JP2003069841A - Apparatus and method for converting color and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To fast perform an interpolation operation of n (n>=4)-dimensional data suitable to software processing. SOLUTION: A computer 102 performs color conversion processing of CMYK image data generated by an image forming apparatus 100 to convert the CMYK image data into an RGB image and stores the RGB image in a storage medium, etc. Color conversion processing of CMYK four-dimensional data is subjected to four-dimensional memory map interpolation operation (four-dimensional color space is divided into a plurality of unit prisms each having 16-vertexes, and an interpolation parameter allocated to a unit prism having 16-vertex to which an input signal belongs is used to perform an interpolation operation) suitable to CPU processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color conversion apparatus, a color conversion method and a recording medium for inputting a color image signal and performing color conversion into a color signal for a color output apparatus, a color facsimile, a color printer, and a color printer. This technology is applied to color conversion devices such as hard copy and software for color printers that run on workstations.

[0002]

2. Description of the Related Art In a color image forming apparatus such as a color printer or a color copying machine, a CMYK image data for printing is stored in a storage device such as a hard disk so that a CMYK image generated for printing in the image forming apparatus can be repeatedly used. Some have functions that can be stored in. By converting this accumulated CMYK image into an RGB image that can be easily reused by a computer, it can be displayed on an image display device connected to the computer and viewed as a soft copy image. It is also possible to display the converted RGB image data on a display device or the like, perform processing such as adding characters or symbols thereto, and reproduce the processed digital image data again as a hard copy.

As a general method for efficiently color-converting a CMYK image into an RGB image, a memory map interpolation calculation method using a four-dimensional lookup table has been proposed (see Japanese Patent No. 2903808).

In the memory map interpolation method described above, the four-dimensional color space formed by four signals is divided into a plurality of five vertex bodies, and the input color data is included by using the upper bits of the input signal.
Highly accurate color conversion is realized over the entire color space by selecting a vertex body, reading the intensity for interpolation calculation corresponding to the five vertex bodies from the four-dimensional lookup table, and performing linear interpolation calculation.

Another method for efficiently color-converting a CMYK image into an RGB image is, for example, Japanese Patent Laid-Open No. 9-28.
There is a device described in Japanese Patent No. 4579. In this device, four-dimensional interpolation calculation is performed by linearly interpolating the results of a plurality of three-dimensional interpolation calculations. For example, in the case of CMYK signal input, a plurality of three-dimensional interpolation calculation results of C, M, and Y spaces are performed, and two of the three-dimensional interpolation calculation results are selected based on the upper bits of the K signal, and the lower order of the K signal is selected. A four-dimensional interpolation calculation result can be calculated by linearly interpolating with bits.

[0006]

However, in the former method described above, since the unit cube is divided into 24 five-vertex bodies, the judgment processing of 24 patterns is required by using the lower bits of the input signal. . Therefore, when executed by hardware, there is no problem because this judgment processing can be processed in parallel, but when executed by CPU (software) processing, it is necessary to successively judge the magnitude of the lower bits, so processing time is reduced. It takes.

In the latter device, since a plurality of three-dimensional interpolation operations are executed in parallel, the CPU
When processing is performed, it takes a considerable amount of time. For example, in the case of performing a four-dimensional interpolation operation in which each of the CMYK axes is divided into eight, the three-dimensional interpolation operation corresponding to each of the nine K signals is performed, and the three-dimensional interpolation operation is executed nine times in total. It will be.

The present invention has been made in view of the above problems, and an object of the present invention is to provide an n-type suitable for software processing.
An object of the present invention is to provide a color conversion device, a color conversion method, and a recording medium that can perform interpolation calculation of (n> = 4) -dimensional data at high speed.

