according to the spectral change features and based on Gini index, constructing a decision tree classification model to obtain classification results of the black-odorous water body and the general water body, calculating a classification accuracy; analyzing the classification results to obtain spatiotemporal distribution changes of black-odorous water bodies

Abstract: The present invention discloses a CEEMDAN-based method for screening and monitoring soil moisture stress in farmland, characterised by the steps: preprocessing of remote sensing images, construction of NDVI long time series, CEEMDAN decomposition, calculation of statistical descriptors, screening of soil moisture stress sequences, ground data measurement, construction of soil moisture stress characteristic curves, fitting of soil moisture stress response characteristic curves and predicting the content of soil moisture stress. The invention adopts CEEMDAN decomposition, which solves the problems of noise residue and low reconstruction accuracy in the previous methods, and the high reconstruction accuracy of decomposed component data is more conducive to capturing the transient effects of soil moisture stress, and realizes the screening and extraction of soil moisture stress by combining with the ground measured data.

Abstract: The present disclosure provides a data compression method for quantitative remote sensing with an unmanned aerial vehicle. The method performs preprocessing on a multispectral image acquired by an unmanned aerial vehicle, successively performs a three-dimensional convolution and a two-dimensional convolution on the multispectral image by an encoder to obtain deep feature information, performs quantizing and entropy encoding on the deep feature information, optimally distributes a loss and a code rate of the image through end-to-end joint training to obtain an optimal compressed image, and reconstructs the optimal compressed image by using a decoder. Image reconstruction quality and a compression ratio are improved by performing a plurality of convolutions on a multispectral pattern; quantizing and entropy encoding are performed on the convoluted deep feature information, to remove redundancy in a feature image, so as to improve the image reconstruction quality and the compression ratio.