CN103106888A - Three-dimensional color gamut boundary description method - Google Patents

Abstract

The invention discloses a three-dimensional color gamut boundary description method, which is implemented according to the following steps: step 1, setting electronic color target data; step 2, adjusting the environment settings around the display; step 3, extracting the color gamut boundary of the device from all Select the data on the outer surface of the cube from the sample points, and then extract the RGB values of the six circumscribed surfaces; step 4, extract the discrete points of the color gamut boundary of the RGB space from all the characteristic modeling data, and one circumscribed point of the RGB cube Generate 128 triangles on the surface; step 5, describe the appearance of the RGB color gamut; step 6, convert the RGB values of all triangle vertices into CIELAB values obtained by measurement, and perform closed patches on all triangles in the CIELAB space to obtain the appearance of the color gamut Figure, Serve. The method of the invention has simple and convenient steps and fast operation, and can accurately extract the color gamut boundary of the equipment type.

Description

A 3D Color Gamut Boundary Description Method

technical field

The invention belongs to the technical field of image color processing, and relates to a method for describing a three-dimensional color gamut boundary.

Background technique

In the currently commonly used three-dimensional color gamut boundary description algorithm, Herzog regards the printer color gamut as a deformed hexahedron, based on the corresponding 8 vertices in the CIELAB space, and describes it by analyzing the degree of deformation of 12 sides and 6 faces The degree of color gamut deformation, which is more accurate in describing equipment with a relatively regular color gamut, but for ordinary equipment, its color gamut regularity is poor. M. Mahy proposed a color gamut analysis model based on the Neugebauer equation, which decomposes the n-color color rendering system into multiple n-1 color systems, but limited by the model, it is not applicable to devices based on the RGB additive color method. Huang Qingmei of Beijing Institute of Technology also regards the device color gamut as a closed solid with six curved surfaces as boundary surfaces, and uses Zernike polynomials to represent the boundary surfaces. Dr. Wang Yong from Zhejiang University proposed a TVHOP bivariate high-order polynomial analytical model for the calculation of the device color gamut boundary. When describing the color gamut of a CRT display, the GOG model and edge (surface) judgment criteria are first used to calculate the color in the RGB color space. Gamut boundary, and then use bivariate high-order polynomials to describe each surface of the color gamut boundary in the CIELAB color space. These models use complex analytical formulas when describing the color gamut boundary, which is not convenient to use. In addition, Raja B. proposed to use the improved convex hull algorithm to describe the color gamut, which extracts the values in the boundary part of the discrete sample points through the convex hull algorithm, but because some internal points are often regarded as boundary points, the calculated color gamut Boundaries generally exceed 10% of the actual color gamut. Morovic proposed a partition maximization algorithm to calculate the color gamut. The advantage of this algorithm is that it is fast and convenient, but often there are boundary points that cannot be found in some partitions and need to be reconstructed by interpolation, which also introduces errors to a certain extent.

Contents of the invention

The purpose of the present invention is to provide a three-dimensional color gamut boundary description method, which solves the problem in the prior art that it is difficult to quickly and accurately obtain and visualize the color gamut boundaries of RGB or CMY color devices.

The technical solution adopted in the present invention is a method for describing the boundary of a three-dimensional color gamut, which is implemented according to the following steps:

Step 1. Set the electronic color target data

Adjust according to the output and measurement time controlled by the user. The default setting is that the R, G, and B single channels are evenly divided into 9 levels, that is, the discrete value range of the single-channel drive value is [0 32 64 96 128160 192 224 255 ], if the driving signal range of the user equipment is 0~1, then the value range of the above driving value is [0 32 64 96 128 160 192 224 255]/255, in the established electronic color target, the RGB driving value of the color block is according to The above value ranges are mixed with gradients, which contain a total of 9×9×9=729 color blocks;

