Hyperspectral reconstruction method based on attention mechanism and dense network and storage medium

The invention relates to the technical field of hyperspectral image acquisition, in particular to a hyperspectral reconstruction method based on an attention mechanism and a dense network and a storage medium, and the method comprises the steps: S1, inputting an original RGB image into a shallow feature extraction module, carrying out the shallow feature extraction, and outputting a feature image T1; s2, the feature image T1 is input into a deep feature extraction module, a feature image T2 is output, the deep feature extraction module is formed by connecting a plurality of dense attention modules, and each dense attention module is composed of an attention module and a dense module; s3, inputting the feature image T2 into a spectral feature recombination module, and outputting a hyperspectral image T3; and S4, the error between the real hyperspectral image data G and the reconstructed hyperspectral image T3 is calculated, the weight of the hyperspectral reconstruction network is optimized, channel informatio