Hyperspectral recovery from two images

Abstract

Hyperspectral imaging methods, devices and systems are described that improve the hyperspectral accuracy, and reduce manufacturing costs, by using two images that enable the recovery of hyperspectral images with high fidelity. One hyperspectral imaging device includes one or more imaging lenses, one or more sensors positioned to receive light associated with an object from the one or more imaging lenses, and a spectral filter. The hyperspectral imaging device is configured to capture a first image produced without using the spectral filer and to capture a second image produced with the spectral filter. The first image and the second image have different spectral contents, and the first and the second images are processed using a trained neural network for producing hyperspectral imaging data associated with the first and second images.