JP2016012782A - Image processing apparatus, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide accurate calibration that does not use a reference chart, and even if there are an individual difference and a change over time in characteristic between scanners, can remove the influence of the difference and change.SOLUTION: An image processing apparatus acquires image reading data created by an image reading device 110 reading an inter-device difference correction patch 121 that is formed into an image by a predetermined image forming apparatus 120; estimates, on the basis of image reading data created by a plurality of image reading devices 110 and image reading data created by an image reading device 110 to be calibrated, spectral response characteristics of the image reading device 110 to be calibrated; and calculates correction parameters on the basis of the estimated spectral response characteristics.

Description

  The present invention relates to an image processing apparatus, method, and program, and more particularly, to calibration (color calibration) of an image reading apparatus such as a scanner.

  When color image data digitized by a color image scanner is actually used, it is output to an output device such as a monitor, a color printer, or a printing machine. At this time, it is necessary to perform color correction on the read data of the color image scanner in order to match the color of a color print or the like obtained by outputting a document read by the color image scanner. At this time, in order to perform color conversion with high accuracy, calibration that adjusts color correction parameters so as to detect variations in characteristics and changes with time of components constituting the color image scanner and remove the influence may be performed. (For example, patent document 1).

  In Patent Document 1, in order to perform color conversion with high accuracy by correcting individual differences in scanner characteristics with respect to an original image and differences in reading values due to changes over time, scanner reading values of a reference chart made up of a plurality of color patches are stored in advance. Further, it is described that individual differences in scanner characteristics are corrected for a plurality of hue regions based on the reference data.

  As described above, the conventional calibration system for the color reproducibility of the scanner is premised on reading a reference chart managed in color. For this reason, in the adjustment in the manufacturing process as well as in the market, scanner calibration is costly due to the management of the reference chart and the increase in work time.

  In Patent Document 2, for the purpose of eliminating the need for a reference chart, a reference image is generated based on images input from a plurality of scanners, and a correction table is generated for each of the plurality of scanners based on the reference image. Is disclosed.

  According to the technique disclosed in Patent Document 2, it may be possible to suppress variations in relative color reading among a plurality of scanners. However, the problem that color output colors such as color prints do not match the original color due to individual differences in scanner characteristics with respect to the original image and differences in reading values due to changes over time cannot be solved.

  The present invention has been made in view of the above circumstances, and provides a high-precision calibration that can eliminate the influence of individual differences and temporal changes in scanner characteristics without using a reference chart. With the goal.

  In order to achieve the above object, the present invention provides an image processing apparatus for calculating correction parameters used for calibration of an image reading apparatus, wherein the image reading apparatus displays an evaluation color formed on a recording sheet by a predetermined image forming apparatus. Based on the image reading data acquisition means for acquiring the image reading data generated by reading, the image reading data of a plurality of image reading devices, and the image reading data of the image reading device to be calibrated, Spectral sensitivity characteristic estimation means for estimating the spectral sensitivity characteristic of the image reading device; and correction parameter calculation means for calculating a correction parameter based on the spectral sensitivity characteristic estimated by the spectral sensitivity characteristic estimation means. And

  According to the present invention, it is possible to provide a highly accurate calibration that can eliminate the influence of individual differences in scanner characteristics and changes over time without using a reference chart.

It is a block diagram which shows the function structure of embodiment by this invention. It is a figure which shows the specific example of the machine difference correction patch 121 used in the said embodiment. It is a figure which illustrates having variation in the spectral radiation characteristic of the light source of the line sequential CIS scanner used in the said embodiment. It is a figure for demonstrating the concept of the color conversion performed in the said embodiment.

<Example of overall configuration>
FIG. 1 shows a functional configuration of an image processing system 1 according to the present embodiment. A hardware configuration for realizing the configuration of the image processing system 1 is, for example, a multi-function multifunction peripheral (hereinafter, MFP (Multi-Functional Peripherals)). However, it is not essential to be physically integrated, and it can be implemented with a set of scanners and printers.

  The image processing system 1 reads image data from a document, converts the image data (analog signal) into digital data, and outputs it, and performs various correction processes on the read image data (digital data). The image processing apparatus 100 includes an image forming apparatus 120 that forms an image on recording paper based on image data from the image processing apparatus 100.

