CN100573651C - A kind of color domain mapping real-time and real-time treatment circuit - Google Patents
A kind of color domain mapping real-time and real-time treatment circuit Download PDFInfo
- Publication number
- CN100573651C CN100573651C CNB200710121830XA CN200710121830A CN100573651C CN 100573651 C CN100573651 C CN 100573651C CN B200710121830X A CNB200710121830X A CN B200710121830XA CN 200710121830 A CN200710121830 A CN 200710121830A CN 100573651 C CN100573651 C CN 100573651C
- Authority
- CN
- China
- Prior art keywords
- color data
- lattice point
- color
- lattice
- class
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
The present invention relates to a kind of color domain mapping real-time and real-time treatment circuit, wherein disposal route comprises: 1) the representative colors data are deposited in four groups of lookup table memories respectively; 2) input primitive color data to be corrected extract four lattice points according to these primitive color data in the primitive color space, and decoding obtains the memory address of described four lattice points; 3) find the representative correction of color data of described four lattice points respectively according to memory address; 4) according to the representative correction of color data of described four lattice points and the primitive color data to be corrected of input, carry out interpolation arithmetic, draw final correction of color data with the tetrahedral interpolation value-based algorithm.Treatment circuit comprises address decoder, four groups of lookup table memories and interpolating circuit.Compared with prior art, the present invention combines the color map look-up table with the tetrahedral interpolation value-based algorithm, saved hardware resource, has improved frequency of operation simultaneously, thereby has improved arithmetic speed effectively.
Description
Technical field
The invention belongs to the color signal process field, specifically, the present invention relates to Color Gamut Mapping disposal route and treatment circuit.
Background technology
Display device type in the market is various, and in order to realize match colors between the display device with different colour gamuts and color characteristics, people study various Color Gamut Mapping methods.On hardware handles, the method that Color Gamut Mapping can adopt the color map look-up table to combine with three-dimensional interpolation realizes that this method mainly comprises three steps: packing, extract and interpolation.Wherein, packing is meant the 3D grid discrete sampling is carried out in the primitive color space; The raw data of sampling network lattice point correspondence is called representative primitive color data, the correction of color data corresponding with these raw data (dissimilar display devices has different correction of color data) are called representative correction of color data, and the mapping table between representative primitive color data and the representative correction of color data is exactly described color map look-up table.Because color data is made up of three components (as the RGB component) usually, so the color map look-up table usually is a three dimensional lookup table.Typical evenly packaging method, i.e. uniformly-spaced discrete sampling on three component directions of color has been shown in the accompanying drawing 1.After the packing whole original color gamut space is divided into a plurality of cubes of being made up of sampled point (lattice point), is convenient description and understanding hereinafter, will be called cubic units (minimum sampling cubic units) by the minimum cube that adjacent lattice point is formed.Accompanying drawing 3 has provided the example of color map look-up table.Extract, be meant from interpolation point (i.e. Shu Ru primitive color data to be corrected residing position the primitive color space) and select some sampling network lattice points on every side that the sealing physical efficiency that these points surround is enclosed in its inside with this interpolation point.In the extraction step, the interpolation algorithm difference that is adopted, the number that extracts lattice point is also different with selection mode.Interpolation, promptly according to interpolation point and from extraction step obtain the sampling grid lattice point between geometric relationship and the interpolation algorithm that adopted, between the correction of color data of the primitive color data correspondence that the sampling grid lattice point that extraction step obtains is represented, carry out interpolation, calculate the output color data.Relevant interpolation algorithm comprises (tetrahedral interpolation) such as pyramid interpolation, Tri linear interpolation (trilinear interpolation), tetrahedron interpolation.On hardware, directly realize relatively difficulty of color map look-up table, especially be difficult to extract simultaneously the pairing correction of color data of the required some net points of interpolation calculation.Dongil Han is at document [J] .IEEE Trans.on Consumer Electronics, May2004, Vol.50, Issue 2, Page (s): 691 698 " Real-time Color GamutMapping Method for Digital TV Display Quality Enhancement ", [J] .IEEE Trans.on Consumer Electronics, Feb.2005, Vol.51, Issue 1, Page (s): 168 174 " A Cost Effective Color Gamut MappingArchitecture for Digital TV Color Reproduction Enhancement " in introduced that three dimensional lookup table with complexity converts eight one dimension look-up tables to and the hardware algorithm structure of the realization Color Gamut Mapping that combines with Tri linear interpolation algorithm (trilinear interpolation).The weak point of this method is the Tri linear interpolation algorithm owing to will carry out iterated interpolation calculating to eight points, operand is bigger, need expend great amount of hardware resources, therefore also limited allow maximum operation frequency---the experiment on the fpga chip SpartanIII of xilinx company xc3s400 shows, the circuit that eight groups of one dimension look-up tables that propose with Dongil Han and Tri linear interpolation combine and realize, its maximum operation frequency is near the 60M hertz.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, the color map look-up table combined with the tetrahedral interpolation value-based algorithm, so provide a kind of can the economize on hardware resource, improve the color domain mapping real-time and the treatment circuit of maximum operation frequency.
