CN103057292B - Printer calibration algorithm based on characteristics of human eye - Google Patents

Abstract

The invention discloses a printer calibration algorithm based on characteristics of a human eye. The calibration algorithm provided by the invention comprises the following steps: obtaining a calibrated sample by printing calibration ink formulae; measuring the calibrated sample to obtain a measurement spectrum; and predicting a prediction spectrum of the calibration ink formula by using a former model, and calculating to obtain a calibration coefficient by combining a polynomial and the sum in all wavelengths of CIE standard 1964 color observer tristimulus-value functions representing the characteristics of the human eye, so that a printer is calibrated. The printer calibration algorithm provided by the invention has the advantages of being simple in application and high in precision, is applicable to calibrating various printer spectrum characteristic models, and has a very high practical value.

Description

Printer Calibration Algorithm Based on Human Eye Characteristics

technical field

The invention relates to a printer calibration algorithm, in particular to a printer calibration algorithm based on the characteristics of human eyes.

Background technique

With the development of industrial technology, printers have been widely used in people's daily life. At the same time, a large number of scientific workers have carried out corresponding research on ink color management of printers. At present, a related research hotspot is the spectral characterization model of the printer, which can predict the spectral data after printing on a certain paper from the ink formula (ie, the amount of each ink). The existing printer spectral characterization models mainly include three-dimensional look-up table method, Yule-Clapper model and its improved model, Neugebauer model and its improved model, etc.

As far as the current technology is concerned, at least thousands of samples need to be printed and measured to establish a high-precision printer spectral characterization model, so establishing a spectral model is a relatively time-consuming and laborious project. However, the prediction accuracy of the final model is still constrained by multiple factors, such as printer repeatability, paper uniformity, ambient temperature, ambient humidity, ink cartridge replacement, etc. In order to solve this problem, it is necessary to calibrate the printer, that is, by printing some samples under the existing printing conditions to establish the connection with the established model (hereinafter referred to as "the original model"), so as to realize the original model. High-precision predictions under existing printing conditions. Although there are some studies on printer calibration, all printer calibration algorithms do not consider the characteristics of human eyes, and the printed colors are judged or appreciated by people. Therefore, a printer based on the characteristics of human eyes is proposed. Calibration algorithm is very necessary.

Contents of the invention

In order to solve the problems described in the background technology, the present invention discloses a printer calibration algorithm based on the characteristics of human eyes. This method realizes the calibration of the printer by combining the polynomial and the spectral tristimulus value function of the CIE (International Commission on Illumination) standard chromaticity observer representing the characteristics of the human eye. The specific steps are as follows:

1) Select n calibration ink formulas that basically cover each color area, generally n is 30-50;

2) Utilize the printer and paper used to build the original model to output the print sample of the calibration ink formula, that is, the calibration sample, and place the calibration sample in the darkroom for a period of time to ensure that the ink is fully dry;

3) A spectrophotometer is used to measure all calibration samples, and the corresponding measurement spectrum R _m (λ) is obtained, wherein the subscript m indicates measurement, and λ indicates wavelength;

4) Use the original model to predict all calibration ink formulations to obtain the predicted spectrum R _p (λ), where the subscript p represents prediction;

5) The sum S(λ) of the three-stimulus value functions x(λ), y(λ) and z(λ) of the CIE1964 standard chromaticity observer at each wavelength is used as the human eye characteristic function;

6) For wavelength λ, it is assumed that R _m (λ), R _p (λ) and S(λ) satisfy the following quadratic polynomial formula:

S(λ)×R _m (λ)=a(λ)×[S(λ)×R _p (λ)] ² +b(λ)×S(λ)×R _p (λ)+c(λ)

Among them, a(λ), b(λ) and c(λ) are calibration coefficients;

7) Calculate the final calibration coefficients a(λ), b(λ) and c(λ) according to the least square method;

8) For any ink formula, its spectrum R' _p (λ) can be predicted by the original model first, and then according to the formula in step 6) and the calibration coefficients a(λ) and b(λ) calculated in step 7) and c(λ), the final calibrated spectrum R' _m (λ) can be obtained, and its value is

R' _m (λ)＝{a(λ)×[S(λ)×R' _p (λ)] ² +b(λ)×S(λ)×R' _p (λ)+c(λ)} /S(λ).

The invention realizes the calibration of the printer by printing and measuring the measurement spectrum obtained from the calibration sample and the prediction spectrum of the original model, and combining the polynomial and the characteristics of the human eye. The invention has the advantages of simple application, high precision and high practical value.