<Image forming apparatus>
In this embodiment, the image reading device 110 reads color image data (hereinafter referred to as RGB data) of three colors of R (red), G (green), and B (blue). Next, the image processing apparatus 100 performs color conversion of RGB data into four color image data (hereinafter referred to as CMYK data) of C (cyan), M (magenta), Y (yellow), and Bk (black). . Next, the image forming apparatus 120 outputs a color image on a recording sheet based on the CMYK data.

  In addition, the image processing apparatus 100 generates a “mechanical difference correction patch 121” that is a patch including a CMYK plate gradation pattern (internal pattern) for calibration for the image reading apparatus 110, and supplies this to the image forming apparatus 120. Output. FIG. 2 shows a specific example of the machine difference correction patch 121. The image forming apparatus 120 is registered in advance in the image processing system 1 in order to prevent individual differences among the image forming apparatuses from appearing in the machine difference correction patch 121.

<Image reading device>
The image reading device 110 is a scanner. In this embodiment, it is assumed that there are a plurality of scanners having the same configuration. The image reading apparatus 110 includes a light source 111, a transmission optical system 112, a photoelectric conversion element 113, and an output unit 114.

  The light source 111 is an LED light source, for example, and irradiates light onto a recording sheet to be read. The transmission optical system 112 is a lens, an infrared cut filter, or the like. It may not be used. The photoelectric conversion element 113 is an element that converts the read light into digital data. The output unit 114 is a functional block that outputs the digital data output from the photoelectric conversion element 113 to another device or the like.

  In this embodiment, a line-sequential CIS scanner is assumed as a specific example of the image reading apparatus 110. The line-sequential CIS scanner includes a CIS (contact image sensor) composed of LED light sources of three colors (wavelengths corresponding to RGB) and photoelectric conversion elements, an A / D converter, and a drive circuit for driving the three colors. LED light sources are sequentially turned on, and RGB 8-bit digital image data is generated and output from the density information of the original obtained by reading the set original line-sequentially.

  Line-sequential CIS scanners are generally smaller and thinner and consume less power than many other powerful CCD systems (there are many models that do not require a power cable and operate only with USB power supply) and are low in power. It has the advantage of being able to operate immediately without the need for warm-up time. However, it was inferior in reading speed and color reproducibility, and it was said that it was easy to focus on a part away from the table with a raised original or an uneven original such as when an open book was turned down. However, these drawbacks have been improved in recent years due to the progress of technological development, and it has been adopted in many scanners.

  On the other hand, because it consists of LED light sources with a narrow band peak frequency, if there is a deviation in the peak wavelength of the light source, even if the gray balance of RGB output is adjusted with a white shading plate etc., a specific spectral reflectance Will change the reading of the color with.

  In particular, in the line-sequential CIS scanner method in which an LED light source having a wavelength corresponding to RGB is sequentially turned on and a document is read as in this example, a deviation of the peak wavelength of the light source occurs independently at each wavelength corresponding to RGB. Even when compared with a white LED light source, variation in RGB read data for a color original tends to be large, and chromaticity management becomes difficult.

  FIG. 3 illustrates that the spectral radiation characteristics of each LED light source of the line-sequential CIS scanner vary. Not only the light source 111 but also each photoelectric conversion element 113 has variations in spectral sensitivity characteristics. Also, the transmission optical system 112 such as a lens or an infrared cut filter has variations in spectral transmission characteristics. An individual difference (machine difference) of the image reading apparatus 110 is generated due to a comprehensive difference in these characteristics.

  Even in the same image reading apparatus 110, there may be a difference between these previous characteristics and current characteristics due to deterioration over time. In the present specification, the overall characteristics of these characteristics are referred to as “total spectral characteristics”. Note that there are several conditions that cause variations in the overall spectral sensitivity characteristics of the image reading apparatus 110.

  In the image reading device 110, light is applied to a document by a light source 111, and reflected light from the document is collected on a photoelectric conversion element 113 through a transmission optical system 112. The photoelectric conversion element 113 outputs the density of RGB at each point on the document according to the resolution of the image reading device 110 as an image signal to the output unit 114. The output unit 114 performs a filtering process on the image signal and outputs it as RGB data to a subsequent device. In this example, for simplification of explanation, it is assumed that region determination such as a character region and an image region and processing according to the determination result are not performed.