[0009]

According to the present invention, a four-dimensional input color space (CMYK) is divided into a plurality of 16-vertex bodies selected by higher-order data of an input color signal, and one
The 6-vertex body is divided into 6 8-vertex bodies (T1 to T6) selected by the lower data of the input color signal, and the interpolation coefficients (P0 to P15) corresponding to the 8-vertex body to which the input color signal belongs are used. A color conversion device that performs interpolation calculation and converts into an output color signal (RGB), and storage means that stores four-dimensional interpolation coefficients (P0 to P15) corresponding to each of the eight vertex bodies (T1 to T6). Determination means for determining the 8-vertex body to which the input color signal belongs, based on the magnitude relationship of the lower-order data of the input color signal,
A means for reading out a four-dimensional interpolation coefficient corresponding to the determined eight-vertex body from the storage means; a means for generating a coefficient for three-dimensional interpolation calculation (Pa to Pd) from the four-dimensional interpolation coefficient; A high-speed color conversion is performed on an input signal of four or more dimensions even by CPU processing, by including means for performing a three-dimensional interpolation operation using the lower-order data of the input color signal and the coefficient for the three-dimensional interpolation operation. The unit polyhedron can be determined at high speed.

According to the present invention, color conversion can be executed at high speed even for a color signal containing a large number of color components such as multi-band data.

According to the present invention, the product-sum operation suitable for CPU processing can be executed at high speed.

According to the present invention, a CPU whose product-sum operation is not fast
Even in the case of processing, color conversion can be executed at high speed for multidimensional input signals.

In the present invention, the access time of the look-up table is shortened when color-converting an input signal of four or more dimensions.

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be specifically described below with reference to the drawings. (Embodiment 1) (Configuration of image processing system) FIG. 1 shows an example of the configuration of an image processing system according to the present invention. In the configuration example of FIG. 1, the image processing system includes image forming apparatuses 100 and 101, a computer 102, a display 103, and a printer control apparatus 104.

Here, the image forming apparatus 100 (P1) is
An image forming apparatus for supplying CMYK image data to an external device, and a color printer, a color copying machine, or a color facsimile machine having a data storage device such as a hard disk or a RAM can be used.

The computer 102 is the image forming apparatus 10.
CMYK image data supplied by 0 (P1) is RG
It has a color conversion function for converting to B image data. Also, software such as various applications and printer drivers can be implemented, and color-converted RG
The B image data is displayed on the display 103, or a printer driver corresponding to the image forming apparatus 101 is operated to output the image forming apparatus P2.

The image forming apparatus 101 (P2) receives drawing data from the computer 102 and performs hard copy output, and is a color printer or the like having a printer controller 104. Printer control device 104
Is a processing device for converting drawing data transmitted from the computer 102 into image data for hard copy output, such as a rendering processing unit, a band buffer, a color conversion processing unit, a gradation processing unit, and a page memory (not shown). Composed of.

In this embodiment, the printer controller 1
Although 04 is described as being mounted in the image forming apparatus P2, it may be an apparatus independent of the image forming apparatus P2, or a part of the function thereof may be realized by a printer driver of the computer 102 or the like. It may be executed.

(Operation of Image Processing System) Next, the operation of the entire image processing system of FIG. 1 will be described. First,
The image forming apparatus 100 (P1) stores the CMYK image data generated during the hard copy output in a built-in hard disk or the like. The hard copy output operation means, for example, a printing operation using a color printer or a copying operation using a color copying machine. The CMYK image data generated in the image forming apparatus 100 (P1) is transferred to the computer 102 through the network.
Transferred to. Computer 102 transferred CMY
Color conversion processing is performed on the K image data to convert it to an RGB image, and the RGB image is stored in a storage medium such as a hard disk included in the computer 102.

The operator of the computer 102 can display the RGB image data stored in the computer 102 on the display for viewing, processing, or outputting to another color printer for printing, if necessary. You can

(Explanation of Color Conversion System) Now, in the above-mentioned image processing system, the four-dimensional data of CMYK signals is converted to RG.
The computer 102 executes a color conversion operation for converting into three-dimensional data such as a B signal. In the present invention, this 4
The color conversion processing of the dimensional data is performed by using the four-dimensional memory map interpolation calculation suitable for the CPU processing.

Therefore, the principle of the memory map interpolation method for four-dimensional data according to the present invention will be described first. The four-dimensional memory map interpolation calculation divides a four-dimensional color space having four signals of C, M, Y, and K as axes into a plurality of unit 16-vertex bodies, and assigns them to the unit 16-vertex bodies to which the input signal belongs. The interpolation calculation is performed using the interpolation parameters.

