Prediction method of spot color overprinting spectrum based on density superposition principle
Through the spot color overprinting spectral prediction model based on the density superposition principle, the problem of large color difference of traditional prediction models is solved, the accuracy improvement of spot color overprinting prediction and production application needs are achieved, and theoretical support for prepress design and digital proofing is provided. .
Patent Information
- Application Number
- CN202211395723.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-08
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2042-11-08
AI Technical Summary
The traditional four-color overprint prediction model has complex calculations and too many demand samples, resulting in large color difference in spot-color overprint prediction and low optimization accuracy, making it difficult to meet the actual printing production needs.
Based on the principle of density superposition, a spot color overprint spectral prediction model is established. Through sample design, spectral measurement, model optimization and verification, a brand new spot color overprint prediction method is provided. The specific steps include sample design, spectral measurement, model establishment, Optimize and validate.
The accuracy of spot color overprinting is improved, providing a theoretical basis in pre-press design and digital proofing, meeting the needs of actual printing production, reducing color aberration and improving the feasibility and accuracy of prediction.
Smart Images

Figure CN115866152B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of digital proofing methods for gravure spot colors, and in particular relates to a method for predicting spot color overprinting spectrum based on the density superposition principle. Background Art
[0002] In the printing process, the latter spot color ink is attached to the film layer of the previous spot color ink, which is called spot color overprinting. Spot color overprinting has good visual effects and anti-counterfeiting effects, so many manufacturers often use spot color overprinting in packaging products when designing products. However, the traditional four-color overprinting prediction model is difficult to apply to spot color overprinting prediction due to problems such as complex calculations and too many required samples. At present, in the field of spot color overprinting prediction, prediction models are mostly established by describing the complex relationship between ink, substrate and light source. However, these models have problems such as large color difference in color prediction and low optimization accuracy, which are difficult to meet the actual printing production needs. Therefore, the present invention intends to establish a spot color overprinting color spectrum prediction model to meet the needs of prepress design and digital proofing in the actual printing production process. Summary of the invention
[0003] The purpose of the present invention is to provide a spot color overprinting color spectrum prediction method based on the density superposition principle, to solve the current spot color overprinting color prediction has large color difference and limitations in prepress design and digital proofing applications, and to provide a new idea for spot color overprinting color prediction.
[0004] The technical solution adopted by the present invention is a method for predicting the spectrum of spot color overprinting based on the density superposition principle, and the specific operation steps are as follows:
[0005] Follow the steps below to implement it:
[0006] Step 1, sample design and output:
[0007] The background color and foreground color are uniformly sampled within the dot area rate of 0-100% to generate a superimposed color spectrum.
[0008] Print out spot color overprint color spectrum;
[0009] Step 2, measuring the data such as the overprint color spectrum value and Lab value;
[0010] Step 3, establishing a spot color overprinting color spectrum prediction model based on the density superposition principle;
[0011] Step 4, optimizing the spectrum prediction model to obtain an optimized spectrum prediction model;
[0012] Step 5, verifying the accuracy of the optimized spectral prediction model;
[0013] The present invention is also characterized in that:
[0014] In step 1, the first printed spot color is defined as the back color; the second printed spot color is defined as the foreground color. The specific steps for generating the overprint color spectrum are: uniformly sample and design the foreground and back color scales with a dot area rate of 1%-10% as a step length; copy the scale upward with the foreground color scale as the horizontal coordinate to generate the foreground color spectrum; copy the scale to the left with the back color scale as the vertical coordinate to generate the back color spectrum; overprint the foreground color and back color spectrum to obtain the spot color overprint color spectrum; as shown in the specific steps. Figure 1-3 shown.
[0015] In step 2, under the conditions of D50 or D65 light source and 2° to 10° field of view, a spectrophotometer is used to obtain relevant data such as foreground color gradient, background color gradient, spot color overprinting color spectrum value, Lab value, etc.;
[0016] In step 3, based on the superposition principle of density, we get:
[0017] D p =K*(D b +D f ) (1)
[0018]
[0019]
[0020] Where D p is the spot color overprint density, D b is the background color density, D f is the foreground color density, and K is the model parameter variable.
