JPH0568165A - Color converter - Google Patents

Abstract

PURPOSE:To process three-dimensional or more than optional-dimensional optional color conversion at real time by simplifying an interpolating operation method of a color conversion value for a color image signal and reducing the capacity of a table memory in a color converter to be used for a color printer, a color display device, a color TV camera, a color recognizing device, and so on. CONSTITUTION:Respective input color signals are divided into upper bit signals, lower bit signals, the upper bit signals are inputted to color conversion value difference value storing table memories 115 to 117 and a color conversion value reference value storing table memory 118 and the lower bit signals are inputted to weight coefficient supplying means 109 to 111 and a unit tetrahedron selecting means 132. Thus color conversion having no discontinuousness is obtained at real time by arithmetic means 121, 122 for executing three-dimensional interpolation based upon these color conversion value difference values, color conversion value reference values and weight coefficient values.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an application for inputting a color image and a color signal and performing arbitrary color conversion in real time, for example, a color hard copy device, a color display device, a color television camera device, and color recognition. The present invention relates to a color conversion device such as a device.

[0002]

2. Description of the Related Art Conventionally, in image processing of a monochrome image, information contained in one pixel of an image is one-dimensional information called lightness (density), and lightness conversion is so-called gamma curve conversion.
LUT (look-up table) for various non-linear curves
It was possible to convert in real time by writing in. Even if the image to be handled is a color image, it is treated as three monochrome images of R (red) plane, G (green) plane, and B (blue) plane for the purpose of performing color conversion in real time, and each is independent. It was often converted by the LUT. However, in this type of processing, the color conversion that can be handled is essentially one-dimensional processing, and R '= hR (R), G' = hG (G), B '= hB (B). Only color conversion is possible.

In color image processing, the information contained in one pixel is three-dimensional information (R, G, B), and color conversion in the original sense means three-dimensional conversion R '= fR. (R, G, B) G '= fG (R, G, B) B' = fB (R, G, B).

For example, as a technique which has become important in color image processing in recent years, in HLS conversion for converting a color represented by (R, G, B) into a hue H, a lightness L, and a saturation S,
Like H = H (R, G, B), one output is a function of three inputs and belongs to the above three-dimensional conversion. However, if it is attempted to convert these with a general-purpose table, assuming that one color is an 8-bit signal, 16 (Mbyte) is required for conversion per color.
The memory capacity of e) is also required. Therefore, conventionally, there is required hardware that can perform three-dimensional color conversion universally for arbitrary color conversion and in real time. Therefore, at present, new hardware is designed for each purpose in a real-time color image processing device.
A general-purpose real-time color conversion device has not been realized.

However, color signals for color correction for color hard copy and color scanner are divided into a plurality of color spaces, a plurality of pieces of color correction information located at the vertices thereof are selected, weighted and interpolated and output. There is an example of an interpolation method (Japanese Patent Publication No. 58-16180). In this example, a unit cube, which is a basic solid in a three-dimensional color signal space, is set in the interpolation process, and this unit cube is divided into a plurality of tetrahedra to simplify the interpolation calculation.

By using this as a color conversion device, it is possible to perform non-linear free color conversion at high speed and while maintaining the gradation of an image, such as changing the color of a specific color in a color space. However, in the example of the previous patent, since the output value for the input value (the value of the second signal system in the patent) is stored in the memory device used for interpolation, when obtaining the interpolation value for the input value, There was a limitation that the four coefficients must be multiplied by the known values at the corresponding vertices of the four tetrahedra, and each added, and that the memory device addresses form a cube.

[0007]

First, in order to construct a high-speed color conversion device, interpolation calculation of color conversion values must be performed at high speed, and multiplication and addition required for interpolation are performed in parallel. There is a need. However, the interpolation method of the above patent requires four multiplications because there are four multiplications to be performed in parallel. In actual hardware, there is a demand to reduce the number of multipliers to 3 for simplification.

Secondly, in the above patent, the table memory stores the second signal system which is the output value to be obtained with respect to the first signal system which is the input value. It is necessary to directly allocate the dynamic range to the storage bit width of the table memory. For example,
Even when the value of the second signal system does not fluctuate as much as that of the first signal system, the second signal system is forced to store a very large numerical value as its absolute value. This does not mean that the memory is effectively used, and it is hard to say that it is the best method when the storage capacity is limited.

