Spectral image construction method and device, equipment and storage medium

Abstract

The invention provides a spectral image construction method and device, equipment and a storage medium, and the method comprises the steps: obtaining a to-be-processed spectral image and a color image corresponding to the to-be-processed spectral image; preprocessing the spectral image to be processed to obtain a preprocessed spectral image; and inputting the preprocessed spectral image and the color image into a feature learner to perform multi-layer feature learning processing to obtain a target spectral image, the multi-layer feature learning processing including multiple deep feature learning processing and multiple shallow feature learning processing. According to the spectral image construction method, device and equipment and the storage medium, all the characteristics of the color image from the deep layer to the shallow layer are fully excavated, and the high-precision high-resolution hyperspectral image is rapidly reconstructed.

Description

technical field [0001] The present application relates to the field of image processing, in particular, to a construction method, device, equipment and storage medium of a spectral image. Background technique [0002] Existing image fusion methods can be divided into traditional physical model methods and deep learning methods. Traditional physical models model by studying the imaging principles of RGB cameras and hyperspectral imagers, extract the spatial structure information of high-resolution RGB images and the spectral information of low-resolution hyperspectral images, and fuse them based on the established physical model. However, in practical applications, due to the complexity of the real world and imaging equipment, it is very difficult to take into account all the influencing factors when modeling, and most of the traditional model algorithms based on image fusion have hyperparameters, and their reconstruction accuracy has a great impact on hyperparameter sensiti...

AI Technical Summary

Problems solved by technology

However, the method of deep learning requires a lot of computing power and data support, and its network parameters need to be readjusted for training between different data sets.

[0004] Therefore, the process of image fusion using traditional physical models and deep learning methods is complex, and it is difficult to quickly reconstruct high-precision high-resolution hyperspectral images

Images