Here, the C, M, Y, and K signals are each 8 bits.
An example will be described in which each axis is divided into 15 parts by data (divided into 15 squared units of 16 vertex bodies). 8-bit signal is 1
If divided into 5, one step is 255/15 = 17. Therefore, Tp = T / 17, ΔT = 17 (T is C,
Assuming that each color signal value of M, Y and K and Tp and ΔT are integer values, the input signal X is separated into upper data and lower data (C = 17 * Cp + ΔC, M = 17 * Mp + ΔM, Y =
17 * Yp + ΔY, K = 17 * Kp + ΔK).

The vertex coordinates (= lattice points) of the 16-vertex body surrounding the input signal X are [C, M, Y, K] = [17 *
Cp, 17 * Mp, 17 * Yp, 17 * Kp], [17
* Cp, 17 * Mp, 17 * Yp, 17 * (Kp +
1)], [17 * (Cp + 1), 17 * Mp, 17 * Y
p, 17 * Kp], [17 * (Cp + 1), 17 * M
p, 17 * Yp, 17 * (Kp + 1)], [17 * C
p, 17 * (Mp + 1), 17 * Yp, 17 * Kp],
[17 * Cp, 17 * (Mp + 1), 17 * Yp, 17
* (Kp + 1)], [17 * Cp, 17 * Mp, 17 *
(Yp + 1), 17 * Kp], [17 * Cp, 17 * M
p, 17 * (Yp + 1), 17 * (Kp + 1)], [1
7 * (Cp + 1), 17 * (Mp + 1), 17 * Yp,
17 * Kp], [17 * (Cp + 1), 17 * (Mp +
1), 17 * Yp, 17 * (Kp + 1)], [17 *
(Cp + 1), 17 * Mp, 17 * (Yp + 1), 17
* Kp], [17 * (Cp + 1), 17 * Mp, 17 *
(Yp + 1), 17 * (Kp + 1)], [17 * Cp,
17 * (Mp + 1), 17 * (Yp + 1), 17 * K
p], [17 * Cp, 17 * (Mp + 1), 17 * (Y
p + 1), 17 * (Kp + 1)], [17 * (Cp +
1), 17 * (Mp + 1), 17 * (Yp + 1), 17
* Kp], [17 * (Cp + 1), 17 * (Mp +
1), 17 * (Yp + 1), 17 * (Kp + 1)].

This unit 16 vertex body is shown in FIG. In FIG. 3, the unit 16-vertex body is originally a 4-dimensional solid, but since it cannot be shown, it is simply divided into two 3-dimensional cubes. That is, the left cube is a CMY unit interpolation solid when the K value is fixed to Kp × Δ, and the right cube is a CMY space unit interpolation when the K grid point is (Kp + 1) × Δ. It is a solid (Δ is the grid size of the unit interpolation solid, and in the case of 15 divisions, Δ = 17).

As an interpolation calculation method for this unit 16-faced body, there is known a method in which a unit 16-vertex body is divided into 24 5-vertices bodies to perform 5-point interpolation calculation.
Since there are also 4 patterns, the determination of 5 vertex bodies becomes complicated. Therefore, in the present invention, the 16-vertex body is divided into six 8-vertex bodies as shown in FIG. 4, and the 8-point interpolation calculation is used.

First, one unit 16-vertex body is selected from a plurality of unit 16-vertex bodies in the upper data of the input color signal. Next, in the 8-point interpolation calculation, the lower data Δ of C, M, Y
C, ΔM, and ΔY are determined to be large or small, and the 8-vertex body determination process is performed. FIG. 5 shows a judgment formula for this size judgment.

