Hyperspectral remote sensing image reconstruction method

The invention discloses a hyperspectral remote sensing image reconstruction method. The method comprises the following steps: acquiring a hyperspectral remote sensing image; sequentially carrying out geometric correction and motion blur elimination processing on the hyperspectral remote sensing image; obtaining a preprocessed hyperspectral remote sensing image; synthesizing the preprocessed hyperspectral remote sensing images into corresponding RGB images, and constructing a training data set; training the dense connection convolutional neural network model by adopting the training data set until the dense connection convolutional neural network model converges to obtain an optimized dense connection convolutional neural network model; and inputting a to-be-reconstructed RGB image into the optimized dense connection convolutional neural network model to output a corresponding hyperspectral remote sensing image. According to the method, the spectrum reconstruction precision can be improved, meanwhile, the dens