Spectral reconstruction method based on sample feature matching

Abstract

The invention provides a spectral reconstruction method based on sample feature matching, which is characterized by comprising the following steps of: firstly, obtaining color response values of a training sample set and a test sample through digital imaging equipment under an ambient light source; screening the training sample set by using a sample feature matching method to obtain a clustering training sample color response value set and a spectrum set; performing normalization processing on the clustering training sample color response value and the test sample color response value, and performing calculation by using a least square method to obtain a coefficient matrix; and finally, reconstructing the spectral reflectivity of the surface of the object by using the coefficient matrix and a curve smoothing method. The spectral reconstruction method provided by the invention is high in reconstruction precision and simple in calculation, fully considers the physical characteristics ofspectral reflectance, and utilizes the characteristics of the sample to screen.

Description

technical field [0001] The invention relates to a spectral reconstruction method, in particular to a spectral reconstruction method based on sample feature matching, which can be widely used in the fields of textile color, printing detection, art reproduction, e-commerce and the like. Background technique [0002] There are two ways to quantitatively represent object color information: color information based on chromaticity and color information based on object surface spectral reflectance. Color information based on chromaticity is usually represented by color tristimulus values. For example, colors in ordinary digital cameras, LCD monitors or multimedia projectors are usually represented by R, G, and B tristimulus values, and these tristimulus values ​​can be converted to Other corresponding tristimulus values, such as CIE tristimulus values ​​X, Y, Z or Lab values, etc. The color tristimulus value is the color value obtained by integrating the relative spectral power di...

AI Technical Summary

Problems solved by technology

However, the above-mentioned spectral reconstruction method only considers the optimal value in the mathematical sense, ignoring the characteristics of the sample itself and the physical meaning of the spectral reflectance (for example, the value of the spectral reflectance has a value not greater than 1 or not less than 0), resulting in reconstruction There is a large deviation between the physical value and smoothness of the spectral reflectance and the actual object

Images