Subsequently, an interpolation calculation process is performed using the eight vertex bodies selected in the above-mentioned determination process. As an example, (T1)
In the case where is selected, the interpolation calculation method will be described with reference to FIG. In the interpolation calculation method according to the present invention, first, two corresponding vertices of eight vertices are linearly interpolated to create one tetrahedron. In the example of FIG. 6, Pa = P0 + (P8−P0) × Δk (Equation 1-a) Pb = P1 + (P9−P1) × Δk (Equation 1-b) Pc = P3 + (P11−P3) × Δk (Equation 1-c) Pd = P7 + (P15−P7) × Δk (Equation 1-d), and Pa, Pb, Pc, and Pd are calculated. However, Δ
k = ΔK / Δ (0 ≦ Δk ≦ 1.0), and Pi means the output value assigned to each vertex. By linearly interpolating the corresponding two vertices using the lower K data, the four-dimensional interpolation operation can be replaced with the three-dimensional tetrahedral interpolation operation.

When the tetrahedron is created, P = αΔC / Δ + βΔM / Δ + γΔY / Δ + δ (Equation 2-a) = (Pb-Pa) ΔC / Δ + in the same manner as the normal tetrahedral interpolation calculation in the three-dimensional space. The output value P can be calculated as (Pc−Pb) ΔM / Δ + (Pd−Pc) ΔY / Δ + Pa (Equation 2-b). The same calculation is performed when (T2) to (T6) are selected. The correspondence between each of the eight vertex bodies and the coefficients α, β, and γ in the above equation is as shown in FIG. 7. However, P (i, j) is a value obtained by linearly interpolating the output values assigned to the vertices Pi and Pj by Δk.

In the present invention, four-dimensional interpolation calculation is performed by the above calculation. In FIG. 3, the 16-vertex body of the interpolation target area is shown as two cubes, but the present invention does not limit the interpolation-target area to that, and the 16-vertex body with different side lengths or polar coordinate display. It can also be applied to the case of a modified 16-vertex body corresponding to an input signal or the like.

(Structure / Operation of Color Conversion Processing) A specific example of a color conversion device that realizes the above-described color conversion method will be described below. In the image processing system of FIG. 1, the computer 1
02 controls the system to convert the CMYK image transmitted from the image forming apparatus P1 into an RGB image. FIG. 2 shows a configuration example of the computer 102 that executes the color conversion operation. The computer 102 is a NIC for performing data communication with an external device, a DISK for temporarily storing image data sent from the NIC, a 4D-LUT storage unit for storing interpolation parameters, a central operation for executing color conversion processing and the like. It is composed of a processing device (= CPU), a RAM and the like.

First, when image data is sent via the NIC, the image data is temporarily stored in the DISK. Next, the CPU uses the image from DISK
Data is recalled and stored in RAM. Then, color conversion processing is performed while accessing the RAM. To perform the color conversion process, the image data is read from the RAM in pixel units, the CMYK values of the read pixels are determined, and 4D-L
The read address of the UT storage unit is determined. Then, after the interpolation parameters required for the interpolation calculation are read out, the pixel data is subjected to color conversion and converted into RGB signals or the like. Sequentially, color conversion is performed while calling pixel data from the RAM, and the conversion result is stored in DISK. When the color conversion of the CMYK image data held in DISK is completed, the image data is transferred to the external device via the NIC again.

The color conversion operation executed by the CPU will be described with reference to the flowchart of FIG. First, in step S101, the 8-vertex body corresponding to the input data is determined. FIG. 9 shows a flowchart of the 8-vertex body determination process. As shown in FIG. 9, according to the present invention, the determination process, which is not suitable for the CPU process, is performed only by the lower order data of the three signals of C, M, and Y, which can be executed in a maximum of 3 steps.

When the 8 vertices are determined, the next step S10 is performed.
In 2, the interpolation parameter is read from the 4D-LUT storage unit. The necessary parameters are as shown in FIG. 4D
The addressing of the LUT storage unit is performed based on the upper data Cp, Mp, Mp, Kp of C, M, Y, and K and the determination result of the 8-vertex field, which will be described in detail later. When the parameters of 8 vertices are read, P (i, j) is calculated in step S103, and three-dimensional interpolation calculation is performed in step S104 to calculate output RGB.

