JP4411230B2 - COLOR CONVERSION DEVICE, ITS CONTROL METHOD, AND PROGRAM - Google Patents

Description

The present invention is a color conversion apparatus and a control method thereof of the image data, and to program.

  In various systems that handle images, such as image output by a printer, color conversion processing for images is required. When the input and output for color conversion processing are in a relatively simple relationship, it is possible to perform color conversion processing by masking processing using matrix operation, but in general, there is a difference between input and output for color conversion. It is not always possible to approximate the relationship with a simple matrix operation with high accuracy.

  In such a case, a color conversion method using a three-dimensional lookup table (hereinafter referred to as a three-dimensional LUT if omitted) and an interpolation operation (for example, a polyhedral interpolation operation such as a tetrahedral interpolation operation) is used. Often adopted.

  FIG. 9 is a block diagram showing a configuration of a conventional color conversion apparatus using such a three-dimensional LUT and interpolation calculation. The color conversion apparatus shown in FIG. 9 has three-dimensional input values 901, 902, and 903. Is converted into an output value 913 by using a three-dimensional LUT for the lattice points and interpolation calculation.

  Specifically, first, the upper bits of the input values 901, 902, and 903 are extracted, and the corresponding lattice point numbers in the three-dimensional LUT are calculated in the lattice point number calculation units 904, 905, and 906. The three grid point numbers obtained by the calculations in the grid point number calculation units 904, 905, and 906 are given to the conventional three-dimensional LUT reference unit (3D LUT reference unit) 914. The three-dimensional LUT reference unit 914 converts the lattice point number into a corresponding address on the three-dimensional LUT 915 and reads the data at the address from the three-dimensional LUT 915 to obtain a lattice point output value. Then, in order to obtain the grid point output values 907, 908, 909, and 910 as necessary in the next interpolation calculation unit 912, after repeatedly referencing the three-dimensional LUT 915, the obtained output value is interpolated by the interpolation calculation unit 912. The final output value 913 is obtained. The interpolation coefficient value used in the interpolation calculation unit 912 is calculated from the input value by the interpolation coefficient value calculation unit 911 and passed to the interpolation calculation unit 912.

  The color conversion method using the three-dimensional LUT 915 and the interpolation operation as shown in FIG. 9 can perform color conversion with higher accuracy than the matrix operation method, but in order to obtain high color conversion accuracy, 3 It is necessary to set the size of the dimension LUT 915 to be large to some extent. Therefore, the size of the lookup table increases, which may increase the cost or decrease the color conversion processing speed.

Therefore, a color conversion device that saves memory capacity for holding data in the lookup table by compressing and holding data in the lookup table for each output color is conventionally known (for example, , See Patent Document 1).
JP 2000-22974 A

  However, in the conventional technique disclosed in Patent Document 1, although the memory capacity for holding data in the three-dimensional lookup table can be saved, when the lookup table is actually referred to, it is held in the lookup table. Therefore, it is not possible to save the memory capacity for holding the data at that time.

  Also, especially when considering systems with low processing speed but a large memory capacity storage system and systems with high processing speed but a small memory capacity storage medium, the data has already been looked up In a table, data cannot be stored on a storage medium with a high processing speed and a small memory capacity, and therefore a storage medium with a low processing speed and a large memory capacity is arranged, which causes a reduction in the processing speed. I will.

  Also, if the bit accuracy of data output from the three-dimensional lookup table is increased from 8 bits to 10 bits, for example, the size of the lookup table must be increased by the amount of increased bit accuracy. Memory capacity will increase.

  Further, the address calculation when referring to the look-up table is different between the 8-bit accuracy and the 10-bit accuracy, and the address calculation process becomes complicated when trying to cope with both accuracy.

  In order to reduce the memory capacity during actual use of the lookup table, it is conceivable to store and use data with the compression process performed on the three-dimensional lookup table. In order to refer to the lookup table with respect to the above data, since data compression / decompression processing is required when referring to the lookup table, there is a problem that the processing operation becomes complicated.