For achieving the above object, color domain mapping real-time provided by the invention comprises the steps:
1) deposits representative correction of color data in four groups of lookup table memories respectively; Wherein, the representative correction of color data with first kind lattice point deposit in first group of storer; The representative correction of color data of the second class lattice point are deposited in second group of storer; The representative correction of color data of the 3rd class lattice point are deposited in the 3rd group of storer; The representative correction of color data of the 4th class lattice point are deposited in the 4th group of storer; Described first, second, third and fourth class lattice point is defined as follows: appoint a minimum sampling cubic units of getting in the original color gamut space, a lattice point in this cubic units is defined as first kind lattice point, another lattice point that is positioned on the described first kind lattice point body diagonal is defined as the second class lattice point, three lattice points adjacent with described first kind lattice point are defined as the 3rd class lattice point, three lattice points adjacent with the described second class lattice point are defined as the 4th class lattice point, and all lattice points in the original color gamut space all are defined as the first, second, third or the 4th class lattice point uniquely; Such definition makes that resulting four lattice points must be stored in respectively in four groups of storeies in extraction step, and promptly any two lattice points that extract can not be arranged in same group of storer;
2) input primitive color data to be corrected extract four lattice points according to these primitive color data in the primitive color space, and decoding obtains the memory address of described four lattice points (i.e. four representative primitive color data);
3) described four groups of lookup table memories are respectively according to step 2) memory address that draws with four representative correction of color data sync export interpolating circuit to;
4) according to the representative correction of color data of described four lattice points and the primitive color data to be corrected of input, carry out interpolation arithmetic, draw final correction of color data with the tetrahedral interpolation value-based algorithm.
In the technique scheme, in the described step 1), the 3D grid discrete sampling is carried out in the primitive color space, the representative correction of color data of the lattice point that sampling is obtained deposit in respectively in four groups of lookup table memories with certain sortord then.
In the technique scheme, described step 2) in, described four lattice points that extract in original color gamut space belong to same cubic units, and the interpolation point of described primitive color data correspondence is positioned at the tetrahedron inside that described four lattice points constitute.
In the technique scheme, in the described step 4), described interpolating circuit carries out interpolation arithmetic according to primitive color data to be corrected and four representative correction of color data receiving with the tetrahedral interpolation value-based algorithm, draws final correction of color data.
In the technique scheme, described step 2) in, described address decoder is after extracting described four lattice points, the positional information (i.e. four representative primitive color data) of these four lattice points is exported to described interpolating circuit, in the described step 4), described interpolating circuit is according to the positional information calculation interpolation weights of described four lattice points.
In the technique scheme, described step 2) in, described as follows in the method that original color gamut space extracts four lattice points that belong to same minimum sampling cubic units: as at first to find the minimum sampling cubic units that comprises these primitive color data to be corrected according to primitive color data to be corrected; Be two common points with summit that belongs to first kind lattice point on this cubic units and the summit that belongs to the second class lattice point again, to be not that two end points of any rib of end points are all the other two summits or not, be six tetrahedrons of splicing mutually with this cubic units subdivision in above-mentioned two summits any; Find the solution these primitive color data to be corrected then and be contained in the tetrahedron which described subdivision goes out, these tetrahedral four summits are four lattice points being asked.
For realizing another goal of the invention of the present invention, the real-time treatment circuit of Color Gamut Mapping provided by the invention comprises: address decoder, lookup table memories and interpolating circuit;
Described lookup table memories has four groups, the representative correction of color data that each group storer is stored the class lattice point in the original color gamut space respectively; Wherein, the representative correction of color data of first group of memory stores first kind lattice point; The representative correction of color data of second group of memory stores second class lattice point; The representative correction of color data of the 3rd group of memory stores the 3rd class lattice point; The representative correction of color data of the 4th group of memory stores the 4th class lattice point; Described first, second, third and fourth class lattice point is defined as follows: appoint a minimum sampling cubic units of getting in the original color gamut space, a lattice point in this cubic units is defined as first kind lattice point, another lattice point that is positioned on the described first kind lattice point body diagonal is defined as the second class lattice point, three lattice points adjacent with described first kind lattice point are defined as the 3rd class lattice point, three lattice points adjacent with the described second class lattice point are defined as the 4th class lattice point, and all lattice points in the original color gamut space all are defined as the first, second, third or the 4th class lattice point uniquely;
Described address decoder is used to receive primitive color data to be corrected, and extracting four lattice points that belong to same cubic units in original color gamut space according to the primitive color data, the interpolation point of described primitive color data correspondence is positioned at the tetrahedron inside that described four lattice points constitute; Simultaneously described address decoder decoding draws the memory address of representative correction of color data in described four groups of lookup table memories of described four lattice points;
Described interpolating circuit is used for obtaining synchronously from four groups of lookup table memories respectively the representative correction of color data of described four lattice points, in conjunction with described primitive color data to be corrected, carry out interpolation arithmetic with the tetrahedral interpolation value-based algorithm, draw final correction of color data number.
In the technique scheme, described lookup table memories is also exported the positional information of described four lattice points for described interpolating circuit, be used to calculate interpolation weights.
In the technique scheme, described each group lookup table memories includes three storeies, stores three color components of color data respectively.
In the technique scheme, described each group lookup table memories can be made of single memory, stores three color components of color data on the different pieces of information position of this storer respectively.
In the technique scheme, the real-time treatment circuit of described Color Gamut Mapping also comprises synchronizer, and this synchronizer is described address decoder, and lookup table memories and interpolating circuit provide synchronizing signal.
Compared with prior art, the present invention can reach following technique effect:
Method of the present invention converts the three dimensional lookup table of complexity to four groups of one dimension look-up tables, and provide the address decoder method, make that extracting four representative correction of color data simultaneously according to the input color data becomes possibility, and then be that real-time processing provides the foundation such as the interpolation algorithm that needs such as tetrahedron interpolation extract four net points.The Color Gamut Mapping circuit that adopts such scheme to realize in the overwhelming majority is used, can be placed on the inner realization of logic IC or asic chip, thereby simplify single board design, reduces cost; If representative correction of color data volume is excessive, can not be placed on logic IC or asic chip inside, then can be at logic IC or ASIC exterior arrangement a slice memory chip data-carrier store as look-up table; The present invention also provides a kind of real-time treatment circuit of realizing by such scheme of Color Gamut Mapping, and this circuit structure is simple, and it is few to take hardware resource, and computing velocity is fast.
