Spectral curve reconstruction method, device and system

Abstract

A spectral curve reconstruction method, device and system. Obtain the sampling result of the first sampling point of the object under test, the sampling result of the first sampling point includes the sampling results of the M channels of the sampling point; according to the light source output spectral function and the spectral response of the M sampling channels of the sampling device The product of the function determines M sampling functions, and the M sampling functions include overlapping areas; the sampling array matrix is ​​determined according to the impact function and the sampling function, and the impact function group includes multiple unit impact functions; the undershoot is determined by the sampling array matrix and the sampling result A system of definite equations, any solution of the system of underdetermined equations is the spectral function of the first sampling point. In this application, by changing the transfer response function of the light source or the camera, overlapping areas exist between sampling points with different output spectral functions and sampling functions formed by the spectral response function. Avoid the fence effect caused by undersampling and improve sampling accuracy.

Description

technical field [0001] The present application relates to the technical field of detection through spectral curves, in particular to a spectral reconstruction method, device and system. Background technique [0002] At present, the requirements for color quality control of printed matter are constantly increasing. This puts forward higher requirements for the corresponding color detection means. It is necessary to achieve high-precision real-time measurement of the spectral reflectance function of the printed matter surface. With the improvement of printing quality requirements for printed matter, the requirements for color quality control of printed matter are also continuously increased. This puts forward higher requirements for the corresponding color detection means. Especially for online color detection of printed matter, not only requires accurate measurement of printed color, but also requires extremely high detection speed. The essence of high-precision color det...

AI Technical Summary

Problems solved by technology

Due to the limitation of the number of sampling points, it leads to undersampling relative to the sampling accuracy requirements

Undersampling will lead to the fence effect when reconstructing the curve. The fence effect is that the surrounding area of ​​the sampling point cannot be covered, and the area that cannot be covered is the observation blind area

As a result, the accuracy of the spectral reconstruction results is insufficient

Images