The present invention has been made to solve the problems of these prior art, and its object is color which can save memory space required for the actual use of the multidimensional look-up table in which storage means conversion apparatus and a control method thereof, and to provide a program.

In order to achieve the above object, the color conversion apparatus of the present invention comprises a coefficient value storage means for storing coefficient values corresponding to each area obtained by dividing the range of the multidimensional lookup table into a plurality of areas, and an input Corresponding from the coefficient value storage means depending on which of the plurality of areas the lattice point calculation means for calculating the lattice points in the multi-dimensional lookup table corresponding to the value, a coefficient value reference means for selecting the coefficient values, and the function value calculating means for calculating a function value by using the calculated coefficients selected values for the lattice points, a function value the function value computing means has computed Error value storage means for storing the difference from the value to be output as an error value corresponding to each of the grid points , and obtaining the error value corresponding to the calculated grid point from the error value storage means Error value And resulting unit, the function value calculating means that is adding the error value said error value acquisition means has acquired for function values calculated, adding means for calculating an output value of the calculated grid points, wherein Interpolation calculation means for obtaining an output value corresponding to the input value by interpolation calculation based on the output value calculated by the addition means .

According to the present invention, it is possible to save the memory capacity required for actual use of the error value storage means is a multi-dimensional look-up table.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First embodiment)
First, a first embodiment of the present invention will be described with reference to FIGS.

  First, the principle of the present invention will be described using an example of a one-dimensional LUT.

  FIG. 1 is a diagram for explaining the principle when the present invention is applied to a one-dimensional LUT. In FIG. 1, the vertical axis indicates an output value, the horizontal axis indicates an input value, and a solid line L1 indicates a solid line L1. The output value of the LUT is shown.

  In a normal LUT configuration, a lookup table that stores an output value corresponding to the solid line L1 is used for each input value. Therefore, the size of the lookup table is the number of values that can be taken by the number of output values. The number of bits required to store the range is required.

  The present invention reduces the number of bits required to store the range of the output value, which is realized by using an approximate function value as follows.

  In FIG. 1, a plurality of broken lines L2 are output values of approximate function values of the output value of the LUT (solid line L1), and the input value is an approximate function value division position indicated by a two-dot chain line L3 in FIG. Different approximate function values are used depending on which part of the divided area is included. Here, the reason why the input value is divided into a plurality of regions and different approximation function values are used for the respective areas is that the error between the output value of the LUT (solid line L1) and the output value of the approximation function value (dashed line L2). This is to make it as small as possible.

  In FIG. 1, when the error (difference) between the output value of the LUT indicated by the solid line L1 and the output value of the approximate function value indicated by the broken line L2 is viewed, it becomes narrower than the range of the output value of the original LUT. The error value can be expressed by a smaller number of bits than the output value of the LUT.

  Next, the operation of the color conversion apparatus according to the present embodiment for the one-dimensional LUT will be described with reference to FIGS. 2 and 3 based on the description of FIG.

  FIG. 2 is an explanatory diagram of approximate function values in the color conversion apparatus according to the present embodiment, in which the vertical axis indicates the output value of the function value and the horizontal axis indicates the input value.

  FIG. 3 is an explanatory diagram of an error value storage table in the color conversion apparatus according to the present embodiment, in which the vertical axis indicates an error value and the horizontal axis indicates an input value.

  When the input value of the one-dimensional LUT is given, first, the approximate function value shown in FIG. 2 is calculated, and the approximate function value is output. The approximate function value is a plurality of linear function values in FIG. 2, and different coefficient values are used for each range of input values divided into four by the broken line L4 in FIG. In the case of a linear function value, since there are two coefficient values necessary for the calculation, 2 × 4 = 8 coefficient values are required as a whole.

