Machine learning-based coating embellishing method and system

Abstract

The invention discloses a machine learning-based coating embellishing method and system. The method comprises the following steps of: pre-establishing a basic sample card and completing mapping whichhas a formula concentration achieving a tri-stimulus value on the basis of machine learning model training; obtaining a reflection rate of a target color lump, calculating a tri-stimulus value of thetarget color lump, and inputting the reflection rate and the tri-stimulus value into a machine learning model to obtain a current formula result; and making a sample according to the current formula result, calculating an aberration, if the aberration is greater than a preset threshold value, obtaining a tri-stimulus value of the sample, inputting the tri-stimulus value of the sample into the machine learning model to obtain a new formula result, and calculating a new corrected formula result according to the current formula result and the new formula result to serve as a new current formula result. The method is capable of effectively solving the problems of long time consumption, high cost and bad effect in the coating embellishing industry; through importing a machine learning method, the system can obtain satisfied embellishing results in continuous evolution and learning; and the method and system have the advantages of being high in intelligence, high in expansibility and high inprecision.

Description

technical field [0001] The invention relates to paint color correction technology in the paint industry, in particular to a paint color correction method and system based on machine learning. Background technique [0002] Computer color matching can provide guidance for specific production practices and simplify the color matching process by using color theory to accurately describe the color properties of pigments. Computer color matching can be widely used in industries involving color such as paint, textiles, garments and automobiles. Computer color matching technology began in the 1930s. CIE (International Commission on Illumination) created the tristimulus value colorimetry system. Hardy successfully designed an automatic recording reflectance multi-angle spectrophotometer; The theory of absorption and scattering in the medium, which is the theoretical basis of most computer color matching systems in the present period - Kubelka-Munk theory, referred to as K-M theory, ...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to provide a method and system for paint color correction based on machine learning in order to solve the problem of excessive color difference caused by the first color matching of the existing color matching model in the current paint color repair industry. Introduce a spectrophotometer that can precisely measure color data, and propose a method based on the combination of particle swarm algorithm and machine learning for the coating industry to accurately predict the formula and realize the color correction function, which can realize high-precision computer color correction in the coating industry

Images