Reflectance spectrum reconstruction method based on iterative threshold method

Abstract

A reflectance spectrum reconstruction method based on an iterative threshold method includes the following steps that principal component analysis is carried out on a training sample set to obtain primary function vectors of the training sample set, and first three primary function vectors are chosen according to the contribution rate to serve as reconstruction primary function vectors; a standard illuminant and a common RGB camera are adopted to obtain RGB tri-stimulus values of a multi-spectral test image, and by means of a conversion matrix, CIE1931 standard colorimetric observer spectral tri-stimulus values are obtained; according to the obtained primary function vectors of the training sample set, CIE1931XYZ tri-stimulus values of the multi-spectral test image, a CIE1931 standard colorimetric observer color matching function and relative spectral power distribution of the standard illuminant, a reflectivity spectral image of the multi-spectral test image is reconstructed by the adoption of the iterative threshold method.

Description

technical field [0001] The invention relates to a reflectance spectrum reconstruction method, in particular to a reflectance spectrum reconstruction method based on an iterative threshold method. Background technique [0002] The existing reflectance spectral reconstruction methods can be divided into three categories: direct reconstruction methods, interpolation reconstruction methods and learning-based reconstruction methods. The direct spectral reconstruction method must first know the spectral characteristic parameters of the spectral imaging system, and also needs to characterize the system noise, which makes it difficult for this type of method to be widely used; the interpolation reconstruction method has a large amount of calculation; the learning-based reconstruction method estimates There is a good correlation between the reflectance spectrum and the output value of the imaging system, which is in line with the physical characteristics of the data, but the accuracy...

AI Technical Summary

Problems solved by technology

[0003] Aiming at the problems of low reconstruction efficiency and low reconstruction precision in existing reflectance spectrum reconstruction methods, the present invention proposes a reflectance spectrum reconstruction method based on iterative threshold method, which can make full use of the spatial sparseness of reflectance spectrum and improve Spectral reconstruction efficiency and reconstruction accuracy of reflectance

Images