Description of drawings
Fig. 1 is the synoptic diagram of evenly packing;
Fig. 2 is the synoptic diagram of tetrahedron interpolation;
Fig. 3 is the synoptic diagram of color map table;
Fig. 4 is the process flow diagram of one embodiment of the invention;
Fig. 5 is that one embodiment of the invention deposits representative correction of color data respectively in is the synoptic diagram of four groups of lookup table memories;
Fig. 6 is the synoptic diagram of first kind lattice point and the second class lattice point in the one embodiment of the invention;
Fig. 7 is a definition first kind lattice point in the one embodiment of the invention, and the second class lattice point and the position between the two concern synoptic diagram;
Fig. 8 is the synoptic diagram of definition the 3rd class lattice point in the one embodiment of the invention;
Fig. 9 is the synoptic diagram of definition the 4th class lattice point in the one embodiment of the invention;
Figure 10 A is a synoptic diagram of respectively organizing grid point distribution in the one embodiment of the invention on the verso;
Figure 10 B is a synoptic diagram of respectively organizing grid point distribution in the one embodiment of the invention on the recto;
Figure 11 is the process flow diagram of the address decoder method of one embodiment of the invention;
Figure 12 is in the one embodiment of the invention, sets up the synoptic diagram of coordinate system in minimum sampling cubic units inside;
Figure 13 is in the one embodiment of the invention, minimum sampling cubic units is carried out the principle schematic of tetrahedron subdivision;
Figure 14 is the electrical block diagram of one embodiment of the invention.
Embodiment
Below in conjunction with drawings and Examples the present invention is done and to describe in further detail.
Fig. 4 is the inventive method implementing procedure figure, comprises the steps:
1) deposits representative correction of color data in four groups of lookup table memories respectively;
2) input primitive color data to be corrected extract four lattice points according to these primitive color data in the primitive color space, and decoding obtains the memory address of described four lattice points (i.e. four representative primitive color data);
3) find the representative correction of color data of described four lattice points respectively according to memory address;
4) according to the representative correction of color data of described four lattice points and the primitive color data to be corrected of input, carry out interpolation arithmetic, draw final correction of color data with the tetrahedral interpolation value-based algorithm.
In above-mentioned four steps, the step that the representative colors data is deposited in four groups of lookup table memories respectively is the prerequisite step, and 3 steps in back are real-time treatment steps that concrete color data is shone upon.
The lookup table memories that relates in the present embodiment can be at logic IC chip such as FPGA, CPLD etc., or realizes in that application-specific integrated circuit ASIC is inner; When representative correction of color data are too big, also can realize by peripheral memory such as EEPROM.
In the described step 1), described representative primitive color data are the sampled data set of the original equipment color space being carried out the 3 d-dem sampling, obtain by the packing step usually.Packaging method commonly used is even packing.Fig. 1 is to the primitive color space synoptic diagram of evenly packing.The primitive color data of its sampling grid lattice point correspondence (being called representative primitive color data) can be the set of following data: low order LSB (least significant bits) is color data collection zero, any value of high significance bit MSB (most significantbits) entirely; Because above-mentioned data acquisition can not comprise the border color data in the primitive color space, so representative raw data also should comprise next group color data collection, the low order LSB that promptly wherein has at least a color component to obtain maximal value, all the other color components is zero color data collection entirely.As RGB color signal with 24 binary representations, can get respectively high 3 as MSB, hang down 5 as LSB, to be sampled at 0,32,64,96,128,160,192,224,255 places respectively on then arbitrary color component, be to obtain 9 sampling GTGs on each sample direction, obtain 729 sampling network lattice points altogether.Described representative primitive color data between corresponding correction of color data (being called representative correction of color data), promptly constitute color lookup table, as shown in Figure 3.
Shown in Figure 5 is that in the present embodiment representative correction of color data to be deposited in respectively be the synoptic diagram of four groups of lookup table memories.In this example, describedly deposit representative correction of color data in four groups of lookup table memories respectively and be meant that the representative correction of color data with these 729 sampled point correspondences deposit four groups of storeies respectively in.Wherein, every group of storer can be set to three, a certain road color data in each storage RGB color data; Every group of storer also can be set to one, and method is that every group of RGB color data merged into one, as merging into road color data with 24 binary representations with three road RGB color data of 8 binary representations respectively.
Shown in Figure 6 is location first kind lattice point 11 in three-dimensional sampling grid, and the representative correction of color data of the primitive color data correspondence of this class lattice point representative will be deposited in first group of storer.Fig. 6 has also expressed the part and the second class lattice point 22, and in Fig. 7 the relation of the position between above-mentioned two class net points is further specified, and promptly the two always is positioned at wherein body diagonal two ends of cubic units.Fig. 6 has also expressed the position relation of the 3rd class lattice point 33 and the 4th class lattice point 44 unit on same cube, and promptly they are respectively adjacent to first kind net point 11 and the second class net point 22 (can with reference to figure 8 and Fig. 9).