Step 2. Adjust the environment settings around the monitor

Set the color temperature to 6500K, and set the Gamma value; after preheating, display all the color blocks on the color target, measure the corresponding CIELAB value with a spectrophotometer, and store the RGB value and CIELAB value of each color block correspondingly;

Step 3. In the RGB color space, the appearance of its color gamut is a regular cube structure. When extracting the boundary of the device color gamut, it is necessary to select the data on the outer surface of the cube from all sample points, and use the default color target data in step 1. Extract the boundary information of the RGB cube. For the color information of the discrete boundary points of all 12 sides, among the three channels, the color value of one channel changes gradually between 0 and 255, while the other two channels are fixed at 0 or 255. Value status, so each of the extracted 12 edges contains 9 boundary points, and the corresponding RGB values are:

L1: R=[0 32 64 96 128 160 192 224 255]; G=0; B=0,

L2: R=[0 32 64 96 128 160 192 224 255]; G=0; B=255,

L3: R=[0 32 64 96 128 160 192 224 255]; G=255; B=0,

L4: R=[0 32 64 96 128 160 192 224 255]; G=255; B=255,

L5: G=[0 32 64 96 128 160 192 224 255]; R=0; B=0,

L6: G=[0 32 64 96 128 160 192 224 255]; R=0; B=255,

L7: G=[0 32 64 96 128 160 192 224 255]; R=255; B=0,

L8: G=[0 32 64 96 128 160 192 224 255]; R=255; B=255,

L9: B=[0 32 64 96 128 160 192 224 255]; G=0; R=0,

L10: B=[0 32 64 96 128 160 192 224 255]; G=0; R=255,

L11: B=[0 32 64 96 128 160 192 224 255]; G=255; R=0,

L12: B=[0 32 64 96 128 160 192 224 255]; G=255; R=255,

In addition, the RGB color gamut contains 6 circumscribing surfaces. In the boundary point information of these 6 circumscribing surfaces, one channel value is at the extreme value, and the other two channel values are gradually changing. Therefore, each of the 6 circumscribing surfaces is Contains 81 boundary points, and the RGB values of the 6 circumscribed surfaces are:

F1: R=0; G=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160 192224 255],

F2: R=255; G=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160192 224 255],

F3: G=0; R=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160 192224 255],

F4: G=255; R=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160192 224 255],

F5: B=0; G=[0 32 64 96 128 160 192 224 255]; R=[0 32 64 96 128 160 192224 255],

F6: B=255; G=[0 32 64 96 128 160 192 224 255]; R=[0 32 64 96 128 160192 224 255];

Step 4, using the data information in step 3, extract the discrete points of the color gamut boundary in RGB space from all the characteristic modeling data, and use the 81 boundary points in each circumscribed surface to first divide each of the four adjacent The boundary points form a square, and then use the diagonal line formed by the upper left corner and the lower right corner of each square to divide it into two triangles, so that a total of 128 triangles are generated on one circumscribed surface of the RGB cube;

Step 5. Describe the appearance of the RGB color gamut

For the triangles generated by the division of the 6 external surfaces of the RGB cube in step 4, fill them with color. When filling each triangle with color, according to the information of the three vertices of each triangle, perform color filling in the RGB space, and its color The value is represented by the following formula:

T _i =(A _i +B _i +C _i )/3,

Where i represents the color channel R, G or B of the vertex, Ti is the color value to be filled, A, B, and C are the three vertices of the triangle, and all triangles are closed and filled to obtain the appearance of the RGB cube color gamut;

Step 6. Use the RGB cube color gamut boundary points extracted in step 4 to search in the measurement data stored in step 2 to find out the corresponding CIELAB values, describe the discrete forms of these boundary points in CIELAB space, and then use RGB The 6 circumscribed surfaces in the cube and 128 triangles in each circumscribed surface are constructed sequentially, and the RGB values of all triangle vertices are converted into CIELAB values obtained by measurement, and all triangles are closed and pasted in CIELAB space to obtain the color Domain appearance map, that is.

