A computer color matching method for leather finishing
A computer color matching and finishing technology, applied in computing, manufacturing computing systems, neural learning methods, etc., can solve problems such as the lack of application of leather finishing color matching
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0039] Taking a 5-color database as an example, there are 5 pigment pastes in the database, namely red coffee (hereinafter referred to as coffee), pink (hereinafter referred to as red), sky blue (hereinafter referred to as blue), lemon yellow (hereinafter referred to as Yellow) and basic white (hereinafter referred to as white).
[0040] 1) Calculate the eigenvalue K of commonly used dyes i (λ), S i (λ); Prepare characteristic samples with multiple concentrations according to the common process coating, and then measure the K / S(λ) spectra of the characteristic samples respectively, see Table 1, take 4 ratios and use the least squares method to calculate K i (λ), S i (λ) value; here are selected to mix with white, the selected ratio is 2:8, 4:6, 6:4, 8:2.
[0041] x 1 K 1,λ -x 1 (K / S) s,λ S 1,λ +x 2 K 2,λ -x 2 (K / S) s,λ S 2,λ =0 (Formula 1)
[0042] Suppose: A 1,λ =x 1 A 3,λ =x 1 *(K / S) s,λ
[0043] A 2,λ =x 2 A 4,λ =x 2 *(K / S) s,λ
[0044] Therefore...
Embodiment 2
[0095] Aforesaid scheme runs according to embodiment 1, then carry out following steps according to following scheme again: (1) measure the reflectance of standard sample (red: blue 7:3), and convert into model reflectance; Measure and adopt Datacolor600 spectrophotometer , wavelength range 400 ~ 700nm, wavelength interval 10nm. use R t1 (λ) means that the R t2 (λ) is substituted into the DKM model formula (1), and transformed into the model reflectance K / S t2 (λ); (2) in K / S t1 (λ) is the input layer, apply the trained BP neural network for calculation, and obtain K / S wt2 (λ); prediction method: K / S wt (λ)=sim(net,K / S t (λ)); (3) with monochromatic model reflectance K i , S i (λ) Fitted weight-average model reflectance K / S wt2 (λ), use the least square method to obtain formula C1 (red: blue 69.99: 30.01); (4) According to the forecasted formula C 1 Make a proof and measure the reflectance R of the proof color c2 (λ); calculate the standard sample R t2 (λ) and proof...
Embodiment 3
[0097] The aforementioned scheme runs according to embodiment 1, and then carries out the following steps according to the following scheme: (1) measure the reflectance of the standard sample (blue: yellow: white 20:5:75), and convert it into a model reflectance; measure using Datacolor600 spectrophotometer, the wavelength range is 400-700nm, and the wavelength interval is 10nm. use R t3 (λ) means that the R t3 (λ) is substituted into the DKM model formula (1), and transformed into the model reflectance K / S t3 (λ); (2) in K / S t3 (λ) is the input layer, apply the trained BP neural network for calculation, and obtain K / S wt3 (λ); prediction method: K / S wt (λ)=sim(net,K / S t (λ)); (3) with monochromatic model reflectance K i , S i (λ) Fitted weight-average model reflectance K / S wt3 (λ), use the least squares method to obtain the formula C 1 (blue:yellow:white 19.67:3.16:77.17); (4) Formula C according to forecast 1 Make a proof and measure the reflectance R of the proof ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More