Chromatic aberration recognition method, system, readable storage medium and chromatograph

By acquiring dark field and illumination images of sensor pixels to calculate correction coefficients, constructing a standard sample library and performing color difference comparison, the problem of low efficiency in traditional color difference detection is solved, and rapid color difference detection of printed surfaces is achieved.

CN115471500BActive Publication Date: 2026-07-21SHENZHEN HEFEI TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHENZHEN HEFEI TECH CO LTD
Filing Date
2022-10-27
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing color difference detection technologies are inefficient and cannot effectively detect color differences on printed surfaces, leading to increased production difficulties.

Method used

By acquiring dark field images and illuminated images of different pixels of the target sensor, calculating correction coefficients for correction, constructing a standard sample image library, and using a detection frame of preset size to collect the image under test for color difference comparison, automatic and rapid color difference detection of printed surfaces is achieved.

Benefits of technology

It greatly improves the efficiency of color difference recognition, and can quickly identify color differences on the printed surface, replacing the traditional method of only detecting points and improving detection efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115471500B_ABST
    Figure CN115471500B_ABST
Patent Text Reader

Abstract

The application provides a color difference recognition method, system, readable storage medium and chromatograph. The method comprises the following steps: acquiring dark field images obtained by imaging different image elements in a target sensor, and acquiring illumination images corresponding to a preset standard white board under a uniform light field according to a preset light field condition; acquiring correction coefficients corresponding to all image elements according to the dark field images and the illumination images under the same image element, so as to correct the image elements of the target sensor according to the correction coefficients; acquiring a standard sample image in a corrected environment, and constructing a standard sample image library according to the standard sample image; acquiring a measured image corresponding to a detection area in a to-be-measured sample according to a detection frame with a first preset size, and performing color difference comparison on the measured image according to the standard sample image library, so as to output a color difference comparison result. The color difference recognition method can greatly improve the color difference detection accuracy and detection efficiency.
Need to check novelty before this filing date? Find Prior Art