CN101969571B - Pixel processing method - Google Patents
Pixel processing method Download PDFInfo
- Publication number
- CN101969571B CN101969571B CN 200910109109 CN200910109109A CN101969571B CN 101969571 B CN101969571 B CN 101969571B CN 200910109109 CN200910109109 CN 200910109109 CN 200910109109 A CN200910109109 A CN 200910109109A CN 101969571 B CN101969571 B CN 101969571B
- Authority
- CN
- China
- Prior art keywords
- value
- pixel
- color
- color composition
- composition value
- 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.)
- Active
Links
Images
Abstract
The invention provides a pixel processing method. The method comprises the following steps: determining a first dispersion according to the difference between a preset color component value of a target pixel and a preset color component value of a peripheral pixel on a first direction; determining a second dispersion according to the difference between the preset color component value of the target pixel and a preset color component value of a peripheral pixel on a second direction; generating a first judgment value according to the preset color component values of a plurality of pixels on the first direction and the first dispersion; generating a second judgment value according to the preset color component values of a plurality of pixels on the second direction and the second dispersion; and selecting a plurality of specific pixels according to the first judgment value and the second judgment value to calculate a target color component value.
Description
Technical field
The present invention relates to a kind of pixel processing method, relate in particular to the pixel processing method of the hue component that lacks in a kind of reconstruction one pixel.
Background technology
In a traditional images treatment system, when an image inductor is responded to an external image, this image inductor can be sent to the hue component value of each pixel one image-signal processor (Image SignalProcessor, ISP) line buffer (Line Buffer), and be temporary in this line buffer, wherein each pixel only includes an a kind of color composition value of color, as shown in Figure 1.Shown in Figure 1 is that this image inductor of this traditional images treatment system is responded to the color filter array 10 that this external image produces.Each row of color filter array 10 is to be formed alternately with the different color composition value of each pixel (pixel).For example in Fig. 1, first row of color filter array 10 is to store each pixel color composition value in red green (being RG) alternate mode, and then second row is to store each pixel color composition value in turquoise (being GB) alternate mode, by that analogy.Yet, because a display is when showing this external image, its each pixel all must have the value with three kinds of hue components of RGB (being RGB), therefore the hue component value of each pixel in the color filter array 10 must just send a display driver to through handling after the composition value with three kinds of colors of RGB of calculating each pixel.Then this display driver drives this display to show this external image according to the composition value of three kinds of colors of RGB of each pixel.
Please refer again to Fig. 1, (be to have red color-values R among Fig. 1 if will calculate pixel 33
33Pixel) green tint value G
33, general known way is to obtain a level difference dh and a vertical differentiation dv of pixel 33 earlier, and then comparison level difference dh and vertical differentiation dv are to determine the COLOR COMPOSITION THROUGH DISTRIBUTION of pixel 33.Then, decide the method for a kind of interpolation (Interpolation) to come interpolation to go out green tint value G according to this COLOR COMPOSITION THROUGH DISTRIBUTION again
33In addition, known way is to extrapolate level difference dh and vertical differentiation dv, i.e. dh=abs (R with the color residual quantity of same-type color
33-R
31)+abs (R
33-R
35)+abs (G
32-G
34), dv=abs (R
33-R
13)+abs (R
33-R
53)+abs (G
23-G
43), abs (R wherein
33-R
31) refer to R
33With R
31The absolute value of difference.Because known way is to extrapolate level difference dh and vertical differentiation dv with the color residual quantity of same-type color, therefore known way can't be told some specific COLOR COMPOSITION THROUGH DISTRIBUTION of color filter array 10.For instance, work as R
13=G
23=R
33=G
43=R
53=G
34, and R
33=R
31=R
35, G
32=G
34, R
31≠ G
32The time (for example vertical stripes pattern that is spacing with a pixel), level difference dh and vertical differentiation dv can be 0, this moment, known way will be according to the G around the pixel 33
23, G
34, G
43, G
32On average go out the green tint value G of pixel 33
33, thus, the green tint value G that calculates
33Will depart from this extraneous actual value.Otherwise more suitable way should be the G with vertical direction
23, G
43Come interpolation to go out the green tint value G of pixel 33
33Therefore, how to judge effectively a pixel in the color filter array around COLOR COMPOSITION THROUGH DISTRIBUTION become the industry urgent problem.