Describedly deposit representative correction of color data in four groups of lookup table memories respectively and sort as follows:
(1) described three-dimensional sampling grid is decomposed into several layers along a certain color component direction, every layer is called one page, and with highest significant position sign this page sequence number of this page up-sampling lattice point on this color component, when this color component is obtained maximal value, add this page of value sign sequence number of 1 with its highest significant position; Each page or leaf is divided into row and column according to all the other two color components again, and indicates the row and column sequence number respectively with the highest significant position of these two color components, when the investigation color component is obtained maximal value, adds this row of value sign of 1 or/and the row sequence number with its highest significant position; Visit all sampling lattice points then line by line page by page successively, when the lattice point that is accessed to belongs to the first kind lattice point 11 or the second class lattice point 22, this lattice point is arranged as one-dimensional sequence according to the sequencing that is accessed to, all accessed and arrange and to finish up to all such lattice points, the correction of color data with such lattice point correspondence deposit corresponding storer in according to same putting in order again;
(2), be parallel to which sample direction according to the line of itself and adjacent first kind lattice point 11 and be divided into three groups to described the 3rd class lattice point 33; Method according to the above-mentioned arrangement first kind lattice point and the second class lattice point sorts to described three group lattice points respectively again, then three group lattice points after sorted are connected to an one-dimensional sequence, the correction of color data with this one-dimensional sequence lattice point correspondence deposit the 3rd thin storer 3 in according to same putting in order at last;
(3), be parallel to which sample direction according to the line of itself and the adjacent second class lattice point 22 and be divided into three groups to described the 4th class lattice point 44; Method according to the above-mentioned arrangement first kind lattice point and the second class lattice point sorts to described three group lattice points respectively again, then three group lattice points after sorted are connected to an one-dimensional sequence, the correction of color data with this one-dimensional sequence lattice point correspondence deposit the 4th group of storer in according to same putting in order at last.
Such classification is in order to make that resulting four lattice points must be stored in respectively in four groups of storeies in extraction step, and promptly any two lattice points that extract can not be arranged in same group of storer.The mode classification of present embodiment is not unique.
Understand for convenient, further describe below in the present embodiment, representative correction of color data are deposited in the specific implementation of four groups of lookup table memories respectively.
In the present embodiment, described three-dimensional sampling grid is decomposed into 9 layers along a certain color component direction (such as the B component direction), every layer is called one page, and indicate this page sequence number with the highest significant position MSB of this page up-sampling lattice point on this color component B (in this example with 3 binary representations), as with first footers be 0, second footers be 1 ... .., the 8th footers are 7, the 9th page of up-sampling lattice point obtained maximal value on this color component B, and its high significance bit identical with the 8th page, so the time add value 8 this page of sign sequence numbers of 1 with its highest significant position; Each page or leaf is divided into row and column according to all the other two color components again and (is divided into row such as making on the R component, be divided into row on the G component), and indicate the row and column sequence number respectively with the highest significant position of these two color components, equally, when the investigation color component is obtained maximal value, add this row of value sign of 1 or/and the row sequence number with its highest significant position, accompanying drawing 10A and 10B have explained the wherein ranks denotation approach of verso and recto up-sampling grid respectively, and have indicated a kind of grid classification situation on figure; Visit all sampling lattice points then line by line page by page successively, when the lattice point that is accessed to belongs to the first kind lattice point 11 or the second class lattice point 22, this lattice point is arranged as one-dimensional sequence according to the sequencing that is accessed to, all accessed and arrange and to finish up to all such lattice points, the correction of color data with such lattice point correspondence deposit corresponding storer in according to same putting in order again; To described the 3rd class lattice point 33, be parallel to which sample direction according to the line of itself and adjacent first kind lattice point 11 and be divided into three groups, as shown in Figure 8, totally 3 groups of lines among the figure between first kind lattice point 11 and adjacent the 3rd class lattice point 33 are parallel to certain sample direction (color component direction) respectively; Method according to the above-mentioned arrangement first kind lattice point and the second class lattice point sorts to described three group lattice points respectively again, then three group lattice points after sorted are connected to an one-dimensional sequence, described method of attachment, such as ranked first the group lattice point earlier, be right after and arrange second group thereafter, follow the 3rd group, the correction of color data with this one-dimensional sequence lattice point correspondence deposit the 3rd group of storer 3 in according to same putting in order at last; To described the 4th class lattice point 44, be parallel to which sample direction according to the line of itself and the adjacent second class lattice point 22 and be divided into three groups; Method according to the above-mentioned arrangement first kind lattice point and the second class lattice point sorts to described three group lattice points respectively again, then three group lattice points after sorted are connected to an one-dimensional sequence, the correction of color data with this one-dimensional sequence lattice point correspondence deposit the 4th group of storer in according to same putting in order at last.
Described step 2) in, address decoder calculates four addresses of four groups of lookup table memories of visit according to the input color data.Described address decoder method is relevant with the method that above-mentioned representative correction of color data deposit storer in.Accompanying drawing 11 has provided address decoder algorithm flow chart of the present invention.First three step described in this figure all is for four in unique definite input color data field sampling lattice points, thereby obtains four corresponding correction of color data, finishes described extraction step.