(Reading Parameters from 4D-LUT Storage Unit) A method of calling interpolation parameters from the 4D-LUT storage unit will be described. The four-dimensional lookup table is a storage device that stores output values at four-dimensional divided grid points. For example, in the case of 15 divisions, [C, M, Y, K] = [0, 0, 0, 0], [0, 0,
0,15], [0,0,0,30] ... [0,0,
0, 255], [0, 0, 15, 0], [0, 0, 1
5,15], [0,0,15,30], ...
[0,0,15,255], :: [255,255,255,0], [255,25
5,255,15], [255,255,255,3
0] ..., [255, 255, 255, 255]
The output values corresponding to CMYK are sequentially stored in the 4D-LUT memory.

When the parameters are stored in the memory in the order as described above, [C, M, Y, K] = [c, m,
The parameters [y, k] and [c, m, y, k + Δ] are always located at consecutive addresses in the memory. Therefore, since the parameters required to perform the linear interpolation calculation of (Equation 1) are continuously stored in the memory, two parameters can be called simultaneously with one memory access and high-speed parameter reading is possible. become.

For example, when reading the above-mentioned color conversion parameters for the 8-vertex body T1, since P1 and P9 are continuous, two bytes can be called at the same time. Similarly, P
Since 0 and P8, P3 and P11, and P7 and P15 can be read at the same time, a total of four memory accesses are required.

As described above, according to the present invention, the CPU is determined by the simple determination of the 8-vertex body and the high-speed memory access.
Color conversion suitable for processing can be executed.

Further, the calculation formula of the formula 1 is in the form of sum of products operation. CPUs of recent years have very high processing capabilities, and in particular, there are some that execute a product-sum operation at high speed in one clock, or calculate a plurality of product-sum operations in parallel. Considering such a situation, there is no problem in processing speed even if the sum of products operation of Expression 1 is executed as it is. However, when the CPU is unsuitable for sum-of-products calculation, the effect of simplifying the determination process is offset.

Therefore, when the CPU does not have a structure for accelerating the product-sum operation, the multiplication part of Expression 1 is replaced with a memory lookup. For example, in the case of (Equation 1-a), by holding the correspondence between the difference value (P8-P0) and Δk and the multiplication result in a static storage memory, the multiplication can be executed by the memory lookup process. .

Since Δk is the lower data of the K signal, the range of possible values is limited. For example, in the case of 15 divisions,
ΔK takes only 17 different values. In addition, the difference value (P8
Since −P0) is sufficient for 8 bits or less, the memory size required for this multiplication is at most 256 (bytes) × 17.
= 4.25 Kbytes, which is very small, does not impose a load on the CPU processing. The same multiplication memory can also be used in Equation 2, in which case multiplication in the interpolation operation is unnecessary. Therefore, in CPU processing, 8
By performing the vertex body determination at high speed, the performance of the entire interpolation calculation can be improved.

(Example 2) In Example 1, a 16-vertex body was divided into 8-vertex bodies, and three-dimensional tetrahedral interpolation calculation was performed by linear interpolation calculation. In this embodiment, the 16-vertex body is divided into two 12-vertex bodies based on the same idea.

FIG. 10 shows an example in which a unit 16-vertex body is divided into two 12-vertex bodies. In FIG. 10, the left and right cubes of the unit 16-vertex body are each divided into triangular prisms. In this case, since the determination process only compares the lower data of only two signals of the four signals, the determination process may be performed once.

The interpolation calculation of the 12-vertex body is the same as that of the first embodiment, and the corresponding 2 vertices are linearly interpolated with the lower data of the K signal to form a three-dimensional triangular prism, and then a general triangular prism interpolation calculation is performed. Is to do.