Thirdly, it cannot be said that each axis in the actual color space reflects the human visual characteristics evenly, and in the normal red, green, and blue color spaces, the green axis is particularly visual. There are consequences that are important. Therefore, if the total number of bits is constant, the interpolation accuracy can be increased by allocating more bits to the green axis, and the overall interpolation accuracy can be improved. From this point of view, it is desirable that the unit interpolation section of the color space is a rectangular parallelepiped, but the above-mentioned patent has a limitation that it is a cube.

The present invention solves the above-mentioned problems of the prior art, and provides an apparatus capable of processing in real time any color conversion whose inputs and outputs are three-dimensional or more than arbitrary dimensions by an interpolation method without discontinuity. The purpose is to do.

[0011]

According to the present invention, a set of three-dimensional or arbitrary-dimensional color signals is input, and the number of bits is different for each input signal, or is divided into an upper signal and a lower signal based on an equal bit distribution. Then, the color conversion value difference value storage table memory selected by the pair of higher-level signals, and the color conversion value reference value storage table memory selected by the pair of higher-level signals, and the weighting coefficient selected by the pair of lower-level signals. Providing means, a plurality of unit tetrahedron selecting means obtained by dividing a unit cube or unit rectangular parallelepiped in a color space selected by the higher-order signal, and color conversion value differences at a plurality of vertices of the unit tetrahedron An arithmetic means for performing three-dimensional or arbitrary-dimensional interpolation of the color space from the value and the color conversion value reference value and the weighting coefficient value; an arithmetic function setting means for confirming the arithmetic value of the arithmetic means; Note When, in which it said provided a selector for switching signals of any bit width with the output of the external memory in the lower signal, and bit manipulation means for performing a shift only the arbitrary bit width.

[0012]

According to the present invention, the value stored in the memory is changed from the value of the second signal system at each vertex of the tetrahedron to the difference value of the second signal system at one reference vertex of the tetrahedron according to the above configuration. Next, the interpolation method is changed to three multiplications of the three interpolation coefficients and the second signal system difference values at the three vertices and addition of the second signal system reference value. By this means, the number of multipliers required for interpolation can be reduced to three.

At the same time, the second signal system differential value storage memory stores a small absolute value of the signal difference value, so that the memory can be efficiently used as a result. Further, each axis of the input color space is divided into its upper signal and lower signal with different bit allocations, and a rectangular parallelepiped in the color space is used as the basic unit of interpolation, so that interpolation suitable for human visual characteristics is performed.

[0014]

The concept of the present invention will be described below. First, the signal interpolation method of Japanese Patent Publication No. 58-16180 will be described. For simplicity, the case where the input space is one-dimensional is shown. Figure 2
In, the output value (P) at an arbitrary point P included in the unit interpolation section AB is obtained by using the output values (A) and (B) at A and B and the two coefficients a and b.

[0015]

[Equation 1]

However,

[0017]

[Equation 2]

[0018] In this case, the multiplication is required twice. So, another interpolation formula

[0019]

[Equation 3]

Introduce That is, if the output value reference value (A) and the output value difference value [(B)-(A)] are stored at the input points A and B, the coefficient b for (B) is used (Equation 3). As described above, the multiplication can be performed by one interpolation.

FIG. 3 shows a case where the same thing is expanded three-dimensionally. The basic interpolation section in three dimensions is a tetrahedron ABCD, and the output values at each vertex are (A) (B) (C).
(D). In the signal interpolation method of the above patent, by storing the output value at each vertex in the memory, the coefficient a,
Using b, c, d, the internal point (P),

[0022]

[Equation 4]

However, if this is the case, four multiplications are required. Therefore, let A be the reference vertex of the four vertices, and
The output value reference value (A) is stored in B, C, and D, and the difference value between the original output value and the output value reference value [(B)-
(A)], [(C)-(A)], [(D)-(A)] are accumulated,

[0024]

[Equation 5]

Interpolation can be performed by using three multiplications as described above.
According to the present invention, free color conversion can be performed at high speed by using a simple interpolation method as in the above (Equation 5).

Next, a method of dividing an input color space into a tetrahedron according to the present invention will be described. Input color (x, y, z)
As an 8-bit numerical value represented by
1, y1, z1), the lower signals are (x2, y2, z
2) is written. Assuming that the upper signal takes 3 bits of each signal, the upper signal is a unit solid to which the input color belongs in the input color space created by (x, y, z), in this case, a unit cube is 512 (= 1 out of 8x8x8)
The lower signal has a role of individually selecting, and the lower signal has a role of indicating the existence 1 in the unit solid selected by the upper signal.
FIG. 4 shows how the entire color space is divided into unit cubes, in this case, unit cubes. In FIG. 5, one unit cube is divided into six unit tetrahedra (unit 0 to unit 5). Indicated. Which of the unit tetrahedrons the given color signal belongs to is determined as shown in (Table 1) by the relational expression of (x2, y2, z2).