Summary of the invention
Therefore, a purpose of the present invention is to provide the pixel processing method of the hue component that lacks in a kind of reconstruction one pixel.
According to one embodiment of the invention, it provides a kind of a plurality of pixels of handling a corresponding color filter array to have the pixel processing method of the hue component that is lacked in each pixel of single color composition (color component) with reconstruction, include: decide one first residual quantity according to a pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the first direction of an object pixel, wherein this pre-color composition value of this object pixel is to distinguish corresponding different hue component with this pre-color composition value of this surrounding pixel on being positioned at this first direction; This pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the second direction according to this object pixel decide one second residual quantity, wherein this pre-color composition value of this object pixel be positioned at this second direction on this pre-color composition value corresponding different hue component respectively of this surrounding pixel, and the extension of the extension of this first direction and this second direction intersects at this object pixel; Produce one first judgment value according to the pre-color composition value and this first residual quantity that are positioned at a plurality of pixels on this first direction; Produce one second judgment value according to the pre-color composition value and this second residual quantity that are positioned at a plurality of pixels on this second direction; And according to the comparative result of this first, second judgment value, select the pre-color composition value of a plurality of specific pixel to calculate an object color component composition value of the object color component composition that this object pixel lacks, and store this object color component composition value.
According to another embodiment of the present invention, it provides a kind of image processing system, includes an imageing sensor and an image-signal processor.This imageing sensor can produce the color filter array with a plurality of pixels, and wherein each pixel has single colour component (color component).This image-signal processor is in order to receive these a plurality of pixels, and the pixel of rebuilding the hue component that is lacked in this each pixel with following method, as follows: decide one first residual quantity according to a pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the first direction of an object pixel, wherein this pre-color composition value of this object pixel is distinguished corresponding different hue component with this pre-color composition value of this surrounding pixel on being positioned at this first direction; This pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the second direction according to this object pixel decide one second residual quantity, wherein this pre-color composition value of this object pixel be positioned at this second direction on this pre-color composition value corresponding different hue component respectively of this surrounding pixel, and the extension of the extension of this first direction and this second direction intersects at this object pixel; Produce one first judgment value according to the pre-color composition value and this first residual quantity that are positioned at a plurality of pixels on this first direction; Produce one second judgment value according to the pre-color composition value and this second residual quantity that are positioned at a plurality of pixels on this second direction; And according to the comparative result of this first, second judgment value, select the pre-color composition value of a plurality of specific pixel to calculate an object color component composition value of the object color component composition that this object pixel lacks, and store this object color component composition value.
Thus, just can obtain preferable, real object color component composition value.
Description of drawings
Fig. 1 is that an image inductor of a traditional images treatment system is responded to the color filter array schematic diagram that an external image produces;
Fig. 2 A is 1 embodiment of the functional block diagram of image processing system of the present invention;
Fig. 2 B is a kind of a plurality of pixels of handling a corresponding color filter array of the present invention have the pixel processing method of the hue component that is lacked in each pixel of single color composition with reconstruction an embodiment flow chart;
Fig. 3 is that size of the present invention is the schematic diagram of a color filter battle array of the block of 5 * 5 pixels;
Fig. 4 is the schematic diagram of a color filter battle array of one first specific image;
Fig. 5 is the schematic diagram of a color filter battle array of one second specific image.