  Further, the error value storage table shown in FIG. 3 is referred to using the input value of the one-dimensional LUT. This error value storage table is a table in which the difference between the output value of the LUT of FIG. 1 and the output value of the approximate function value is stored for each input value, and since the range is narrow, there are fewer bits than the output value of the LUT. You can use a number table.

  Finally, the output value of the approximate function value calculated in FIG. 2 and the output value (error value) of the error value storage table referred to in FIG. 3 are added, and the output of the LUT corresponding to the solid line L1 in FIG. A value is obtained.

  If the number of coefficient values necessary for calculating the approximate function value is small, the size of the error value storage table and the size of the coefficient value storage area can be added to be smaller than the original one-dimensional LUT size. The memory capacity when using the LUT can be saved.

  In the case of a color conversion apparatus, since a three-dimensional input value is usually handled, a three-dimensional LUT will be described with reference to FIG.

  FIG. 4 is an explanatory diagram of a three-dimensional LUT in the color conversion apparatus according to the present embodiment, and the three axes L5, L6, and L7 in the same figure indicate the three dimensions of the input.

  In the three-dimensional LUT used for color conversion, a corresponding output value is stored for each grid point obtained by dividing an input value into a grid at appropriate intervals.

  In FIG. 4, a three-dimensional LUT is configured by storing corresponding output values for each vertex of a cube or rectangular parallelepiped obtained by division.

  In the three-dimensional case, when the number of lattice divisions increases, the number of lattice points increases abruptly as compared with the one-dimensional case. Therefore, the demand for reducing the table size of the three-dimensional LUT is further increased.

  When the present invention is applied to a three-dimensional LUT, the principle is exactly the same as that of the one-dimensional LUT described above. First, an approximate function value is calculated, and then an error value storage table is referred to. The final LUT output value is obtained by adding the approximate function value and the error value.

  When obtaining the approximate function value for the three-dimensional LUT, the range of the input value is divided into a plurality of regions, and different coefficient values are applied to each region. An example of the region division is shown in FIG.

  In FIG. 5, three axes L8, L9, and L10 indicate the three dimensions of the input, and here, since each dimension is divided into two, it consists of 2 × 2 × 2 = 8 regions as a whole.

If a linear function value is used as an approximate function value, an output value o for an input value (x, y, z) is obtained by using four coefficient values a, b, c, d,
o = a * x + b * y + c * z + d
Therefore, the number of necessary coefficient values as a whole is 4 × 8 = 32.

  Further, the error value storage table is the same as the one-dimensional case except that the number of dimensions of the lookup table is three-dimensional, but the number of input points to be handled is increased in the three-dimensional case compared to the one-dimensional case. Therefore, the effect of reducing the number of bits of the data of each lookup table becomes larger.

  Hereinafter, the color conversion apparatus according to the first embodiment of the present invention will be described with reference to FIG. 6 based on the principle described for the above-described one-dimensional LUT and three-dimensional LUT.

  FIG. 6 is a block diagram showing the configuration of the color conversion apparatus according to this embodiment. In FIG. 6, the same parts as those in FIG.

  6 differs from FIG. 9 in that the configuration of the three-dimensional LUT reference unit, the addition of the coefficient value lookup table (coefficient value LUT) 606 to the configuration of FIG. 9, and the output value of FIG. Instead of the three-dimensional LUT (3D LUT) 915, a three-dimensional LUT (3D LUT) 607 holding an error value is provided.

  The color conversion device converts the three-dimensional input values 901, 902, and 903 into an output value 913 by the three-dimensional LUT reference unit and the interpolation calculation unit 912.

  The output value 913 of the color conversion device is generally multidimensional in many cases, but only one dimensional output is shown here. In the multidimensional case, a plurality of configurations similar to those in FIG. 6 are provided in parallel. Alternatively, an output value for a necessary dimension can be obtained by applying the process a plurality of times.

  The three-dimensional LUT reference unit 601 in the color conversion apparatus according to the present embodiment includes a function approximating unit 601a and an error value correcting unit 601b.