Described address decoder is visited four correction of color data correspondences of four groups of storer gained respectively from same input color data behind address decoder four sampling lattice points belong to four summits of certain on the same minimum cubic units the sampling grid, and two of two correction of color data of gained correspondence sampling lattice points belong to the position adjacent relation from the 3rd group of storer and the 4th group of storer, so this cubic units can be split into six tetrahedrons of splicing mutually.The coding/decoding method of this address decoder also further comprises:
(1) is somebody's turn to do the minimum cubic units of importing color data place in sampling grid with the high significance bit MSB addressing of input color data, and be initial point with the summit that belongs to first kind lattice point on this minimum cubic units, the summit that order belongs to the second class lattice point is positioned at first quartile, sets up the coordinate system of this minimum cubic units inside;
(2) described color data has three color components, therefore can be used as color vector, so the color vector with described input color data deducts the color vector that belongs to the summit of first kind lattice point on the cubic units that comprises this input color vector, again the gained vectorial difference is asked absolute value, promptly try to achieve the coordinate of this color vector in above-mentioned cubic units intrinsic coordinates is, establish usefulness (r, g, b) expression;
(3) sampling interval of establishing three directions is Δ R, Δ G, Δ B, be six tetrahedrons with the cubic units subdivision then to described tetrahedron subdivision method, can determine what tetrahedron is the input color data be contained in according to the comparison operation between the above-mentioned coordinate figure (r, g, b), as: if
And
Then described input color data belongs to some tetrahedrons,
Be designated as I;
If
And
Then described input color data belongs to another tetrahedron,
Be designated as II;
If
And
Then described input color data belongs to another tetrahedron,
Be designated as III;
If
And
Then described input color data belongs to another tetrahedron,
Be designated as V;
If
And
Then described input color data belongs to another tetrahedron,
Be designated as VI;
Under all the other situations, described input color data belongs to another tetrahedron, is designated as VII;
(4) the tetrahedral method in above-mentioned location can be determined four lattice points extracting, carry out inverting by method again to grouping in the described storer and ordering, can try to achieve the address of described four lattice points after grouping and ordering, thereby calculate the storage address of four correction of color data in data-carrier store that extracts, finish address decoder; Simultaneously four primitive color data of above-mentioned four lattice point correspondences are delivered to the interpolating circuit unit, carry out interpolation and calculate.
The sortord, the decoding process that are pointed out that present embodiment are not unique.As long as can successfully calculate the address of the representative correction of color data of four lattice points that extract according to the primitive color data of input, any sortord, decoding process all belong to category of the present invention.
In the present embodiment, should import the minimum cubic units at color data place in sampling grid with the high significance bit MSB addressing of input color data, be meant the high significance bit MSB that keeps this input color data and make that its low order is that the color data 0 of zero back gained must be that sample in eight summits of cubic units one of the most close initial point of the minimum that comprises this input color data (is the point at ater color place entirely, might belong to the arbitrary class in the described four class lattice points), three color components of described gained color data 0 add respectively the color data P that obtains after the corresponding sampling interval then the body of representative color data 0 place lattice point so can uniquely determine above-mentioned minimum sampling cubic units to angular vertex.To above-mentioned sampling lattice point number 729 embodiment, as import color data for (R, G, B)=(19,72,38), then the MSB data splitting of this color is (0,2,1), the MSB data splitting of its body diagonal angle lattice point is (0+1,2+1,1+1) i.e. (1,3,2), the primitive color data of their correspondences are (0,64,32) and (32,96,64), wherein (32,96,64) also can pass through (0,64,32)+(32,32,32) obtain (32 is three sampling interval on the sample direction).If the input color data is (R, G, B)=(190,253,237), then the MSB data splitting of this color is (5,7,7), and the MSB data splitting of its body diagonal angle lattice point is (6,8,8), but represents with triad because of MSB in this example, its maximal value is less than 8, so the MSB data splitting of its body diagonal angle lattice point should be (6,7,7), but the primitive color data of described body diagonal angle lattice point correspondence should be (192,255,255).
Accompanying drawing 12 has been explained described method at minimum sampling cube internal structure coordinate system.By by described method construct coordinate system, can unify the tetrahedron subdivision and locate the input color to be contained in what tetrahedral method.According to described method, at first should locate the distribution of first kind lattice point at cubic units internal structure coordinate system.Lattice point sorting technique among accompanying drawing 10A and the 10B has supposed that the lattice point that described capable sequence number, row sequence number and page or leaf sequence number are even number belongs to first kind lattice point.So, the sampled point that belongs to first kind lattice point in the whole sampling grid of described embodiment has 25 * 5=125.In the circuit design, as long as the position, end of three MSB data is zero entirely on the lattice point, then this lattice point belongs to first kind lattice point.On circuit, also be easy to realize judging which class lattice point investigates lattice point belongs to according to similar method.The last position combination of the MSB of arbitrary input color data has eight kinds of combinations, wherein has only the corresponding first kind lattice point of a kind of combination, promptly when it is combined as (0,0,0); When described MSB last position was not 0 entirely, last position was not that 0 MSB data add 1, so can find the sampled point that belongs to first kind lattice point all the time in unit cube.And deduct the color vector of the described sampled point that belongs to first kind lattice point that finds with described input color data (vector), promptly try to achieve described coordinate.As import color for (R, G, B)=(19,72,38), then the MSB data splitting of this color is (0,2,1), obviously comprise the MSB combination that belongs to first kind lattice point in the cubic units of importing data and should be (0,2,1)+(0,0,1)=(0,2,2), then the color data of its representative is (0,64,64), so the coordinate of asking is (r, g, b)=| (19,72,38)-(0,64,64) |=(19,8,26).By the tetrahedral method in described location, b＞r＞g is arranged, again so this input color dot belongs to described VII class tetrahedron.Can determine the position on these four summits of tetrahedron according to the tetrahedron subdivision method, and, find the solution lattice point address after sorted, promptly read the address of four representative correction of color data in four groups of storeies according to the above-mentioned method that all kinds of lattice points are divided into groups and sort.In addition, above-mentioned corresponding four primitive color data of positional information that extract four summits of tetrahedron, four primitive color data of this that solves in the decode procedure are sent to the interpolative operation unit, to carry out interpolation arithmetic.Owing to relate to sequential processing, should adopt Synchronization Design in the circuit design, but it is dispensable that four correction of color data sync that above-mentioned four primitive color data and address decoder read out from four groups of storeies arrive the interpolative operation unit, the former can arrive the previous one or several clock period, carries out the calculating of interpolation weights to save time.