[primary clustering symbol description]
| 10、300、400、500 | The |
| 2000 | |
| 2100 | |
| 2200 | Image- |
| 2300 | |
| 2400 | |
| 402、404、406、408、410 | Row |
| 402a-402e、404a-404e、 406a-406e、408a-408e、 410a-410e、502a-502e、 504a-504e、506a-506e、 508a-508e、510a-510e | Pixel |
| 502、504、506、508、510 | OK |
Embodiment
In the middle of specification and claims, used some vocabulary to censure specific assembly.The person with usual knowledge in their respective areas should understand, and hardware manufacturer may be called same assembly with different nouns.This specification and claims book is not used as distinguishing the mode of assembly with the difference of title, but the criterion that is used as distinguishing with the difference of assembly on function.Be an open term mentioned " comprising " in the middle of specification and the claim item in the whole text, so should be construed to " comprise but be not limited to ".In addition, " couple " word and comprise any indirect means that are electrically connected that directly reach at this, therefore, be coupled to one second device if describe one first device in the literary composition, then represent this first device and can directly be electrically connected in this second device, perhaps be electrically connected to this second device indirectly by other device or connection means.
Please refer to Fig. 2 A, Fig. 2 A is 1 embodiment of the functional block diagram of image processing system of the present invention.Image processing system 2000 comprises imageing sensor 2100, image-signal processor 2200, driver 2300 and display device 2400.Imageing sensor 2100 can be photosensitive coupling component (Charge Coupled Device, CCD) or complementary matal-oxide semiconductor (Complementary Metal-Oxide Semiconductor, CMOS) imageing sensor, it is to receive light source, and the light source that foundation receives forms the color filter array with a plurality of pixels, wherein each pixel has single colour component, and imageing sensor 2100 formed color filter arrays are similar to color filter array shown in Figure 1.Imageing sensor 2100 is sent to image-signal processor 2200 with formed color filter array one with being listed as.Image-signal processor 2200 utilizes a plurality of line buffers (line buffer is not shown among Fig. 2 A) to store each row pixel.Wait to obtain after the certain number of pixels, 2200 pairs of pixels that obtain of image-signal processor are handled, and the pixel after will handling again is sent to driver 2300.Pixel after driver 2300 combined treatment makes it be presented at display 2400 again.Wherein, the processing carried out of 2200 pairs of pixels of image-signal processor will with follow-up Fig. 2 B, 3,4,5 and relevant narration explain.
Please refer to Fig. 2 B, is a kind of a plurality of pixels of handling a corresponding color filter array of the present invention have the pixel processing method 200 of the hue component that is lacked in each pixel of single color composition (color component) with reconstruction an embodiment flow chart shown in Fig. 2 B.Please note, this color filter array of the present invention is to be that the block (block) of 5 * 5 pixels is an operational block with a size, wherein a center pixel of this operational block is pixel processing method 200 object pixel V3 to be processed of the present invention, the pre-color composition of object pixel V3 can be the pixel of the hue component of the hue component of a blueness (B) or redness (R), as shown in Figure 3, shown in Figure 3 is that size of the present invention is the schematic diagram of color filter battle array 300 of the block of 5 * 5 pixels.Please note, the present invention is not limited to the block of 5 * 5 pixels, and those skilled in the art are after reading the disclosed invention of this paper, via suitably modifying embodiments of the present invention, also can be applied to the block of other big or small pixel, for example the block of 7 * 7 pixels.Because pixel processing method 200 of the present invention does not need to use pixels all in the color filter battle array 300 in the process of an object color component composition value of an object color component composition that calculates object pixel V3, therefore the color filter battle array 300 shown in Fig. 3 only indicates required a plurality of pixel code names, be pixel (V2, G2, V3, G3, V4) and the tertial pixel (V1, G1, V3, G4, V5) of the third line, other pixel is all only represented with grid.On the other hand, if can reach identical result substantially, do not need necessarily to carry out according to the step order in the flow process shown in Fig. 2 B, and the step shown in Fig. 2 B not necessarily will carry out continuously, namely other step also can be inserted wherein.Pixel processing method 200 includes:
Step 202: according to the pre-color composition value v of object pixel V3
3With the one first pre-color composition value g that is positioned at one first surrounding pixel G2 on the horizontal direction
2Between difference and the one second pre-color composition value g of one second surrounding pixel G3
3Between difference decide one first residual quantity d1, the pre-color composition value v of object pixel V3 wherein
3Pre-color composition value g with first, second surrounding pixel G2, G3
2, g
3The hue component that difference is corresponding different;
Step 204: according to the pre-color composition value v of object pixel V3
3With the one first pre-color composition value g that is positioned at one first surrounding pixel G1 on the vertical direction
1Between difference and the one second pre-color composition value g of one second surrounding pixel G4
4Between difference decide one second residual quantity d2, the pre-color composition value v of object pixel V3 wherein
3With first, second pre-color composition value g that is positioned at first, second surrounding pixel G1, G4 on this vertical direction
1, g
4It is corresponding different respectively hue components;
Step 206: a plurality of pixel V2, G2, V3, G3, V4 divide other pre-color composition value v on this horizontal direction according to being positioned at
2, g
2, v
3, g
3, v
4Produce one first judgment value dh with the first residual quantity d1;
Step 208: a plurality of pixel V1, G1, V3, G4, V5 divide other pre-color composition value v on this vertical direction according to being positioned at
1, g
1, v
3, g
4, v
5Produce one second judgment value dv with the second residual quantity d2;
Step 210: according to the comparative result of first, second judgment value dh, dv, select the pre-color composition value of a plurality of specific pixel in the color filter battle array 300 to calculate this object color component composition value of this object color component composition that object pixel V3 lacks, and store this object color component composition value.
Please note, pixel processing method 200 of the present invention is not limited to handle the pixel (V2, G2, V3, G3, V4) of the third line of this horizontal direction and the tertial pixel (V1, G1, V3, G4, V5) of this vertical direction, after suitably modifying, pixel processing method 200 of the present invention also can be handled a plurality of pixels on any direction with mutually orthogonal characteristic in the color filter battle array 300, and it also is category of the present invention place.
On the other hand, the first surrounding pixel G2, the second surrounding pixel G3, the first surrounding pixel G1 and the second surrounding pixel G4 all are close to object pixel V3.In addition, according to embodiments of the invention, this pre-color composition of object pixel V3 is same as the hue component of pixel V1, V2, V4, V5.In other words, when this pre-color composition of object pixel V3 is the hue component of blueness, the hue component of pixel V1, V2, V4, V5 also is blue, or when this pre-color composition of object pixel V3 was the hue component of redness, the hue component of pixel V1, V2, V4, V5 also was red.On the other hand, the hue component of the first surrounding pixel G2, the second surrounding pixel G3, the first surrounding pixel G1 and the second surrounding pixel G4 is green.In other words, the purpose of pixel processing method 200 of the present invention is to calculate the green tint composition value of hue component of the green of object pixel V3, and stores this green tint composition value of object pixel V3.
In step 202, the first pre-color composition value g of the first surrounding pixel G2 on this horizontal direction
2The second pre-color composition value g with the second surrounding pixel G3
3Can be read out, to calculate the first pre-color composition value g respectively
2Pre-color composition value v with object pixel V3
3One first difference value (g
2-v
3), and the second pre-color composition value g
3Pre-color composition value v with object pixel V3
3One second difference value (g
3-v
3), wherein first, second surrounding pixel G2, G3 all are close to object pixel V3.Then, with the first difference value (g
2-v
3) and the second difference value (g
3-v
3) add up to produce one and add total value, and add total value and according to this and adjust coefficient and calculate an absolute value with as the first residual quantity d1.