  The function approximating unit 601a includes a function value calculating unit 602 and a coefficient value LUT reference unit 603, and the coefficient value LUT reference unit 603 is connected to the coefficient value LUT 606. The function value calculation unit 602 calculates a function value using a coefficient value selected using a multidimensional input value. The coefficient value LUT reference unit 603 corresponds to the coefficient value LUT 606 depending on which area of the plurality of areas (area obtained by dividing the multidimensional input value range into a plurality of areas). A numerical value is selected and passed to the function value calculation unit 602.

  The error value correcting unit 601b includes an adding unit 604 and an error value LUT reference unit 605, and the error value LUT reference unit 605 is connected to the 3D LUT 607. The adder 604 adds an error value to the function value calculated by the function value calculator 602 and outputs the result. The error value LUT reference unit 605 acquires an error value corresponding to the input value with reference to the 3D LUT 607 and passes the error value to the addition unit 604.

  The coefficient value LUT 606 is a lookup table (coefficient value storage means) holding coefficient values. The 3D LUT 607 is a lookup table (error value storage means) that holds error values corresponding to each multidimensional input value, and is hereinafter referred to as an error value LUT as necessary.

  Here, the error value is a difference between the function value calculated by the function value calculation unit 602 and the value that should be output originally.

  In FIG. 6, first, input values 901 to 903 are converted into lattice point numbers by lattice point number calculation units 904 to 906 in the same manner as the configuration of the prior art shown in FIG. 9, and are input to the three-dimensional LUT reference unit 601. Entered. Next, the interpolation calculation unit 912 performs interpolation calculation using the grid point output values 907 to 910 output from the three-dimensional LUT reference unit 601, and the final output value 913 is obtained as in FIG. is there.

  As described above, the three-dimensional LUT reference unit 601 includes the function approximation unit 601a including the function value calculation unit 602 and the coefficient value LUT reference unit 603, and the error value correction unit including the addition unit 604 and the error value LUT reference unit 605. 601b includes two parts corresponding to the calculation of the approximate function value of the principle and the reference of the error value LUT607. Further, the function approximating unit 601a is connected to a coefficient value LUT 606 for storing the coefficient value of the approximating function corresponding to each region divided into a plurality of regions in the input value range, and the error value correcting unit 601b is connected to the error value correcting unit 601b. A value LUT 607 is connected.

  First, the operation of the function approximating unit 601a will be described.

  The input three-dimensional grid point number is determined by the coefficient value LUT reference unit 603 as to which area the input value is divided as shown in FIG. 5, and an address on the coefficient value LUT 602 corresponding to the area is determined. Is calculated. Using this address, the coefficient value LUT reference unit 603 refers to the coefficient value LUT 606, obtains a coefficient value necessary for calculating the approximate function, and passes it to the function value calculation unit 602. The function value calculation unit 602 calculates an approximate function value using the input three-dimensional grid point number and coefficient value, and passes the output value to the addition unit 604 of the error value correction unit 601b.

  Next, the operation of the error value correction unit 601b will be described.

  The error value LUT reference unit 605 converts the input three-dimensional grid point number into a corresponding address on the three-dimensional LUT 607, and refers to the three-dimensional LUT 607 using this address, and an error value corresponding to the input value. And the error value is passed to the adder 604. The adder 604 adds the function value obtained from the function value calculator 602 and the error value obtained from the error value LUT reference unit 605, and outputs the result as an output value of the three-dimensional LUT reference unit 601.

  By adopting the configuration as shown in FIG. 6 and having the contents held in the three-dimensional LUT 607 not with the output value but with an error value with a narrow range, the memory capacity during actual use of the three-dimensional LUT 607 can be saved.

  In the above description, a linear function is used as an approximate function. However, a more complex function such as a quadratic function or other arithmetic functions / logical functions can be used as long as the circuit scale allows.