[0021] From the definition formula of density, we can get:
[0022] R(λ)=10 -D(λ) (4)
[0023] The definition formula of density is shown in formula (5):
[0024]
[0025] Where D(λ) represents density and R(λ) is the spectral value.
[0026] The density value is converted into the spectrum value using formula (4), as follows:
[0027]
[0028] After conversion, we get:
[0029] R p =(R b ·R f ) K (7)
[0030] Where R p is the spot color overprint color spectrum value, R b is the background color spectrum value, R f is the foreground color spectrum value.
[0031] Considering the influence of paper, the spot color overprint prediction model can be obtained as shown in formula (8):
[0032]
[0033] Where R w is the paper spectrum value;
[0034] In step 4, the model is optimized by considering factors such as printing color sequence, overprinting rate, and ink transparency to establish an optimized spot color overprinting color spectrum prediction model. The specific model is shown in formula (9):
[0035]
[0036] Where K1 and K2 are optimization model parameter variables;
[0037] In step 5, the optimization model is verified from two aspects: technical feasibility and accuracy.
[0038] Technical feasibility verification: Based on Matlab, verify whether K1 and K2 exist to make the spectral prediction model valid, record and save the K1 and K2 values of the corresponding tones, and bring the K1 and K2 values into the model to calculate the color difference between the predicted spectral values and the measured spectral values to complete the technical feasibility test of the model.
[0039] Model accuracy verification: Use the prediction model to predict the spectral value of the spot color overprinting color, convert the spectral value obtained by the prediction model into Lab value through the colorimetric formula, and compare it with the measured Lab value of the overprinting color, and calculate the color difference to test the model accuracy.
[0040] In step 5, the Lab value is calculated using the relevant colorimetric formula, and the CIE2000 L*a*b* color difference formula is used for calculation. The colorimetric formula is shown in formula (10) and (11), and the color difference formula is shown in formula (12):
[0041]
[0042] In the formula, X, Y, and Z are the three stimulus values of the color sample, and k is the normalization coefficient. is the standard chromaticity observer tristimulus value, is the spectral power distribution function, λ is the wavelength, 380-780nm.
[0043]
[0044] in
[0045]
[0046] Where, X n , Y n , Z n is the tristimulus value of CIE standard illuminant, L * is the psychometric brightness, a * 、b * For psychometric chromaticity.
[0047]
[0048] In the formula, ΔL′, ΔC′, and ΔH′ are the lightness difference, chroma difference, and hue difference, respectively. K L , K C , K H is the correction factor related to the use conditions, S L , S C , S H is the weight function, R T is the rotation function.
[0049] The beneficial effect of the present invention is that the present invention establishes a spot color overprinting color spectrum prediction model based on the density superposition principle, and on the basis of research and experiments on foreground colors, background colors and spot color overprinting colors, makes reasonable assumptions based on the density superposition principle, and proposes a simple-to-operate spot color overprinting color spectrum prediction model, which provides a theoretical basis for the application of spot color overprinting color prediction in prepress design and digital proofing. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1 is the foreground color spectrum adopted in the present invention;
[0051] Figure 2 is the background color spectrum adopted in the present invention;
[0052] Figure 3 It is the spot color overprinting color spectrum adopted in the present invention. DETAILED DESCRIPTION
[0053] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0054] The present invention establishes a spectrum prediction model for spot color overprinting based on the density superposition principle, which is specifically implemented according to the following steps:
[0055] Step 1, sample design and output:
[0056] In the dot area rate of 0%-100%, the foreground and background color scales are evenly sampled and designed with the dot area rate of 1%-10% as the step length; the foreground color scale is used as the horizontal coordinate to copy the scale upward to generate the foreground color spectrum; the background color scale is used as the vertical coordinate to copy the scale to the left to generate the background color spectrum; the foreground color and background color spectrum are superimposed to obtain the spot color overprinting color spectrum. According to the process and printing materials selected by the customer, the spot color overprinting color spectrum is printed out;
[0057] Step 2, measuring the spot color overprinting color spectrum value, Lab value and other data:
[0058] Use an Exact spectrodensitometer under D50 or D65 light source and 2° to 10° field of view to obtain spot color overprinting color spectrum spectral values, Lab values and other related data using a spectrophotometer.