Accompanying drawing 13 has been explained described method of minimum sampling cube being carried out the tetrahedron subdivision.Press shown in the figure, may be interpreted as: find to belong to first kind lattice point respectively on the cubic units or/and two summits of the second class lattice point, and with they all tetrahedral two public vertex for being gone out by subdivision, selecting in 12 ribs not with above-mentioned two summits is that the two ends of a rib of end points are certain tetrahedral remaining two summit that will be gone out by subdivision, and this cube is split into six tetrahedrons of splicing mutually.
In the described step 4),, obtain the output color data by carrying out interpolation at four groups of representative colors correction datas that in four groups of lookup table memories, check in according to address decoder.From the spatial spreading data, extract in the interpolation algorithm of four points, typical in the tetrahedral interpolation value-based algorithm.Accompanying drawing 2 is a tetrahedron interpolation synoptic diagram.Among the figure, four sampling lattice point A, B, C, D are mapped as four some A ' in the object space in certain luv space, B ', and C ', D ' is contained in interpolation point P among the tetrahedron ABCD and is mapped as some P ' among tetrahedron A ' B ' C ' D '.If the value (can be scalar, perhaps vector) of each point representative represents that with Vi (i=A, B, C, D, P, A ', B ', C ', D ', P ') then the tetrahedral interpolation value-based algorithm can be expressed as among the figure:
V
P′＝α×V
A′+β×V
B′+γ×V
C′+κ×V
D′
In the formula, α, beta, gamma, κ are respectively four some A ', B ', and C ', the interpolation weights of D ', its computing method are volume ratios of four little tetrahedron BCDP, ACDP, ABDP, ABCP and tetrahedron ABCD.
Accompanying drawing 14 provides realizes a kind of electrical block diagram of the present invention, comprise address decoder, lookup table memories and interpolating circuit (being the interpolative operation unit), also comprise a synchronizer simultaneously, for described address decoder, lookup table memories and interpolating circuit provide synchronizing signal.
Color domain mapping real-time provided by the invention and treatment circuit, be applicable to the shades of colour System with Real-Time, such as the Color Gamut Mapping of the panchromatic demonstration of laser, the colour gamut coupling that handset displays and LCD show etc. also are applicable to the mapping field in other two three-dimensional data spaces.The Color Gamut Mapping circuit that adopts such scheme to realize in the overwhelming majority is used, can be placed on the inner realization of logic IC or asic chip, thereby simplify single board design, reduces cost; If representative correction of color data volume is excessive, can not be placed on logic IC or asic chip inside, then can be at logic IC or ASIC exterior arrangement a slice memory chip data-carrier store as look-up table.In order to improve processing speed, can in specific design, adopt multistage flowing water, to satisfy the needs of two-forty real time data processing.Experiment on the fpga chip SpartanIII of xilinx company xc3s400 shows that adopt the circuit of method construct of the present invention, the maximum operation frequency of its permission can reach the 80H hertz.
In the present embodiment,, be appreciated that the present invention also can be used for the Color Gamut Mapping or the mapping of other three-dimensional data of other three-dimensional color data though employing is the RGB color data.
Be pointed out that also sampling of the present invention must be a uniform sampling also, also can be inhomogeneous sampling, when adopting inhomogeneous sampling, minimum sampling volume unit is not a cubic units, but common rectangular parallelepiped unit.
It should be noted last that above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is had been described in detail with reference to embodiment, those of ordinary skill in the art is to be understood that, technical scheme of the present invention is made amendment or is equal to replacement, do not break away from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (11)
1. a color domain mapping real-time comprises the steps:
1) deposits representative correction of color data in four groups of lookup table memories respectively; Wherein, the representative correction of color data with first kind lattice point deposit in first group of storer; The representative correction of color data of the second class lattice point are deposited in second group of storer; The representative correction of color data of the 3rd class lattice point are deposited in the 3rd group of storer; The representative correction of color data of the 4th class lattice point are deposited in the 4th group of storer; Described first, second, third and fourth class lattice point is defined as follows: appoint a minimum sampling cubic units of getting in the original color gamut space, a lattice point in this cubic units is defined as first kind lattice point, another lattice point that is positioned on the described first kind lattice point body diagonal is defined as the second class lattice point, three lattice points adjacent with described first kind lattice point are defined as the 3rd class lattice point, three lattice points adjacent with the described second class lattice point are defined as the 4th class lattice point, and all lattice points in the original color gamut space all are defined as the first, second, third or the 4th class lattice point uniquely;
2) input primitive color data to be corrected extract four lattice points according to these primitive color data in the primitive color space, and decoding obtains the memory address of described four lattice points;
3) described four groups of lookup table memories are respectively according to step 2) memory address that draws with four representative correction of color data sync export interpolating circuit to;
4) according to the representative correction of color data of described four lattice points and the primitive color data to be corrected of input, carry out interpolation arithmetic, draw final correction of color data with the tetrahedral interpolation value-based algorithm.
2. color domain mapping real-time according to claim 1, it is characterized in that, in the described step 1), the 3D grid discrete sampling is carried out in the primitive color space, and the representative correction of color data of the lattice point that sampling is obtained deposit in respectively in four groups of lookup table memories with certain sortord then.