  In FIG. 5, the range of the input value is equally divided into two for each dimension, but the number of divisions for each dimension is arbitrary as long as the memory capacity of the coefficient value LUT 606 does not become too large, or is always equal. There is no need.

  As described above, according to the color conversion apparatus according to the present embodiment, the coefficient value LUT reference unit 603 refers to the coefficient value LUT 606 and sets the input values 901 to 903 to the input values 901 to 903. A coefficient value corresponding to the region to which the image belongs is selected, an approximate function value for the selected coefficient value and input values 901 to 903 is calculated by the function value calculation unit 602, and the calculated approximate function value and error value LUT reference unit 605 are calculated. Thus, the output value of the three-dimensional LUT 607 is obtained by adding the error values corresponding to the input values 901 to 903 obtained by referring to the error value LUT 607 by the adding unit 604. Therefore, the memory capacity of the three-dimensional LUT 607 is obtained. Can be saved.

(Second Embodiment)
Next, a second embodiment of the present invention will be described with reference to FIG.

  Among the problems to be solved by the above-described invention, this embodiment solves the increase in memory capacity and the correspondence to a plurality of bit precisions when the bit precision is raised.

  FIG. 7 is a block diagram showing the configuration of the main part of the color conversion apparatus according to the present embodiment, which corresponds to the three-dimensional LUT reference unit 601, the coefficient value LUT 606, and the three-dimensional LUT 607 in FIG. A three-dimensional LUT reference unit 701, a coefficient value LUT 705, and a three-dimensional LUT 709 are shown. Other input values, grid point number calculation unit, grid point output value, interpolation coefficient value calculation unit, interpolation calculation unit, output value, etc. Is omitted.

  The three-dimensional LUT reference unit 701 is divided into a function approximation unit 701a and an error value correction unit 701b.

  The function approximation unit 701a includes a function value calculation unit 702, a coefficient value LUT reference unit 703, and a function output accuracy selection unit 704. The coefficient value LUT reference unit 703 is connected to the coefficient value LUT 705. The function value calculator 702 calculates a function value using a coefficient value selected using a multidimensional input value. The coefficient value LUT reference unit 703 is a function corresponding to the coefficient value LUT 705 depending on which of the plurality of regions (region obtained by dividing the multidimensional input value range into a plurality of regions). A numerical value is selected and passed to the function value calculation unit 702. The function output accuracy selection unit 704 selects and outputs the output value of the function value calculation unit 702 from multi-bit accuracy.

  The error value correction unit 701b includes an addition unit 706, an addition position selection unit 707, and an error value LUT reference unit 708. The error value LUT reference unit 708 is connected to a three-dimensional LUT (3D LUT) 709. The adder 706 adds the error value to the function value calculated by the function value calculator 702 and outputs the result. The addition position selection unit 707 selects the output value of the error value LUT reference unit 708 from a plurality of bit positions, aligns the digits, and outputs the result to the addition unit 706. The error value LUT reference unit 708 acquires an error value corresponding to the input value with reference to the three-dimensional LUT 709 and passes it to the addition unit 706.

  The coefficient value LUT 705 is a lookup table (coefficient value storage means) that stores coefficient values. The three-dimensional LUT 709 is a lookup table (error value storage means) that holds error values corresponding to each multi-dimensional input value, and is hereinafter referred to as an error value LUT as necessary.

  For convenience of explanation, the bit precision of the grid point output value originally corresponding in FIG. 6 is 8 bits, and the bit precision of the corresponding grid point output value in FIG. 7 is 8 bits and 10 bits.

  First, the function approximation unit 701a of the three-dimensional LUT reference unit 701 will be described.

  The coefficient value LUT reference unit 703 refers to the coefficient value LUT 705 using the input three-dimensional grid point number value, and the processing until the coefficient value used in the approximation function is passed to the function value calculation unit 702 has been described above. This is exactly the same as in the first embodiment.