<Image processing device>
In this embodiment, the RGB data output by the output unit 114 is converted into CMYK data by the image processing apparatus 100. This is called color conversion, and is executed by the color conversion means 106. As shown in FIG. 1, the image processing apparatus 100 includes an image reading data acquisition unit 101, a principal component score analysis unit 102, a spectral sensitivity characteristic estimation unit 103, a correction parameter calculation unit 104, a data storage unit 105, a color conversion as functional blocks. Means 106 and correction means 107 are provided.

  The color conversion means 106 calculates a hue component of the RGB data and unifies it linearly using a masking coefficient set for each area divided for each hue in order to unify the predetermined characteristics. As shown in FIG. 4, the hue is divided into RGB data by dividing the entire three-dimensional RGB color space by a plane extending radially around the achromatic color axis (Dr = Dg = Db).

  In color conversion, RGB data as an input color is “scanner vector”, CMYK data as an output color is “printer vector”, and linear conversion from the scanner vector to the printer vector is performed. Here, the scanner vector before the conversion is corrected by using the correction parameter calculated by the calibration, so that the characteristics are unified. The correction unit 107 executes the correction of the scanner vector.

<Calibration>
At the time of calibration, the user causes the image reading device 110 to be calibrated to read the machine difference correction patch 121. Then, RGB data that is image reading data obtained by reading the machine difference correction patch 121 is input to the image processing apparatus 100 through the process described above. Using the RGB data, calibration is executed by the image reading data acquisition unit 101, the principal component score analysis unit 102, the spectral sensitivity characteristic estimation unit 103, the correction parameter calculation unit 104, and the data storage unit 105.

  The image reading data acquisition unit 101 acquires RGB data from the output unit 114 of the image reading device 110. The principal component score analyzing unit 102 performs principal component analysis on the gradation pattern when reading the machine difference correction patch 121 including the gradation pattern of the process color.

  The spectral sensitivity characteristic estimation unit 103 estimates the spectral sensitivity characteristics of the calibration target image reading apparatus 110 based on the image reading data of the plurality of image reading apparatuses 110 and the image reading data of the calibration target image reading apparatus 110. To do. The correction parameter calculation unit 104 calculates a correction parameter based on the spectral sensitivity characteristic of the calibration target image reading apparatus 110 estimated by the spectral sensitivity characteristic estimation unit 103.

  The principal component score analysis unit 102 according to the present embodiment includes RGB data obtained by reading the machine difference correction patch 121 under a condition in which a variation in the overall spectral sensitivity characteristic occurs in the image reading apparatus 110 to be calibrated, An eigenvector for a difference from the RGB data obtained by reading the machine difference correction patch 121 under a standard condition in which such a variation is assumed not to occur is obtained.

<Calculation of Reading Characteristics of Image Reading Apparatus 110>
The spectral sensitivity characteristic estimation means 103 uses the spectral sensitivity characteristic (S _RGB [λ]) of the image reading device 110, the spectral radiation characteristic (E [λ]) of the light source 111, and the spectral sensitivity characteristic (C _RGB [C] of the photoelectric conversion element 113). λ]), a lens (L [λ]), an infrared cut filter (F [λ]: when used), and the like. Here, various spectral characteristics are handled at a pitch of 380 to 780 nm and 5 nm, and excess or deficiency of data is compensated by integration, interpolation and extrapolation.

Note that the spectral radiation characteristics of the LED light source change according to the chromaticity variation. In addition, the spectral emission characteristic (E [λ]) of the light source 111 of another apparatus or a past apparatus, the spectral sensitivity characteristic (C _RGB [λ]) of the photoelectric conversion element 113, a lens (L [λ]), and an infrared cut Data such as a filter (F [λ]: in use) is stored in the data storage unit 105.

Next, the correction parameter calculation unit 104 sets the output gain (G _RGB ) of the image reading device 110 according to the unit adjustment target and the target value (W _RGB ) of the machine difference correction patch 121 and its spectral reflectance (R [ λ]) based on the following formula (2). In the present embodiment, the output gain (G _RGB ) is a correction parameter.