[0027]

[Table 1]

Now, when the unit tetrahedron to which the input color belongs is determined, the interpolation formula becomes (Equation 1).

[0029]

[Equation 6]

Therefore, as shown in FIG. 6, Δ is the volume of the unit tetrahedron, and Δi is the vertex P1, P2, P3, P of the unit tetrahedron.
New 4 created by connecting 3 of 4 and input color point P
Individual small tetrahedrons are made and the volume of the small tetrahedron not including the point Pi is represented, and Di is the color conversion value difference value at the point Pi,
O1 indicates the color output value at the reference vertex P1.

If the above-described upper signal and lower signal division is performed with different bit allocation for each color, the color space can be divided into a set of rectangular parallelepipeds, and thereafter, interpolation can be performed by almost the same procedure. For example, an input color signal may be a high-order signal (x1, y1, z1) to a low-order signal (x2, y2, z) with non-uniform accuracy.
Divide into 2). Where x1, y1, z1 are 3, 4,
2 bits, x2, y2, and z2 are 5, 4, and 6 bits. In this case, the unit cube is 32, 16,
It has a length of 64, which is 8, 16, 4 in each axial direction.
It will be lined up individually. This bit allocation is one example, but the role of selecting one unit cube from 512 (= 8 × 16 × 4) unit cubes in the color space created by (x, y, z) for the upper signal And the lower signal indicates the position in the unit cube selected by the upper signal. FIG. 7 shows a state in which the entire color space is divided into unit cubes, and FIG. 8 shows a method of dividing one unit cube into six unit tetrahedra (unit 0 to unit 5).

Specific examples of the present invention will be described below.
This will be described with reference to FIG. FIG. 1 is a block connection diagram of a color conversion apparatus according to an embodiment of the present invention. In FIG. 1, only the portion where the output signal R ′ is generated from the input signals (R, G, B) is shown. Input signal 101-10
3 is gamma converted by the gamma conversion table memories 104 to 106. By this gamma conversion, it is possible to select in what color space the grid points storing the color conversion values are set. For example, when a log curve for density conversion is set in the gamma conversion table memory, the density of the density conversion results in a close density of lattice points in the shadow portion of the image, resulting in good interpolation characteristics in the shadow portion. The gamma-converted input signal is represented by (x, y, z). Input signal 12
5 to 127 are higher-order signals (x
1, y1, z1) and lower signals (x2, y2, z2). The high-order signal and the low-order signal are grouped together, respectively.
In the figure, they are represented by (107) and (108), respectively. The upper signal group 107 and the lower signal group 108 may have non-uniform bit division. Further lower signals (x2, y2, z2)
3 types of arbitrary bit width signals (x21, y21,
z21) are respectively taken out from the input signals (125-127). These three types of signals are collectively called an extended signal group in FIG. 1 and are represented by (143). The lower signal group 108 is input to the weighting factor providing means 109 to 111. With these three weighting factor providing means, lower order signals (x2, y2, z2)
When you enter, the unit 4 to which it belongs in the unit cube
Determine one face piece (Unit 0-Unit 5), 4
Weighting factors W2-W4 when using vertices are provided.
This weight coefficient is originally a numerical value of 0 to 1, but it is 0 to 1.
The signals 128 to 130 are output as numerical values of 255. The lower signal group 108 is also the unit tetrahedron selecting means 13
2 is also input, and one of the belonging unit tetrahedrons is selected from 6 units 0 to 5 and is output as a unit selection signal 133 of a 3-bit numerical value from 0 to 5.

On the other hand, the upper-level signal group 107 and the unit selection signal 133 are transmitted via the address bus interface 113 to the color conversion value difference value storage table memories 115 to 117.
Entered in. Similarly, the upper signal group 107 is also input to the table memory 118 for storing the color conversion value reference value. This is because in the method of division into the unit tetrahedron shown in FIG. 8, one end point of the diagonal axis of the unit solid is set at the reference vertices of the six unit tetrahedrons, and This is because the unit can share the reference vertex and the unit selection signal 133 input to (118) can be omitted.