[0059] Step 3, establishing a spot color overprinting color spectrum prediction model based on the density superposition principle;
[0060] Based on the density superposition principle, we can get equations (1) to (3):
[0061] D p =K*(D b +D f ) (1)
[0062]
[0063]
[0064] Combined with the density definition formula (5), the density value can be converted into a spectral value.
[0065] From the definition formula of density, we can get:
[0066] R(λ)=10 -D(λ) (4)
[0067] The definition formula of density is shown in formula (5):
[0068]
[0069] Where D(λ) represents density and R(λ) is the spectral value.
[0070] The density value is converted into the spectrum value using formula (4), as follows:
[0071]
[0072] After conversion, we get:
[0073] R p =(R b ·R f) K (7)
[0074] Where R p is the spot color overprint color spectrum value, R b is the background color spectrum value, R f is the foreground color spectrum value.
[0075] Considering the influence of paper, a spectral prediction model of spot color overprinting based on the density superposition principle can be established. The specific model is shown in formula (8);
[0076]
[0077] Where R w is the paper spectrum value;
[0078] Step 4: Optimize the spectral prediction model and establish an optimized spectral prediction model:
[0079] Considering factors such as printing color sequence, overprint rate, and ink transparency, the spot color overprint spectrum prediction model is optimized to obtain an optimized model. The specific model is shown in formula (9);
[0080]
[0081] Step 5: Verify the accuracy of the optimized spectral prediction model:
[0082] (1) Technical feasibility verification
[0083] 1) Data preprocessing
[0084] The gradation spectrum value with a dot area rate of 10%-100% in the data measured in step 2 is divided by the paper spectrum value to complete the data preprocessing.
[0085] 2) Verify the technical feasibility of the model
[0086] Import preprocessed data into Matlab to verify whether there are appropriate K1 and K2 values to make the spectrum prediction model valid, and record and save K1 and K2 values of different gradations. Bring the obtained K1 and K2 values into the model to calculate the overprint color spectrum value, and calculate the color difference between the predicted spectrum and the measured spectrum value to verify the technical feasibility of the model;
[0087] (2) Model accuracy verification
[0088] Model accuracy verification: Use the prediction model to predict the spectral value of the spot color overprinting color, convert the spectral value obtained by the prediction model into Lab value through the colorimetric formula, and compare it with the measured Lab value of the overprinting color, and calculate the color difference to test the model accuracy.
[0089] In step 5, the Lab value is calculated using the relevant colorimetric formula, and the CIE2000 L*a*b* color difference formula is used for calculation. The colorimetric formula is shown in formula (10) and (11), and the color difference formula is shown in formula (12):
[0090]
[0091] In the formula, X, Y, and Z are the three stimulus values of the color sample, and k is the normalization coefficient. is the standard chromaticity observer tristimulus value, is the spectral power distribution function, λ is the wavelength, 380-780nm.
[0092]
[0093] in
[0094]
[0095] Where, X n , Y n , Z n is the tristimulus value of CIE standard illuminant, L * is the psychometric brightness, a * 、b * For psychometric chromaticity.
[0096]
[0097] In the formula, ΔL′, ΔC′, and ΔH′ are the lightness difference, chroma difference, and hue difference, respectively. K L , K C , K H is the correction factor related to the use conditions, S L , S C , S H is the weight function, R T is the rotation function.
[0098] Example
[0099] Taking the actual spot color overprint prediction as an example, the process of establishing the spot color overprint spectrum prediction model based on the density superposition principle is specifically described. The Lab values of the custom foreground color are: 47.78, 80.02, 11.76; the Lab values of the custom background color are: 79.26, 10.62, 57.86.