Next, the correction unit 107 adds the product of the spectral sensitivity characteristic obtained by adjusting the gain of RGB data (O _RGB ) as a reading value of the image reading apparatus 110 and the spectral reflectance (T [λ]) of the machine difference correction patch 121. (Equation 3 below).

As is apparent from the above equation (Equation 1), the total spectral sensitivity characteristic (S _RGB [λ]) of the image reading device 110 to be calibrated is expressed as “spectral radiation characteristics, spectral sensitivity” of the plurality of image reading devices 110. Product of characteristics and spectral transmission characteristics ”(this data includes data of other apparatuses and past data of the apparatus itself) and“ spectral radiation characteristics, spectral sensitivity characteristics, It is estimated based on the product of spectral transmission characteristics.

Based on the total spectral sensitivity characteristic estimated in this way, the output gain of the image reading apparatus 110 to be calibrated is adjusted (Equation 2). As a result, the RGB data (O _RGB ) that is the calculated read value is a value that eliminates the influence of the image reading apparatus 110 due to machine differences and changes over time. Therefore, according to the present embodiment, it is possible to provide a highly accurate calibration that can remove the influence even if there is an individual difference or temporal change in scanner characteristics.

  Each function of the image processing apparatus 100 described above can be realized by a software program executed by a computer, for example.

DESCRIPTION OF SYMBOLS 1 Image processing system 100 Image processing apparatus 101 Image reading data acquisition means 102 Principal component score analysis means 103 Spectral sensitivity characteristic estimation means 104 Correction parameter calculation means 105 Data storage part 106 Color conversion means 110 Image reading apparatus 111 Light source 112 Transmission optical system 113 Photoelectric conversion element 114 Output means 120 Image forming apparatus 121 Machine difference correction patch

Japanese Patent No. 4607723 JP 2002-262007 A

Claims (6)

An image processing apparatus that calculates correction parameters used for calibration of an image reading apparatus,
Image reading data acquisition means for acquiring image reading data generated by the image reading device reading the evaluation color formed on the recording paper by the predetermined image forming device;
Spectral sensitivity characteristic estimation means for estimating spectral sensitivity characteristics of the calibration target image reading device based on the image reading data of a plurality of image reading devices and the image reading data of the calibration target image reading device;
Correction parameter calculation means for calculating a correction parameter based on the spectral sensitivity characteristic estimated by the spectral sensitivity characteristic estimation means;
An image processing apparatus comprising:   The image processing apparatus according to claim 1, wherein the evaluation color includes a gradation pattern formed with a process color of the predetermined image forming apparatus. The image reading data of the plurality of image reading devices is:
The image processing apparatus according to claim 1, further comprising past image reading data of the calibration target image reading apparatus. The image reading data of the plurality of image reading devices is:
The image processing apparatus according to claim 1, comprising past image reading data of another image reading apparatus that is not the image reading apparatus to be calibrated. A computer-based image processing method for calculating correction parameters used for calibration of an image reading apparatus,
An image reading data acquisition step of acquiring image reading data generated by the image reading device reading the evaluation color formed on the recording paper by the predetermined image forming device;
A spectral sensitivity characteristic estimation step for estimating a spectral sensitivity characteristic of the calibration target image reading device based on the image reading data of a plurality of image reading devices and the image reading data of the calibration target image reading device;
A correction parameter calculation step for calculating a correction parameter based on the spectral sensitivity characteristic estimated by the spectral sensitivity characteristic estimation step;
An image processing method comprising: On the computer,
An image reading data acquisition process for acquiring the image reading data generated by the image reading device reading the evaluation color formed on the recording paper by the predetermined image forming device;
Spectral sensitivity characteristic estimation processing for estimating spectral sensitivity characteristics of the calibration target image reading device based on the image reading data of a plurality of image reading devices and the image reading data of the calibration target image reading device;
A correction parameter calculation process for calculating a correction parameter based on the spectral sensitivity characteristic estimated by the spectral sensitivity characteristic estimation process;
An image processing program for executing

Images