Further, the extension signal group 143 is stored in the external memory 14
It is input to the selector 141 which selects the output of 0. When the selector 141 selects the external memory 140, the information of the external memory 140 is transmitted to the address interface 113 by the selector output signal 144. The usage of the external memory is freely decided according to the purpose. For example, as shown in FIG. 9, a plurality of sets of addresses of the image data address and the color conversion value difference value storage table memories 115 to 117 and the color conversion value reference value storage table memory 118 are stored in the external memory 140 at the same time. The color conversion value difference value storage table memories 115 to 117 and the color conversion value reference value storage table memory 118 have, for example, T1 and T shown in FIG.
2 ,. ． ． ． The color conversion information is stored like Tn + 1. However, the address of the image data is generated in the image specific area determination unit. If you configure the external memory in this way,
Color conversion processing can be individually performed according to the address of the table memory. In the case of FIG. 9, color conversion according to the contents of the color conversion table T1 is performed in the area from address Y1 to address Y2, and similarly, address Y2 to address Y3.
The color conversion table T2 is performed in the area of (3), and the color conversion according to the contents of the color conversion table T3 is performed in the area of addresses Y3 to Y4. On the other hand, when the selector 141 selects the extension signal group 143, this extension signal group 143 is used as the selector output signal 144 in the address interface 113.
Is input to. At this time, the selector output signal 144 is used together with the higher-order signal group 107 to improve the interpolation accuracy. For example, when 3 bits are assigned to the upper signal group 107 and 5 bits are assigned to the lower signal group 108 as shown in FIG. 10 and the bit width of the extension signal is 1 bit, the bit width of the upper signal group 107 is 4 bits and The bit width of the signal group 108 is also 4 bits. Therefore, the extended signal group 143 causes the higher-order signal group 107 to have eight times as much information amount (twice for each RGB color) as compared with the color conversion value difference value storage table memories 115 to 117 and the color conversion value reference value storage table memory 118.
Will be told. At the same time, the memory capacity of the color conversion table memory is also increased eight times. Therefore, when the bit width of the extension signal group 143 is set to 1 bit, the number of unit rectangular parallelepipeds, which is a basic unit of interpolation, is increased by 8 times, and the physical interpolation accuracy is improved by 8 times. On the other hand, the bit width of the lower-order signal is reduced from 5 bits to 4 bits, but the weight coefficient providing means 109-111 are set to 0.
Since the weighting coefficient is output as a numerical value of ~ 255, the lower signal group 108 must be compensated by the reduced bit width. Therefore, the bit operating means 142 performs bit shift by the bit width of the extension signal, and inserts 1 or 0 by the shift amount from the lowest.

In this way, the extension signal 143 and the external memory 1
By selectively using 40 for the selector 141 and the bit operating means 142, it is possible to realize a variety of color conversion functions and an improvement in interpolation accuracy.

The address interface 113 addresses when downloading data from the host computer to the color conversion value difference value storage table memories 115 to 117 and the color conversion value reference value storage table memory 118 and when performing color conversion. The bus 114 and the upper signal 107 are switched. When downloading data, all combinations of the higher-order signal 107 and the unit selection signal 133 are generated as addresses in the color conversion value difference value storage table memory and the color conversion value reference value storage table memory, and are determined at that time. The color conversion value reference value O1 and the color conversion value difference values D2-D4 at the unit tetrahedron reference vertex are calculated, and the data is loaded. Next, at the time of executing color conversion, the signal 137 of the color conversion value reference value O1 and the signals 134 of the color conversion value difference values D2 to D4 at the four vertices selected from the higher-level signal 107 and the unit selection signal 133 signal. 136 is output. The data bus interface 119 changes the data flow direction between the time of downloading data to the table memory for storing color conversion value reference values and the table memory for storing color conversion value difference values and the time of executing color conversion. The weighting factors W2 to W4, the color conversion value reference value O1 and the color conversion value difference values D2 to D4 for the three vertices are executed by the multiplier 121 and the adder 122 (Equation 1). However, since the weighting factor is multiplied by 255 from the original value, the output of the adder is finally shifted by 8 bits to calculate 1/255.
Get a bit color signal. Obtained color conversion interpolation value 138
Is output value R '(124) via the output LUT 123.
To generate. Various nonlinear curves can be loaded in the output LUT 123 depending on how the color conversion device is used.

In the embodiments shown here, the bit precision, bit allocation, etc. are merely examples and other numerical values may be used. Further, in the present embodiment, since the calculation data is complicated, the multiplier 121-0 is provided so that the color conversion interpolation value at each calculation stage can be compared with the simulation expected value.
.About.121-2 and adders 122-0.about.122-2 are provided with arithmetic function setting means 139 capable of arbitrarily setting respective arithmetic functions.

[0038]

As described above, the present invention has three inputs and three outputs.
It is possible to process in real time an arbitrary color conversion having dimensions or more than arbitrary dimensions by an interpolation method having no discontinuity.