  The function value calculation unit 702 calculates the output value of the approximate function using the input value and the coefficient value, and the bit accuracy of the output value is equal to or higher than the highest bit accuracy of the corresponding output value, Here, it is 10 bits or more. Since the calculation result of the function value calculation unit 702 usually has a bit accuracy higher than the bit accuracy of the output value, this can be realized only by increasing the number of bits output by the function value calculation unit 702.

  Next, the output value of the function value calculation unit 702 is input to the function output accuracy selection unit 704, where the bit accuracy is selected to be the same as the bit accuracy of the grid point output value. For example, when 8 bits are selected as the bit precision of the grid point output value, the input value of 10 bits or more is rounded down to 8 bits or rounded (not simply rounded down, but becomes a predetermined value) The compression technique is performed and output. Further, for example, when 10 bits are selected as the bit precision of the grid point output value, the input value of 10 bits or more is rounded down or rounded to 10 bits and output. It is assumed that the bit accuracy of output is set before using the three-dimensional LUT reference unit 701. The approximate function output value having the bit precision of the grid point output value by the function output accuracy selection unit 704 is sent to the addition unit 706 of the error value correction unit 701b.

  Next, the error value correction unit 701b of the three-dimensional LUT reference unit 701 will be described.

The processing up to the point where the error value LUT reference unit 708 refers to the three-dimensional LUT (3D LUT) 709 holding the error value using the three-dimensional grid point number value is the same as that in the first embodiment. Although the same, the error value obtained by referring to the three-dimensional LUT 709 is converted by the addition position selection unit 707 before being input to the addition unit 706. The addition position selection unit 707 has a function of aligning and outputting the input error value by the number of bits set in advance before use, for example, scaling of 1/2 ⁿ times, 1 time, 2 ⁿ times Coefficient value can be set. The adder 706 is the highest corresponding output bit precision, in this example, an adder corresponding to 10 bits. For the 8-bit or 10-bit function value output from the function approximator 701, The final grid point output value is obtained by adding the error values after digit alignment output from the addition position selection unit 707.

  As described above, according to the color conversion apparatus of the present embodiment, the output bit accuracy of the function value calculation unit 702 and the calculation bit accuracy of the addition unit 706 are increased, and the function output accuracy selection unit 704 is provided. Therefore, it is possible to cope with higher output bit precision and easily switch between a plurality of bit precisions.

  Further, the error value obtained by referring to the error value LUT 709 by the error value LUT reference unit 708 can be digit-aligned by the addition position selection unit 707, so that data held in the error value LUT 709 is stored in the error value LUT 709. Even if the number of bits is constant, the error value range after digit alignment can be controlled. Further, the fact that the number of bits of data held in the error value LUT 709 is constant means that the logic for calculating the address in the error value LUT reference unit 708 does not change even if the bit precision changes. It is possible to more easily cope with switching of a plurality of bit precisions.

  In the above description, two bit precisions of 8 bits and 10 bits are given as an example for convenience of explanation, but the corresponding bit precision is not necessarily limited to two, and the number of bits is 8 as well. It is not limited to bits and 10 bits.

  In FIG. 7, an example using both the function output accuracy selection unit 704 and the addition position selection unit 707 is shown, but a configuration using only one of them can be similarly implemented.

(Third embodiment)
Next, a third embodiment of the present invention will be described with reference to FIG.

  In the present embodiment, color conversion is performed using the same principle as that described in the first embodiment.

  The basic configuration of the color conversion apparatus according to the present embodiment is the same as the configuration shown in FIG. 7 in the second embodiment described above, and will be described using the same drawing as necessary. .

  Hereinafter, the operation of the color conversion apparatus according to the present embodiment will be described with reference to the flowchart of FIG.

  The processing steps of the color conversion apparatus according to the present embodiment are roughly divided into a function approximation step S801 that is a process of the function approximation unit 701a and an error value correction step S802 that is a process of the error value correction unit 701b. Steps S801 and S802 include the following steps.

