Method and device for decomposing hyper-spectral intrinsic images

Abstract

The invention discloses a method for decomposing hyper-spectral intrinsic images and a device for decomposing the images. The method comprises the following steps of acquiring hyper-spectral image sequences; decomposing according to Retinex constraint, non-local texture constraint and absolute measure; solving the decomposed hyper-spectral images via a conjugate gradient algorithm, thereby acquiring initial reflected intrinsic images and direct illuminated intrinsic images; performing spectrum-based training and light transmission decomposition to the initial reflected intrinsic images by an indirect illuminated component decomposition algorithm studied based on a hyper-spectral dictionary, thereby obtaining the reflected intrinsic images and the indirect illuminated intrinsic images. According to the method, the accuracy of image decomposition is increased by adding the non-local texture constraint according to the Retinex theory; besides the method is simple and convenient and can better meet the usage requirements of an user.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method and device for decomposing hyperspectral intrinsic images. Background technique [0002] The appearance of an object in people's eyes is determined by the interaction between the object's essential attributes and environmental attributes, such as the shape, color, and material of the object's surface, as well as ambient lighting and reflections from surrounding objects. Among them, illumination and reflection are the two most important attributes: illumination corresponds to the illumination information in the environment, including direct illumination and indirect illumination; reflection corresponds to the inherent properties of objects, including shape, color and material. [0003] A major task in the field of computer vision is to extract intrinsic properties from the scene to describe each pixel of the image. In order to solve this highly underconstrained p...

AI Technical Summary

Problems solved by technology

[0004] Intrinsic image decomposition in related technologies is usually based on a single color or grayscale image, but due to the lack of spectral details of the scene, the effect of reflection and illumination separation is greatly limited

For the more unconstrained lighting decomposition problem, the related technology is to decompose the RGB value of the pixel in the lighting map and add user interaction, but this method is highly subjective and complicated to use.

Images