[0100] Step 1, sample design and output:
[0101] The foreground and back color scales are designed with uniform sampling at a dot area rate of 10% as a step length, and the foreground color scale is used as the horizontal coordinate to copy the scale upward to generate the foreground color spectrum (such as Figure 1As shown); copy the scale to the left with the background color scale as the vertical coordinate to generate the background color spectrum (as shown Figure 2 Overprinting the foreground color and the background color spectrum to obtain the spot color overprinting color spectrum (as shown in Figure 3 As shown), the gravure printing method and the back printing process are selected, the printing material is PET film, the Toyo gravure ink is used as the experimental ink, and the Sansei Seiki gravure proofing machine is used to actually proof and output the sample. This embodiment takes the first column of scales on the left side of the spot color overprinting color spectrum as an example to explain in detail the process of establishing the spot color overprinting color spectrum prediction model;
[0102] Step 2, measuring the overprint color spectrum value, Lab value and other data;
[0103] Using an Exact spectrodensitometer, under D50 light source and 10° field of view, the spectral values, Lab values and other data at different dot areas of the foreground color scale, the background color scale and the first column of scales on the left side of the three spot colors were measured, as shown in Table 1.
[0104] Table 1 Example of background color scale data
[0105]
[0106]
[0107] Step 3, establishing a spot color overprinting color spectrum prediction model based on the density superposition principle;
[0108] 1) Data preprocessing
[0109] Data preprocessing is completed based on formula (13):
[0110]
[0111] Where R1(λ) is the 31-dimensional foreground color spectrum value, R2(λ) is the 31-dimensional back color spectrum value, R3(λ) is the 31-dimensional overprint color spectrum value, and R w (λ) is the 31-dimensional paper spectrum value;
[0112] 2) Fitting model parameter K
[0113] Import the preprocessed data into Matlab, enable the Matlab curve fitting toolbox and input data with R3(λ)′ as the Z axis, R1(λ)′ as the Y axis, and R2(λ)′ as the X axis, using formula (14):
[0114] Z=(X*Y) K (14)
[0115] Fitting is performed to obtain the model parameters K and the determination coefficient R 2, root mean square error RMSE, record and save the K value; bring the K value into the spectrum prediction model to calculate the spectrum value, use formula (10) (11) to convert it into Lab value, compare the result with the measured Lab value to calculate the color difference ΔE1 to verify the technical feasibility of the model. The specific results are shown in Table 2;
[0116] Table 2 Model technical feasibility test
[0117]
[0118] It can be seen from Table 2 that the determination coefficients of the spectral prediction models are all greater than 0.95, and the average color difference of the model is 1.68NBS, which is less than 3NBS and meets the national standard requirements, indicating that the model is reasonable and feasible, and the establishment of the spectral prediction model is completed.
[0119] Step 4, optimizing the model and establishing an optimized spectrum prediction model;
[0120] Considering the influence of factors such as printing color sequence, overprinting rate, and ink transparency, the spot color overprinting spectrum prediction model is optimized to obtain an optimized model. The specific model is shown in formula (15):
[0121]
[0122] Where K1 and K2 are the parameter variables of the optimized spectral prediction model;
[0123] (5) Model verification
[0124] a. Model technical feasibility test
[0125] Import the data preprocessed in step 3 into Matlab, enable the Matlab curve fitting toolbox and use R3(λ)′ as the Z axis and R1(λ) as the ’ is the Y axis, R2(λ) ’ Input data for the X-axis and fit it using formula (16):
[0126]
[0127] The optimized model parameters K1, K2 and determination coefficient R were obtained by fitting. 2 , spectral root mean square error RMSE, record and save K1, K2 values, and bring the fitted K1, K2 values into the model to calculate the spectral value, and use formula (10) (11) to convert it into Lab value, and compare the result with the measured Lab value to calculate the color difference ΔE2 to verify the technical feasibility of the model. The specific results are shown in Table 3:
[0128] Table 3 Technical feasibility test of optimization model
[0129]
[0130] As can be seen from Table 3, the determination coefficients of the optimization models are all greater than 0.95, indicating that the models are reasonable and feasible. The average difference of the model is 0.916NBS, which is less than 3NBS and meets the national standards.
[0131] b. Model accuracy evaluation
[0132] The prediction model is used to predict the spot color overprinting color. The new foreground spot color is used, and its Lab value is: 63.463 -39.131 -28.635; and the background color is used, and its Lab value is: 39.779 1.268 1.684; the K1 and K2 values in Table 3 are called and substituted into the equation to calculate the spectrum value, and converted into Lab value using formula (8) (9). The result is compared with the actual measured overprinting spot color Lab value to calculate the color difference ΔE3. The results are shown in Table 4:
[0133] Table 4 Model color difference distribution
[0134]
[0135] From the results, we can see that the average difference of the model is 2.308NBS, which is less than 3NBS and meets the national standard requirements, indicating that the model is reasonable and feasible.