  In the function approximating step S801, first, a reference address calculating step S803 for calculating the reference address of the coefficient value LUT 705 from the upper bits of the grid point number is executed, and the input three-dimensional grid point number is the range of the input value. The reference address of the coefficient value LUT 705 is determined depending on which area the data is divided into a plurality of areas. Next, the coefficient value LUT reference unit 703 executes a coefficient value LUT reference step S804 that refers to the coefficient value LUT 705, and a coefficient value used for the approximate calculation is obtained from the coefficient value LUT 705. Next, the function value calculation unit 702 executes a function value calculation step S805 for performing a function value calculation using the grid point number and the coefficient value, and calculates an output value of the approximate function. The bit precision of the output value in step S805 is the same as or higher than the highest output bit precision corresponding to this color conversion method. Next, step S806 is executed to round the function output value to the selected bit precision (a technique of compressing to a predetermined value instead of simply truncating), and rounding the function output value to a preselected bit precision. (Technique that compresses to a predetermined value instead of simply truncating).

On the other hand, in the error value correction step S802, first, in step S807 of calculating the reference address of the error value LUT 709 from the grid point number, the address in the error value LUT 709 is calculated from the input three-dimensional grid point number. Next, the error value LUT reference unit 708 executes an error value LUT reference step S808 referring to the error value LUT 709 to obtain an error value. Next, the addition position selection unit 707 executes an addition position selection step S809 in which the output value of the error value LUT 709 is aligned with the selected addition position, and the error value is converted into a preselected scaling ^coefficient value (1/2 ⁿ (1x, ²ⁿ times, etc.)

  Finally, the addition unit 706 executes an addition output step S810 for adding and outputting the function output value and the error output value, and the approximate function output value obtained by the function approximation step 801 described above and the addition position selection step The final grid point output value is obtained by adding the error value after digit alignment obtained in S809 by the adding unit 706.

  According to the color conversion apparatus according to the present embodiment, by having the above-described steps S801 to S810, the memory capacity of the three-dimensional LUT 709 can be saved as in the first and second embodiments described above. It is possible to easily cope with higher output bit accuracy and multiple output bit accuracy.

  Note that the present invention is not limited to the present embodiment, and any one can be applied based on the subject matter of the present invention.

  Another object of the present invention is to supply a storage medium storing software program codes for realizing the functions of the above-described embodiments to a system or apparatus, and store the computer (or CPU or MPU) of the system or apparatus. Needless to say, this can also be achieved by reading and executing the program code stored in the medium.

  In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code and the storage medium storing the program code constitute the present invention.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an operating system (OS) or the like running on the computer based on the instruction of the program code However, it is needless to say that a case where the functions of the above-described embodiment are realized by performing part or all of the actual processing.

  In addition, after the program code read from the storage medium is written into a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer, the function is determined based on the instruction of the program code. It goes without saying that the CPU or the like provided in the expansion card or the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.

  The above-described program only needs to be able to realize the functions of the above-described embodiments by a computer, and the form includes forms such as object code, a program executed by an interpreter, and script data supplied to the OS. But it ’s okay.

  The recording medium for supplying the program includes, for example, RAM, NV-RAM, floppy (registered trademark) disk, optical disk, magneto-optical disk, CD-ROM, CD-R, CD-RW, DVD (DVD-ROM, DVD-RAM, DVD-RW, DVD + RW), magnetic tape, non-volatile memory card, other ROM, and the like may be used as long as they can store the program.

  Further, the program is supplied by downloading from another computer or database (not shown) connected to the Internet, a commercial network, a local area network, or the like.

It is principle explanatory drawing by the one-dimensional LUT of this invention. It is an approximate function explanatory drawing for demonstrating the principle of this invention. It is error value table explanatory drawing for demonstrating the principle of this invention. It is principle explanatory drawing by the three-dimensional LUT of this invention. It is a figure which shows an example of the input value range division | segmentation at the time of three-dimensional input for demonstrating the principle of this invention. 1 is a block diagram illustrating a configuration of a color conversion apparatus according to a first embodiment of the present invention. It is a block diagram which shows the principal part structure of the color conversion apparatus which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows operation | movement of the color conversion apparatus which concerns on the 3rd Embodiment of this invention. It is a block diagram which shows the structure of the conventional color conversion apparatus.