(Embodiment 3) Each of the embodiments described above is a CPU.
Although the embodiment is realized by processing, the present invention can be realized by hardware. FIG. 11 shows a configuration example when the present invention is realized by hardware. When it is realized by hardware, the arithmetic expression of (Expression 2-b) is modified to simplify the circuit. That is, in P = (Pb-Pa) ΔC / Δ + (Pc-Pb) ΔM / Δ + (Pd-Pc) ΔY / Δ + Pa, Pb-Pa = P1 + (P9-P1) × Δk-P0- (P8-P0) It can be written as × Δk = P1−P0 + (P9−P1−P8 + P0) × Δk = (P1−P0) + ((P9−P8) − (P1−P0)) Δk. Therefore, a = P1−P0 and a ′ = P9
By storing a and a'in the 4D-LUT as -P8, the simple product-sum operation result as shown in FIG.
The interpolation calculation can be realized by multiplying by C. Below, P
Similarly for c-Pb, Pd-Pc, and Pa, Pc-Pb = (P3-P1) + ((P11-P9)-(P3-P1)) [Delta] k Pd-Pc = (P7-P3) + ((P15- P11) − (P7−P3)) Δ k Pa = P0 + (P8−P0) × Δk, and thus the equation is generalized as follows: P = ((a′−a) ΔK + a) ΔC + ((b′− b) ΔK + b) ΔM + ((c′−c) ΔK + c) ΔY + (d′−d) ΔK + d. a, a ', b, b', c, c'is 6
Since it differs for each vertex, the memory address of the four-dimensional LUT is controlled and switched based on the determination result of the six vertex body. However, since d and d ′ are parameters common to all 6 vertices, 4D-L is used only for the upper data of C, M, Y, and K without using the result of the tetrahedral determination processing as shown in FIG.
Determine the address of the UT.

(Embodiment 4) In each of the above embodiments, the case where CMYK data is converted into RGB data has been described. However, the present invention is not limited to this, and can be applied to, for example, color conversion of multiband data, which is being considered for use in the field of remote medicine.

Generally, the spectral reflectance characteristics of objects existing in nature vary widely. Therefore, even if the color is the same under a certain lighting environment, when the lighting environment changes, a different color appears, causing metamerism. When metamerism occurs in performing remote medical care, a doctor in a remote place cannot accurately grasp the skin color of a patient, which hinders treatment. Therefore, the use of multi-band data is being studied in order to realize the same color reproduction even if the lighting environment on the patient side and the doctor side are different.

In this multi-band data, since the spectral distribution data of the object is expressed through a large number of filters having different spectral transmission characteristics, the object color is usually expressed by color signals of 10 or more components. Since such multidimensional color data must be finally converted into three-dimensional data for display on the monitor screen, a color conversion means for converting a multidimensional input color signal into a three-dimensional color signal or the like. Will be required.

The color conversion apparatus according to the present invention can also be used when converting n-dimensional data into m-dimensional data having a smaller dimension than n.

A specific method for converting n-dimensional data into m-dimensional data will be described. The n-dimensional data can be converted into the n-1 dimensional data by the linear interpolation of the above-mentioned (Equation 1). Therefore, by repeating this linear interpolation and reducing the number of dimensions, the n-dimensional interpolation calculation can be changed to the m-dimensional interpolation calculation. FIG. 12 shows a flowchart of this embodiment. Similar to the first embodiment, the vertex body is determined in step S202. When tetrahedral interpolation is used as the three-dimensional interpolation calculation, the number of vertex bodies of this n-dimensional data is 4 × 2.
Power of (n-3) (when using triangular prism interpolation, 6
X2 (n-3). Then, in step S203, the interpolation parameters corresponding to the vertices are read out, and step S
At 204, the corresponding two vertices are linearly interpolated. By this linear conversion, the number of vertices is halved, and the number of dimensions m is reduced by one. The linear interpolation process of step S204 is repeated, and when the number of dimensions m reaches 3, in step S206 a three-dimensional interpolation operation such as tetrahedral interpolation or triangular prism interpolation is performed to calculate an output value.

(Explanation of Storage Medium) FIG. 13 shows an image processing system when the computer 102 reads a program recorded on the storage medium and executes the program to execute the color conversion process described above. It is a figure showing an example of hardware constitutions. As shown in FIG. 13, this image processing system is realized by, for example, a workstation, a personal computer, or the like, and a CPU that controls the entire system.
And a ROM that stores CPU control programs, etc.
And a RAM used as a work area for the CPU
A hard disk, a display for displaying image data, and an image forming apparatus such as a color printer.