The beneficial effects of the present invention are: considering the deficiency of the above algorithm, first find the boundary of the device space color gamut, and then convert it to the thinking in the chromaticity space CIELAB, and because the device space color gamut of RGB or CMY equipment is different from the CIELAB color Corresponding to the color gamut of the color space, the steps are simple and the calculation is fast, and the color gamut boundary of the device type can be accurately extracted.

Description of drawings

Fig. 1 is the discrete boundary points in the RGB space in the three-dimensional color gamut boundary description method of the present invention;

Fig. 2 is the triangular division of the RGB color gamut surface in the inventive method;

Fig. 3 is the RGB cube color gamut appearance in the inventive method;

Fig. 4 is the color gamut of the RGB device in the CIELAB space in the method of the present invention.

Detailed ways

The working principle of the three-dimensional color gamut boundary description method of the present invention is: for three-color color equipment, take RGB display as an example, its equipment color space is RGB, and its standard chromaticity space is selected as CIELAB, first make a certain number of electronic color blocks , according to the display of the RGB driving value in each color block in the monitor, record the corresponding CIELAB value by measuring; since the three channels of R, G, and B are independent of each other, that is, the color value of any single channel cannot It is replaced by another one or two channels, so it can be regarded as the corresponding color gamut boundary in the device color space, which corresponds to the color gamut boundary in the CIELAB standard chromaticity space.

The three-dimensional color gamut boundary description method of the present invention is implemented according to the following steps:

Step 1. Set the electronic color target data

Adjust according to the output and measurement time controlled by the user. The default setting is that the R, G, and B single channels are evenly divided into 9 levels, that is, the discrete value range of the single-channel drive value is [0 32 64 96 128160 192 224 255 ], if the driving signal range of the user equipment is 0~1, then the value range of the above driving value is [0 32 64 96 128 160 192 224 255]/255, in the established electronic color target, the RGB driving value of the color block is according to The above value ranges are mixed with gradients, which contain a total of 9×9×9=729 color blocks.

Step 2. Adjust the environment settings around the monitor

Set the color temperature to 6500K, and set the Gamma value to 1.8 for the Apple system and 2.2 for the PC; preheat for about 30 minutes, display all the color blocks on the color target, measure the corresponding CIELAB value with a spectrophotometer, and set each The RGB value and CIELAB value of each color block are stored correspondingly.

Step 3, as shown in Figure 1, in the RGB color space, it can be seen from the positions describing all the color blocks, the appearance of the color gamut is a regular cubic structure, when extracting the boundary of the color gamut of the device, it is necessary to select the points in the cube from all the sample points For the data of the outer surface, use the default color target data in step 1 to extract the boundary information of the RGB cube. For the color information of the discrete boundary points of all 12 sides, among the three channels, the color value of one channel is between 0 and 255 Gradient, while the other two channels are fixed at the extreme value state of 0 or 255, so each of the extracted 12 sides contains 9 boundary points, and the corresponding RGB values are:

L1: R=[0 32 64 96 128 160 192 224 255]; G=0; B=0,

L2: R=[0 32 64 96 128 160 192 224 255]; G=0; B=255,

L3: R=[0 32 64 96 128 160 192 224 255]; G=255; B=0,

L4: R=[0 32 64 96 128 160 192 224 255]; G=255; B=255,

L5: G=[0 32 64 96 128 160 192 224 255]; R=0; B=0,

L6: G=[0 32 64 96 128 160 192 224 255]; R=0; B=255,

L7: G=[0 32 64 96 128 160 192 224 255]; R=255; B=0,

L8: G=[0 32 64 96 128 160 192 224 255]; R=255; B=255,

L9: B=[0 32 64 96 128 160 192 224 255]; G=0; R=0,

L10: B=[0 32 64 96 128 160 192 224 255]; G=0; R=255,

L11: B=[0 32 64 96 128 160 192 224 255]; G=255; R=0,

L12: B=[0 32 64 96 128 160 192 224 255]; G=255; R=255,

In addition, the RGB color gamut contains 6 circumscribing surfaces. In the boundary point information of these 6 circumscribing surfaces, one channel value is at the extreme value, and the other two channel values are gradually changing. Therefore, each of the 6 circumscribing surfaces is Contains 81 boundary points, and the RGB values of the 6 circumscribed surfaces are:

F1: R=0; G=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160 192224 255],

F2: R=255; G=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160192 224 255],

F3: G=0; R=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160 192224 255],

F4: G=255; R=[0 32 64 96 128 160 192 224 255]; B=[0 32 64 96 128 160192 224 255],

F5: B=0; G=[0 32 64 96 128 160 192 224 255]; R=[0 32 64 96 128 160 192224 255],

F6: B=255; G=[0 32 64 96 128 160 192 224 255]; R=[0 32 64 96 128 160192 224 255].

Step 4. Using the data information in Step 3, extract the discrete points of the color gamut boundary in RGB space from all the characteristic modeling data, and use the 81 boundary points in each circumscribed surface. The connection method adopted is to first connect Every four adjacent boundary points form a square, and then use the diagonal line formed by the upper left corner and the lower right corner of each square to divide it into two triangles, so that one circumscribed surface of the RGB cube produces 128 triangles , refer to Figure 2.

Step 5. Describe the appearance of the RGB color gamut

For the triangles generated by the division of the 6 external surfaces of the RGB cube in step 4, fill them with color. When filling each triangle with color, according to the information of the three vertices of each triangle, perform color filling in the RGB space, and its color Values are represented by the following formula:

T _i =(A _i +B _i +C _i )/3,

Where i represents the color channel R, G or B of the vertex, Ti is the color value to be filled, A, B, and C are the three vertices of the triangle, and all the triangles are closed and filled to obtain the RGB cube color as shown in Figure 3 domain appearance.

Step 6. Use the RGB cube color gamut boundary points extracted in step 4 to search in the measurement data stored in step 2 to find out the corresponding CIELAB values, describe the discrete forms of these boundary points in CIELAB space, and then use RGB The 6 circumscribed surfaces in the cube and 128 triangles in each circumscribed surface are constructed sequentially, and the RGB values of all triangle vertices are converted into CIELAB values obtained by measurement, and all triangles are closed and pasted in CIELAB space to obtain the color Domain appearance, see Figure 4, and it's done.

The three-dimensional color gamut boundary description method of the present invention extracts the color gamut boundary points of the device in the RGB color space, and then calculates the CIELAB values of these boundary points by using a measurement method or a color space conversion algorithm, and then in the CIELAB color space Connect and close to obtain the three-dimensional continuous color gamut in the standard chromaticity space, and that's it.

Claims (2)

1. a three-dimensional gamut contour description method, is characterized in that, implements according to following steps:

Step 1, setting electronics look target data

Adjust according to output and the Measuring Time of the required control of user, default setting is that R, G, B single channel evenly are divided into 9 grades, the discrete span that is the single channel motivation value is [0 32 64 96 128 160 192 224 255], if the driving range of signal of subscriber equipment is 0 ~ 1, above motivation value span is [0 32 64 96 128 160 192 224 255]/255, in the electronics look target of setting up, color lump RGB motivation value comprises 9 * 9 * 9=729 color lump altogether according to above span mixed gradient;

Environment setting around step 2, adjustment display

The setting colour temperature is 6500K, sets the Gamma value; After preheating, show all color lumps on the look target, with CIELAB value corresponding to spectrophotometer measurement, with rgb value and the CIELAB value corresponding stored of each color lump;

step 3, in the RGB color space, its colour gamut outward appearance is the cube structure of rule, need to select the data that are in the cube outside surface during extraction equipment gamut boundary from all sample points, adopt the look target data of acquiescence in step 1, extract the cubical boundary information of RGB, divergent boundary point colouring information for all 12 limits, in three passages, one of them passage color value is 0 ~ 255 gradual change, and two other passage all fixedly is in 0 or 255 extreme value state, so on 12 limits of extracting, every limit comprises 9 frontier points, corresponding rgb value is respectively:

L1：R=[0?32?64?96?128?160?192?224?255]；G=0；B=0，

L2：R=[0?32?64?96?128?160?192?224?255]；G=0；B=255，

L3：R=[0?32?64?96?128?160?192?224?255]；G=255；B=0，

L4：R=[0?32?64?96?128?160?192?224?255]；G=255；B=255，

L5：G=[0?32?64?96?128?160?192?224?255]；R=0；B=0，

L6：G=[0?32?64?96?128?160?192?224?255]；R=0；B=255，

L7：G=[0?32?64?96?128?160?192?224?255]；R=255；B=0，

L8：G=[0?32?64?96?128?160?192?224?255]；R=255；B=255，

L9：B=[0?32?64?96?128?160?192?224?255]；G=0；R=0，

L10：B=[0?32?64?96?128?160?192?224?255]；G=0；R=255，

L11：B=[0?32?64?96?128?160?192?224?255]；G=255；R=0，

L12：B=[0?32?64?96?128?160?192?224?255]；G=255；R=255，

In addition, the RGB colour gamut comprises 6 outer junctions, in the frontier point information in these 6 outer junctions, all that a channel value is in extreme value, two other channel value gradual change, so in 6 outer junctions, each outer junction all includes 81 frontier points, the rgb value of 6 outer junctions is respectively:

F1：R=0；G=[0?32?64?96?128?160?192?224?255]；B=[0?32?64?96?128?160?192?224?255]，

F2：R=255；G=[0?32?64?96?128?160?192?224?255]；B=[0?32?64?96?128?160?192?224?255]，

F3：G=0；R=[0?32?64?96?128?160?192?224?255]；B=[0?32?64?96?128?160?192?224?255]，

F4：G=255；R=[0?32?64?96?128?160?192?224?255]；B=[0?32?64?96?128?160?192?224?255]，

F5：B=0；G=[0?32?64?96?128?160?192?224?255]；R=[0?32?64?96?128?160?192?224?255]，

F6：B=255；G=[0?32?64?96?128?160?192?224?255]；R=[0?32?64?96?128?160?192?224?255]；

Step 4, utilize the data message of step 3, extract the gamut boundary discrete point of rgb space from all characterization modeling datas, utilize 81 frontier points in each outer junction, first every four adjacent frontier points are formed a square, the diagonal line that recycles 2 compositions in each square upper left corner and the lower right corner is divided into two triangles, and the cubical outer junction of RGB produces 128 triangles altogether like this;

Step 5, description RGB colour gamut outward appearance

The triangle of cutting apart generation for 6 outer junctions of step 4 pair RGB cube, they are carried out color filling, during to each triangle Fill Color, the information on three summits leg-of-mutton according to each, carry out color filling in rgb space, its color value represents with following formula:

T _i＝(A _i+B _i+C _i)/3，

Wherein i represents Color Channel R, G or the B on summit, the color value of Ti for needing to fill, and A, B, C are leg-of-mutton three summits, and all triangles are sealed filling, obtain RGB cube colour gamut outward appearance;

Step 6, the RGB cube gamut boundary point that utilizes step 4 to extract, search in the measurement data of step 2 storage, find out corresponding CIELAB value, the discrete form of these frontier points is described in the CIELAB space, then according to 128 triangular construction orders in 6 outer junctions in the RGB cube, each outer junction, convert the rgb value of all triangular apex to measure acquisition CIELAB value, in the CIELAB space, all triangles are sealed paster, obtain the colour gamut outside drawing.

2. three-dimensional gamut contour description method according to claim 1, is characterized in that, in described step 2, when setting the Gamma value, system is 1.8 for apple, is 2.2 for PC.

Images