3. color domain mapping real-time according to claim 2, it is characterized in that, described step 2) in, described four lattice points that extract in original color gamut space belong to same minimum sampling cubic units, and the interpolation point of described primitive color data correspondence is positioned at the tetrahedron inside that described four lattice points constitute.
4. color domain mapping real-time according to claim 3, it is characterized in that, in the described step 4), described interpolating circuit is according to primitive color data to be corrected and four representative correction of color data receiving, carry out interpolation arithmetic with the tetrahedral interpolation value-based algorithm, draw final correction of color data.
5. color domain mapping real-time according to claim 4, it is characterized in that, described step 2) in, described address decoder is after extracting described four lattice points, the positional information of these four lattice points is exported to described interpolating circuit, in the described step 4), described interpolating circuit is according to the positional information calculation interpolation weights of described four lattice points.
6. color domain mapping real-time according to claim 3, it is characterized in that, described as follows in the method that original color gamut space extracts four lattice points that belong to same minimum sampling cubic units: as at first to find the minimum sampling cubic units that comprises these primitive color data to be corrected according to primitive color data to be corrected; Be two common points with summit that belongs to first kind lattice point on this cubic units and the summit that belongs to the second class lattice point again, to be not that two end points of any rib of end points are all the other two summits or not, be six tetrahedrons of splicing mutually with this cubic units subdivision in above-mentioned two summits any; Find the solution these primitive color data to be corrected then and be contained in the tetrahedron which described subdivision goes out, these tetrahedral four summits are four lattice points being asked.
7. the real-time treatment circuit of Color Gamut Mapping comprises: address decoder, lookup table memories and interpolating circuit; It is characterized in that,
Described lookup table memories has four groups, the representative correction of color data that each group storer is stored the class lattice point in the original color gamut space respectively; Wherein, the representative correction of color data of first group of memory stores first kind lattice point; The representative correction of color data of second group of memory stores second class lattice point; The representative correction of color data of the 3rd group of memory stores the 3rd class lattice point; The representative correction of color data of the 4th group of memory stores the 4th class lattice point; Described first, second, third and fourth class lattice point is defined as follows: appoint a minimum sampling cubic units of getting in the original color gamut space, a lattice point in this cubic units is defined as first kind lattice point, another lattice point that is positioned on the described first kind lattice point body diagonal is defined as the second class lattice point, three lattice points adjacent with described first kind lattice point are defined as the 3rd class lattice point, three lattice points adjacent with the described second class lattice point are defined as the 4th class lattice point, and all lattice points in the original color gamut space all are defined as the first, second, third or the 4th class lattice point uniquely;
Described address decoder is used to receive primitive color data to be corrected, and extracting four lattice points that belong to same minimum sampling cubic units in original color gamut space according to the primitive color data, the interpolation point of described primitive color data correspondence is positioned at the tetrahedron inside that described four lattice points constitute; Simultaneously described address decoder decoding draws the memory address of representative correction of color data in described four groups of lookup table memories of described four lattice points;
Described interpolating circuit is used for obtaining synchronously from four groups of lookup table memories respectively the representative correction of color data of described four lattice points, in conjunction with described primitive color data to be corrected, carry out interpolation arithmetic with the tetrahedral interpolation value-based algorithm, draw final correction of color data number.
8. the real-time treatment circuit of Color Gamut Mapping according to claim 7 is characterized in that, described lookup table memories is also exported the positional information of described four lattice points for described interpolating circuit, be used to calculate interpolation weights.
9. the real-time treatment circuit of Color Gamut Mapping according to claim 7 is characterized in that, described each group lookup table memories includes three storeies, stores three color components of color data respectively.
10. the real-time treatment circuit of Color Gamut Mapping according to claim 7 is characterized in that, described each group lookup table memories can be made of single memory, stores three color components of color data on the different pieces of information position of this storer respectively.