Here, the CPU, ROM, RAM, and hard disk have the function of the computer 102 of FIG. 1, and as described above, the CPU can also have the function of the color conversion processing device of the present invention. . Note that C
The function of such a color conversion device in the PU is, for example, a software package, specifically, a CD-RO.
It can be provided in the form of an information recording medium such as M. Therefore, in the example of FIG. 13, a medium driving device (program reading device) that drives the information recording medium when it is set is provided.

In other words, the color conversion apparatus and the color conversion method of the present invention make a general-purpose computer system equipped with an image scanner, a display, etc. read a program recorded on an information recording medium such as a CD-ROM, It is also possible to implement in an apparatus configuration in which the microprocessor of this general-purpose computer system executes the color conversion processing and the color conversion profile generation processing. In this case, the program for executing the color conversion processing of the present invention, that is, the program used in the hardware system is provided in a state recorded in the medium. The information recording medium on which the program and the like are recorded is not limited to the CD-ROM, and a ROM, a RAM, a flexible disk, a memory card or the like may be used. The program recorded on the medium can be installed in a storage device, such as a hard disk, built in the hardware system to execute the program and realize the color conversion function.

The program for implementing the color conversion apparatus and the color conversion method of the present invention may be provided not only in the form of a medium but also by communication, for example, by a server.

[0055]

As described above, according to the present invention, the following effects can be obtained. (1) The number of size comparisons required for unit solid determination is reduced, and color conversion of four-dimensional data can be performed at high speed when color conversion is performed by CPU processing or the like. (2) Since the tetrahedral interpolation calculation is performed as the three-dimensional interpolation calculation, the required size determination processing can be performed in three steps when performing the color conversion of the four-dimensional data, and the color conversion is performed by the CPU processing or the like. It is possible to perform color conversion of four-dimensional data at high speed. (3) Triangular prism interpolation calculation is performed as three-dimensional interpolation calculation. Therefore, when performing color conversion of four-dimensional data, necessary size determination processing can be performed in one step, and when performing color conversion by CPU processing or the like. Color conversion of four-dimensional data can be performed at high speed. (4) Color conversion can be performed at high speed even with multi-dimensional color data such as multi-band data. (5) When color conversion is performed using a CPU having a high-speed accelerator function for multiply-accumulate operation, color conversion of four-dimensional data can be performed at high speed. (6) Even a CPU that does not have an accelerator function for multiply-accumulate operations can perform color conversion of four-dimensional data at high speed. (7) Two interpolation parameters can be called at the same time with one memory access at the time of color conversion processing.
The UT can be used efficiently.

[Brief description of drawings]

FIG. 1 shows a configuration example of an image processing system according to the present invention.

FIG. 2 shows a configuration example of a computer that executes a color conversion operation.

FIG. 3 shows a unit 16 vertex body.

FIG. 4 shows six 8-vertex bodies.

FIG. 5 shows a judgment formula for selecting an 8-vertex body.

FIG. 6 is a diagram illustrating an interpolation calculation method according to the present invention.

FIG. 7 shows a correspondence relationship between 8-vertex bodies and coefficients α, β, and γ.

FIG. 8 shows a flowchart of interpolation calculation processing.

FIG. 9 shows a flowchart of an 8-vertex body determination process.

FIG. 10 shows two 12-vertex bodies.

FIG. 11 shows a configuration example when the present invention is implemented by hardware.

FIG. 12 shows a flowchart of the fourth embodiment.

FIG. 13 shows a configuration example when the present invention is implemented by software.