Description of drawings

Figure 1 is a flow chart of the printer calibration algorithm based on the characteristics of the human eye;

Fig. 2 is the tristimulus value function x(λ), y(λ) and z(λ) of CIE1964 standard chromaticity observer;

Fig. 3 is the sum S(λ) of the tristimulus value function of CIE1964 standard chromaticity observer;

Fig. 4 is the scaling factor a (λ), b (λ) and c (λ) of final calculation in the embodiment;

Detailed ways

Taking an HP Designjet1055CM four-ink CMYK (green magenta yellow-black) four-ink printer and an established Yule-Clapper spectral model based on the printer (hereinafter referred to as "the original YC model") as an example, the above-mentioned analysis based on the characteristics of the human eye The specific implementation of the printer calibration algorithm will be described. As shown in Figure 1, the specific steps are as follows:

1) Select 40 calibration ink formulas that basically cover each color area, that is, the combination of CMYK ink consumption;

2) Use the printer and paper used to build the original model to output the printed sample of the calibration ink formula, that is, the calibration sample, and place the calibration sample in the dark room for 1 hour to ensure that the ink is fully dry;

3) Use Datacolor650 spectrophotometer to measure all 40 calibration samples, and measure the corresponding measurement spectrum R _m (λ), where the subscript m means measurement, λ means wavelength, and the measurement range is from 400nm to 700nm with 10nm as the interval;

4) Use the original YC model to predict all calibration ink formulations to obtain the predicted spectrum R _p (λ), where the subscript p represents prediction;

5) The sum S(λ) of the three-stimulus value functions x(λ), y(λ) and z(λ) at each wavelength of the CIE1964 standard chromaticity observer is used as the human eye characteristic function. Among them, the tristimulus value functions x(λ), y(λ) and z(λ) of the CIE1964 standard chromaticity observer are shown in Figure 2, and the sum of the tristimulus value functions of the CIE1964 standard chromaticity observer S(λ) As shown in Figure 3;

6) For wavelength λ, it is assumed that R _m (λ), R _p (λ) and S(λ) satisfy the following quadratic polynomial formula:

S(λ)×R _m (λ)=a(λ)×[S(λ)×R _p (λ)] ² +b(λ)×S(λ)×R _p (λ)+c(λ)

Among them, a(λ), b(λ) and c(λ) are calibration coefficients;

7) Calculate the final calibration coefficients a (λ), b (λ) and c (λ) according to the least squares method, as shown in Figure 4;

8) For any ink formula, its spectrum R' _p (λ) can be predicted by the original model first, and then according to the formula in step 6) and the calibration coefficients a(λ) and b(λ) calculated in step 7) and c(λ), the final calibrated spectrum R' _m (λ) can be obtained, and its value is

R' _m (λ)＝{a(λ)×[S(λ)×R' _p (λ)] ² +b(λ)×S(λ)×R' _p (λ)+c(λ)} /S(λ).

Claims (2)

1. the printer scaling algorithm based on human eye characteristic, is characterized in that comprising the following steps:

1) choose n the basic calibration ink set that covers each look district, n is 30-50;

2) utilize the printer that adopts while setting up master mould and the printing sample of paper output calibration ink set, calibrate sample, and it is fully dry to guarantee ink that calibration sample is placed to a period of time in darkroom;

3) adopt a spectrophotometer to measure all calibration samples, record its corresponding measure spectrum R _m(λ), wherein subscript m represents to measure, and λ represents wavelength;

4) utilize master mould to record in advance prediction spectrum R to all calibration ink sets _p(λ), wherein, subscript p represents prediction;

5) adopt CIE1964 standard colorimetric observer's tristimulus values function x (λ), y (λ) and z (λ) at each wavelength sum S (λ) as human eye characterisitic function;

6), for wavelength X, suppose R _m(λ), R _p(λ) and S (λ) meet following quadratic polynomial formula:

S(λ)×R _m(λ)＝a(λ)×[S(λ)×R _p(λ)] ²+b(λ)×S(λ)×R _p(λ)+c(λ)

Wherein, a (λ), b (λ) and c (λ) are calibration coefficient;

7), according to least square method, calculate final calibration coefficient a (λ), b (λ) and c (λ);

8), for arbitrary ink set, can first by master mould, predict its spectrum R ' _p(λ), then according to step 6) in formula and step 7) in calibration coefficient a (λ), the b (λ) and the c (λ) that calculate, can try to achieve the spectrum R ' after final calibration _m(λ), its value is

R’ _m(λ)＝{a(λ)×[S(λ)×R’ _p(λ)] ²+b(λ)×S(λ)×R’ _p(λ)+c(λ)}/S(λ)。

2. the printer scaling algorithm based on human eye characteristic according to claim 1, it is characterized in that described step 5) in adopt CIE1964 standard colorimetric observer's tristimulus values function x (λ), y (λ) and z (λ) the maximum of each wavelength replace CIE1964 standard colorimetric observer's tristimulus values function x (λ), y (λ) and z (λ) at each wavelength sum S (λ) as human eye characterisitic function, or employing CIE1931 standard colorimetric observer's tristimulus values function replacement CIE1964 standard colorimetric observer's tristimulus values function.