11. the real-time treatment circuit of Color Gamut Mapping according to claim 7 is characterized in that, this real-time treatment circuit also comprises synchronizer, and described synchronizer is described address decoder, and lookup table memories and interpolating circuit provide synchronizing signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200710121830XA CN100573651C (en) | 2007-09-14 | 2007-09-14 | A kind of color domain mapping real-time and real-time treatment circuit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200710121830XA CN100573651C (en) | 2007-09-14 | 2007-09-14 | A kind of color domain mapping real-time and real-time treatment circuit |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101123080A CN101123080A (en) | 2008-02-13 |
| CN100573651C true CN100573651C (en) | 2009-12-23 |
Family
ID=39085388
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200710121830XA Expired - Fee Related CN100573651C (en) | 2007-09-14 | 2007-09-14 | A kind of color domain mapping real-time and real-time treatment circuit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100573651C (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103209331A (en) * | 2013-03-07 | 2013-07-17 | 上海电机学院 | System and method for strengthening image color saturation |
| JP6237291B2 (en) * | 2014-02-06 | 2017-11-29 | 富士通株式会社 | Information processing apparatus, color interpolation method, program, and table generation method |
| JP6310308B2 (en) | 2014-04-15 | 2018-04-11 | キヤノン株式会社 | Color processing apparatus and method |
| CN104795049B (en) * | 2015-04-20 | 2018-05-11 | 合肥工业大学 | A kind of implementation method of display color gamut mapping |
| CN106408540B (en) * | 2016-09-27 | 2020-11-06 | 新奥特（北京）视频技术有限公司 | Extrapolation processing method and device based on 3D LUT table |
| WO2018072270A1 (en) * | 2016-10-17 | 2018-04-26 | 华为技术有限公司 | Method and device for enhancing image display |
| CN108389234B (en) * | 2018-02-05 | 2021-10-08 | 深圳市恒晨电器有限公司 | Flat LCD (liquid Crystal display) heterochromatic detection method, storage device and mobile terminal |
| CN111292235A (en) * | 2019-04-25 | 2020-06-16 | 展讯通信（上海）有限公司 | Tetrahedral interpolation calculation method and device, color gamut conversion method and device, and medium |
| CN112449168B (en) * | 2019-09-03 | 2021-11-23 | 深圳Tcl新技术有限公司 | Color gamut mapping method and system |
| CN113470591B (en) | 2020-03-31 | 2023-11-14 | 京东方科技集团股份有限公司 | Monitor color matching method and device, electronic equipment and storage medium |
| CN112017105B (en) * | 2020-08-13 | 2024-03-26 | 深圳市洲明科技股份有限公司 | Color correction FPGA implementation device and method, color correction equipment and storage medium |
| CN115486060A (en) * | 2021-03-31 | 2022-12-16 | 京东方科技集团股份有限公司 | Color correction method, field programmable gate array, chip and display device |
| WO2024000465A1 (en) * | 2022-06-30 | 2024-01-04 | 京东方科技集团股份有限公司 | Method and apparatus for generating color mapping table, method and apparatus for correcting color mapping table, and medium and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6335800B1 (en) * | 1998-12-11 | 2002-01-01 | Xerox Corporation | Method of multidimensional interpolation for color transformations |
| CN1383542A (en) * | 2000-06-15 | 2002-12-04 | 精工爱普生株式会社 | Image display and color signal adjuster for use therefor |
| CN1694156A (en) * | 2004-04-30 | 2005-11-09 | 三星电子株式会社 | Apparatus and method of transforming three color signals into multi-color signal |
| CN1753452A (en) * | 2004-09-21 | 2006-03-29 | 三星电子株式会社 | Detect gamut boundary and use its equipment that shines upon colour gamut and method |
| JP2006119883A (en) * | 2004-10-21 | 2006-05-11 | Canon Inc | Tetrahedral interpolation apparatus and method |
| CN101031022A (en) * | 2006-01-05 | 2007-09-05 | 夏普株式会社 | Image processing apparatus, image forming apparatus, control method of image processing apparatus |
-
2007
- 2007-09-14 CN CNB200710121830XA patent/CN100573651C/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6335800B1 (en) * | 1998-12-11 | 2002-01-01 | Xerox Corporation | Method of multidimensional interpolation for color transformations |
| CN1383542A (en) * | 2000-06-15 | 2002-12-04 | 精工爱普生株式会社 | Image display and color signal adjuster for use therefor |
| CN1694156A (en) * | 2004-04-30 | 2005-11-09 | 三星电子株式会社 | Apparatus and method of transforming three color signals into multi-color signal |
| CN1753452A (en) * | 2004-09-21 | 2006-03-29 | 三星电子株式会社 | Detect gamut boundary and use its equipment that shines upon colour gamut and method |
| JP2006119883A (en) * | 2004-10-21 | 2006-05-11 | Canon Inc | Tetrahedral interpolation apparatus and method |
| CN101031022A (en) * | 2006-01-05 | 2007-09-05 | 夏普株式会社 | Image processing apparatus, image forming apparatus, control method of image processing apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101123080A (en) | 2008-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100573651C (en) | A kind of color domain mapping real-time and real-time treatment circuit | |
| CN1659595B (en) | Method and apparatus for encoding texture information | |
| CN100498838C (en) | Multi-mode alpha image processing method and system | |
| CN1041568C (en) | Mixed-resolution, N-dimensional object space method and apparatus | |
| US8018458B2 (en) | Close-packed uniformly adjacent, multiresolutional overlapping spatial data ordering | |
| CN105264443B (en) | The device and method of data for computed hologram | |
| CN101258526A (en) | Texture compression based on two hues with modified brightness | |
| Alexandrov | Image representation and processing: a recursive approach | |
| CN106816143B (en) | Display panel gamut conversion method, device, system, display panel and display device | |
| CN104012097A (en) | Method and apparatus for compressing and decompressing data | |
| CN105493497A (en) | Method and processor for efficient video processing in a streaming environment | |
| CN105096286A (en) | Method and device for fusing remote sensing image | |
| CN1133995A (en) | Memory organization and method for multiple variable digital data transformation | |
| Monro et al. | Rendering algorithms for deterministic fractals | |
| CN111009027B (en) | Method for realizing color correction by using 3D LUT and display equipment | |
| CN101287060B (en) | Color conversion circuit and method of color conversion | |
| CN102799657B (en) | System and method for realizing real-time data point mapping processing based on three-dimensional checking | |
| CN103136232A (en) | Method and device for displaying dynamic map | |
| GB2496499A (en) | Interactive comparative analysis of multiple object data visualizations | |
| CN110110132A (en) | A kind of method, apparatus that establishing space lattice system and remote sensing image processing system | |
| CN115866334B (en) | Data processing method for clipping and associating content in video flow | |
| CN106021228A (en) | Method and system for performing text analysis by utilizing knowledge topographic map | |
| CN108038171A (en) | Method for writing data, device and data server | |
| CN103533327B (en) | DIBR (depth image based rendering) system realized on basis of hardware | |
| CN101976448A (en) | Drawing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20091223 Termination date: 20140914 |
|
| EXPY | Termination of patent right or utility model |