[Explanation of symbols]

100, 101 image forming apparatus 102 computer 103 display 104 printer control device

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) H04N 1/46 B41J 3/00 B 5C079 F term (reference) 2C087 AA15 BA07 BD01 BD05 BD53 2C262 AA24 AB19 BA01 BC13 BC15 BC19 EA04 EA06 GA14 5B021 AA01 AA05 AA19 LG07 5B057 CA01 CA08 CB01 CB08 CE18 CH07 CH08 5C077 LL18 MP08 PP32 PP33 PQ12 PQ23 5C079 HB01 HB03 HB12 LB02 MA04 MA11 NA11 PA03 PA05

Claims (10)

[Claims] 1. An n (natural number of 4 or more) -dimensional input color space is divided into a plurality of unit polyhedra, each unit polyhedron is divided into a plurality of solids, and an interpolation coefficient corresponding to a unit polyhedron to which an input color signal belongs is obtained. A color conversion device for performing an interpolation calculation using the conversion means to convert the input color signal into upper data and lower data by storing means for storing an n-dimensional interpolation coefficient corresponding to each solid. For the unit polyhedron selected by the upper data, a determination unit that determines the solid to which the input color signal belongs based on the size relationship of the lower data of the input color signal, and an n-dimensional image corresponding to the determined solid. Means for reading an interpolation coefficient from the storage means, means for generating a coefficient for an n-1 dimensional interpolation operation from the n dimensional interpolation coefficient, lower data of the input color signal and the n-1 dimensional interpolation operation Person in charge And a means for executing an n-1 dimensional interpolation calculation using a number. 2. A solid to which an input color signal consisting of n color signals belongs, in which an n (natural number of 4 or more) -dimensional input color space is divided into a plurality of unit polyhedra, and each unit polyhedron is divided into a plurality of solids. Is a color conversion device for performing an interpolation calculation using an interpolation coefficient corresponding to the above, and converting it into an output color signal, the storage means storing n-dimensional interpolation coefficients corresponding to each of the solids, and n of the input color signals. Of the input color signal based on the magnitude relation of n-1 or less lower data out of the n lower data for the unit polyhedron selected by the upper data. Determining means for determining a solid to which the object belongs, and n corresponding to the determined solid.
Means for reading out the dimensional interpolation coefficient from the storage means, means for generating a coefficient for n-1 dimensional interpolation operation from the n dimensional interpolation coefficient, lower data of the input color signal and the n-1 dimensional A color conversion device comprising: means for executing an n-1 dimensional interpolation calculation using a coefficient for interpolation calculation. 3. The color conversion device according to claim 1, wherein the n is 4, and tetrahedral interpolation calculation is performed as n-1 dimensional interpolation calculation. 4. The color conversion device according to claim 1, wherein the n is 4 and triangular prism interpolation calculation is performed as n-1 dimensional interpolation calculation. 5. A coefficient for interpolation calculation of m (<n) dimensions is generated by repeating a process of generating a coefficient for interpolation calculation of n−1 dimensions from the n-dimensional interpolation coefficient a predetermined number of times. The color conversion device according to claim 1 or 2, characterized in that: 6. The coefficients for the n-1 dimensional interpolation calculation are:
3. The color conversion device according to claim 1, wherein the two n-dimensional interpolation coefficients are generated by linearly interpolating the lower order data of the input color signal. 7. The coefficients for the n-1 dimensional interpolation calculation are:
3. The color conversion device according to claim 1, wherein the color conversion device is generated by a look-up table using the difference value of the two n-dimensional interpolation coefficients and the lower data of the input color signal as an address. 8. The color conversion device according to claim 6, wherein the two n-dimensional interpolation coefficients are arranged at consecutive memory addresses. 9. A color conversion method in which a four-dimensional input color space is divided into a plurality of 16-vertex bodies and interpolation calculation is performed using an interpolation coefficient corresponding to the 16-vertex bodies to which the input color signal belongs to convert into an output color signal. Where the input color signal is separated into upper data and lower data, and 16 vertex bodies selected by the upper data of the input color signal are divided into six 8 vertexes selected by the lower data of the input color signal. Dividing into 8 fields, determining the 8-vertex body to which the input color signal belongs based on the magnitude relation of the lower-order data of the input color signal, and 3 from the 4-dimensional interpolation coefficient corresponding to the determined 8-vertex body. Generating a coefficient for a three-dimensional interpolation operation, and performing a three-dimensional interpolation operation using the lower order data of the input color signal and the coefficient for the three-dimensional interpolation operation to generate the output color signal. Characterized by including How to convert. 10. A computer-readable recording medium recording a program for causing a computer to implement each step of the color conversion method according to claim 8.