Explanation of symbols

601 Three-dimensional LUT reference unit 601a Function approximation unit 601b Error value correction unit 602 Function value calculation unit 603 Coefficient value LUT reference unit 604 Addition unit 605 Error value LUT reference unit 606 Coefficient value LUT
607 3D LUT (3D LUT: error value LUT)
701 Three-dimensional LUT reference unit 701a Function approximation unit 701b Error value correction unit 702 Function value calculation unit 703 Coefficient value LUT reference unit 704 Function output accuracy selection unit 705 Coefficient value LUT
706 Addition unit 707 Conversion position selection unit 708 Error value LUT reference unit 709 Three-dimensional LUT (3D LUT: error value LUT)

Claims (8)

Coefficient value storage means for storing coefficient values corresponding to each area obtained by dividing the range of the multidimensional lookup table into a plurality of areas ;
Grid point calculation means for calculating grid points in the multi-dimensional lookup table corresponding to input values;
Coefficient value reference means for selecting a corresponding coefficient value from the coefficient value storage means depending on which area of the plurality of areas the calculated lattice points belong to;
A function value calculating means for calculating a function value using a coefficient value selected for the calculated lattice point ;
Error value storage means for storing a difference between a function value calculated by the function value calculation means and a value to be output as an error value corresponding to each of the lattice points;
Error value acquisition means for acquiring the error value corresponding to the calculated grid point from the error value storage means ;
By adding the error value said error value acquisition means has acquired for function value the function value calculating means is calculated, adding means for calculating an output value of the calculated grid points,
Interpolation operation means for obtaining an output value corresponding to the input value by performing interpolation operation based on the output value calculated by the addition means . The color conversion apparatus according to claim 1, wherein a lattice point in a multidimensional lookup table corresponding to the input value is calculated based on upper bits of the input value. The color conversion apparatus according to claim 1, wherein the coefficient value is determined so as to make the error value as small as possible in the region. A control method for a color conversion device comprising coefficient value storage means for storing coefficient values corresponding to respective areas obtained by dividing a range of a multidimensional lookup table into a plurality of areas ,
A grid point calculation step of calculating grid points in the multi-dimensional lookup table corresponding to the input value;
A coefficient value reference step of selecting a corresponding coefficient value from the coefficient value storage unit according to which region of the plurality of regions the calculated grid point belongs to;
A function value calculation step of calculating a function value using a coefficient value selected for the calculated grid point ;
The color conversion device further includes error value storage means for storing a difference between the function value calculated in the function value calculation step and a value to be output as an error value corresponding to each of the grid points ,
The control method is:
An error value acquiring step of acquiring the error value corresponding to the calculated grid point from the error value storage means;
An addition step of calculating an output value at the calculated lattice point by adding the error value acquired in the error value acquisition step to the function value calculated in the function value calculation step;
A control method for a color conversion apparatus, comprising: an interpolation calculation step for obtaining an output value corresponding to the input value by performing an interpolation calculation based on the output value calculated in the addition step . 5. The method according to claim 4, wherein a grid point in the multi-dimensional lookup table corresponding to the input value is calculated based on the upper bits of the input value. 6. The method of controlling a color conversion apparatus according to claim 4, further comprising a function value output accuracy selection step of selecting and outputting the function value calculated in the function value calculation step from a plurality of bit accuracy. Any of claims 4 to 6, characterized in that it has the addition position selecting step of outputting to the adder step in the combined digits to select an error value acquired by the error value acquisition step from a plurality of bit positions A method for controlling the color conversion apparatus according to claim 1 . A program comprising computer-readable program code for realizing the color conversion device control method according to any one of claims 4 to 7 .

Images