Claims
1. A method for predicting spot color overprinting spectrum based on the density superposition principle, characterized in that: The specific steps are as follows: Step 1: Sample design and output; Evenly sample the background color and foreground color within the dot area rate of 0-100%, generate the spot color overprinting color spectrum, and print out the spot color overprinting color spectrum; Step 2, measuring the spectral value and Lab value of the spot color overprinting color spectrum; Step 3, establishing a spectrum prediction model based on the density superposition principle; Step 4, optimizing the spectrum prediction model to obtain an optimized spectrum prediction model; Step 5, verifying the accuracy of the optimized spectral prediction model; Among them, in step 1, the first printed spot color is defined as the back color; the second printed spot color is defined as the foreground color; the specific steps of generating the spot color overprinting color spectrum are: uniformly sampling and designing the back color and foreground color scales with a dot area rate of 1%-10% as a step length; copying the scale upward with the back color scale as the horizontal coordinate to generate the back color spectrum; copying the scale to the left with the foreground color scale as the vertical coordinate to generate the foreground color spectrum; overprinting the back color spectrum and the foreground color spectrum to obtain the spot color overprinting color spectrum; In step 3, based on the superposition principle of density, we can get: D p =K*(D b +D f ) (1) Where D p is the spot color overprint density, D b is the background color density, D f is the foreground color density, K is the model parameter; Convert the above density values to spectral values as follows: After conversion, we get: R p =(R b ·R f ) K (7) Considering the influence of paper, the spot color overprint prediction model can be obtained as shown in formula (8): Where R w is the paper spectrum value.
2. The method for predicting spot color overprinting spectrum based on density superposition principle according to claim 1, characterized in that: The spot color overprinting color spectrum was used to obtain the spectral reflectance and Lab value at each dot area ratio by using an Exact spectrodensitometer under the conditions of D50 light source and 10° viewing field.
3. The method for predicting spot color overprinting spectrum based on density superposition principle according to claim 1, characterized in that: In step 4, the spectral prediction model is optimized to obtain an optimized spectral prediction model by considering the printing color sequence, overprinting rate, and ink itself, specifically: Among them, K1 and K2 are parameters of the optimized spectral prediction model, and their value ranges are both 0-1.
4. The method for predicting spot color overprinting spectrum based on density superposition principle according to claim 1, characterized in that: In step 5, the method for performing accuracy test on the optimized spectral prediction model is as follows: The optimized spectrum prediction model is used to calculate the overprint color spectrum value, the calculated spectrum value is converted into Lab value through the colorimetric formula, and the color difference ΔE between the predicted Lab value and the Lab value measured in step 2 is calculated using the CIE2000 L*a*b* color difference formula 00 , if the color difference is less than 3, it means the model optimization is successful; The colorimetric formulas are shown in equations (10) and (11), and the color difference formula is shown in equation (12): Where X, Y, Z are the three stimulus values of the color sample, k is the normalization coefficient, is the standard chromaticity observer tristimulus value, is the spectral power distribution function, λ is the wavelength, and the value range is 380-780nm; in Where X n , Y n , Z n is the tristimulus value of CIE standard illuminant, L * is the psychometric brightness, a * 、b * for psychometric chromaticity; In the formula, ΔL ′ , ΔC ′ , ΔH ′ They are brightness difference, chroma difference, hue difference, K L , K C , K H is the correction factor, S L , S C , S H is the weight function, R T is the rotation function.
Citation Information
Patent Citations
Method for detecting color difference of printed matter
CN101936883A
Spot color overprinted color prediction method based on spectrum
CN105681624A
Fabric image color calibration algorithm based on spectral reflectivity reconstruction
CN111047539A
Design method of intaglio spot color digital ladder ruler based on spectral reflectivity
CN111914316A
Color fiber color measuring and matching method based on global and local features
CN114972549A