At this time, since the conventional three-dimensional color signal interpolation method stores the output value with respect to the input value in the memory content,
There is a problem that the multiplication is used four times in the interpolation formula, but in the present invention, since the reference signal and the difference signal of the output color signal are stored in the memory, the multiplication of the interpolation formula is reduced to three times. It has the great advantage that

Further, since the dynamic range of the numerical values stored in the output value difference value memory becomes small, the effect of reducing the storage bit width or the effect of effectively utilizing the bit width can be expected. Further, there is an advantage that the table memory can be effectively used for visual characteristics by dividing the input color space into unit cubes. Even if the color conversion calculation formula changes, the color conversion value storage table can be rewritten, and since it has a complete interface with the host computer, it is possible to immediately respond to it The effect is extremely large in the color image processing field.

[Brief description of drawings]

FIG. 1 is a block connection diagram of a color conversion device according to an embodiment of the present invention.

FIG. 2 is a conceptual diagram showing a one-dimensional interpolation method which is a premise of the present invention.

FIG. 3 is a conceptual diagram showing stored contents at vertices of a unit tetrahedron in three-dimensional interpolation which is a premise of the present invention.

FIG. 4 is a conceptual diagram in which the entire color space that is the premise of the present invention is divided into unit solids.

FIG. 5 is a diagram showing a method of dividing a unit solid body, which is a premise of the present invention, into unit tetrahedrons (units 0 to 5).

FIG. 6 is a conceptual diagram in which a unit tetrahedron, which is the premise of the present invention, is divided into facets.

FIG. 7 is a conceptual diagram in which the entire color space that is the premise of the present invention is divided into unit cubes.

FIG. 8 is a conceptual diagram showing a method of dividing a unit rectangular parallelepiped into a unit tetrahedron (units 0 to 5) which is a premise of the present invention.

FIG. 9 is a conceptual diagram of use of an external memory, which is a main part of a color conversion device in one embodiment of the present invention.

FIG. 10 is a diagram showing bit allocation of an upper signal and a lower signal by an extension signal of the same color conversion device.

[Explanation of symbols]

104 gamma conversion table memory (red) 105 gamma conversion table memory (green) 106 gamma conversion table memory (blue) 109 weight coefficient W2 providing means 110 weight coefficient W3 providing means 111 weight coefficient W4 providing means 113 address bus interface 115 color conversion value Table memory for storing difference value D2 (rewritable) 116 Color conversion value table memory for storing difference value D3 (rewritable) 117 Color conversion value difference value D4 storage table memory (rewritable) 118 Color conversion value difference value O1 storage Table memory (rewritable) 119 Data bus interface 121-0 Multiplier 121-1 Multiplier 121-2 Multiplier 121-3 Multiplier 122-0 Adder 122-1 Adder 122-2 Adder 123 Output value LUT Thirteen 2 unit tetrahedron selecting means 139 arithmetic function setting means 140 external memory 141 selector 142 bit operating means 145 image specific area determination unit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Teruo Kuro Inventor Matsuo Giken Co., Ltd. 3-10-1 Higashisanda, Tama-ku, Kawasaki City, Kanagawa Prefecture

Claims (2)

[Claims] 1. A three-dimensional or arbitrary-dimensional color signal set is input, and the input signal is divided into higher-order signals and lower-order signals according to different bit number distribution or equal bit distribution, and selected by a higher-order signal set. The table memory for storing the color conversion value reference value, which is the same as the table memory for storing the color conversion value difference value, and the weighting factor providing means selected by the group for the lower signal, and the upper signal, A plurality of unit tetrahedron selecting means obtained by dividing a unit cube or unit rectangular parallelepiped in the color space to be selected, and color conversion value difference values and color conversion value reference values at a plurality of vertices of the unit tetrahedron Arithmetic means for performing three-dimensional or arbitrary-dimensional interpolation of the color space from the weighting factor values, arithmetic function setting means for confirming arithmetic values of the arithmetic means, and an external memory,
A color conversion device comprising: a selector that switches between a signal of an arbitrary bit width in the lower-order signal and an output of the external memory; and a bit operation unit that shifts by the arbitrary bit width. 2. When the input signal is three-dimensional, the unit solid designated by the higher-order signal is divided into a plurality of unit tetrahedrons sharing the diagonal axis, and color is provided at one vertex position forming the diagonal axis. The color conversion device according to claim 1, wherein the conversion value difference value is stored, and the color conversion value difference value is stored at the remaining vertex positions of each tetrahedron.