Note that the present invention does not limit utilizes this to add the order that total value and this adjustment coefficient produce the first residual quantity d1.In other words, this is added total value produce one divided by this adjustment coefficient and add total value after adjusting, and the total value that adds after this adjustment is carried out a signed magnitude arithmetic(al) and produced the first residual quantity d1; Or this is added total value carry out a signed magnitude arithmetic(al) and produce an operation result, and this operation result is produced the first residual quantity d1 divided by this adjustment coefficient be category of the present invention place.On the other hand, this adjustment coefficient settings of the present invention is 1/2
N, and N is a positive integer.Owing to set N=2 in the present embodiment, therefore the first residual quantity d1 can represent by following formula (1):
d1＝abs(((g
2-v
3)+(g
3-v
3))/(2
2))，
＝abs((g
2+g
3-2*v
3)/(2
2))。(1)
In like manner, in step 204, the first pre-color composition value g of this first surrounding pixel G1 vertically upward
1The second pre-color composition value g with the second surrounding pixel G4
4Can be read out, to calculate the first pre-color composition value g respectively
1Pre-color composition value v with object pixel V3
3One first difference value (g
1-v
3), and the second pre-color composition value g
4Pre-color composition value v with object pixel V3
3One second difference value (g
4-v
3), wherein first, second surrounding pixel G1, G4 all are close to object pixel V3.Then, with the first difference value (g
1-v
3) and the second difference value (g
4-v
3) add up to produce one and add total value, and add total value and according to this and adjust coefficient and calculate an absolute value with as the second residual quantity d2.
In like manner, the present invention does not limit and utilizes this to add the order that total value and this adjustment coefficient produce the second residual quantity d2.In other words, this is added total value produce one divided by this adjustment coefficient and add total value after adjusting, and the total value that adds after this adjustment is carried out a signed magnitude arithmetic(al) and produced the second residual quantity d2; Or this is added total value carry out a signed magnitude arithmetic(al) and produce an operation result, and this operation result is produced the second residual quantity d2 divided by this adjustment coefficient be category of the present invention place.On the other hand, this adjustment coefficient settings of the present invention is 1/2
N, and N is a positive integer.Owing to set N=2 in the present embodiment, therefore in the present embodiment, the second residual quantity d2 can represent by following formula (2):
d2＝abs(((g
1-v
3)+(g
4-v
3))/(2
2))，
＝abs((g
1+g
4-2*v
3)/(2
2))。(2)
Then, step 206 of the present invention is that to calculate with object pixel V3 be a level difference of mid point.Further, a plurality of pixel V2, G2, V3, G3, V4 divide other pre-color composition value v to pixel processing method 200 of the present invention on this horizontal direction according to being positioned at
2, g
2, v
3, g
3, v
4Produce the first judgment value dh with the first residual quantity d1.In the present embodiment, the first judgment value dh can represent by following formula (3):
dh＝abs(v
3-v
2)+abs(v
3-v
4)+abs(g
2-g
3)+abs((g
2+g
3-2*v
3)/(2
2))，(3)
Abs (v wherein
3-v
2) be the residual quantity between the pre-color composition value of the pre-color composition value of object pixel V3 and pixel V2, abs (v
3-v
4) be the residual quantity between the pre-color composition value of the pre-color composition value of object pixel V3 and pixel V4, and abs (g
2-g
3) be the residual quantity between the pre-color composition value of the pre-color composition value of pixel G2 and pixel G3.
On the other hand, to be used for calculating with object pixel V3 be a vertical differentiation of mid point to step 208 of the present invention.Further, a plurality of pixel V1, G1, V3, G4, V5 divide other pre-color composition value v to pixel processing method 200 of the present invention on this vertical direction according to being positioned at
1, g
1, v
3, g
4, v
5Produce the second judgment value dv with the second residual quantity d2.In the present embodiment, the second judgment value dv can represent by following formula (4):
dh＝abs(v
3-v
1)+abs(v
3-v
5)+abs(g
1-g
4)+abs((g
1+g
4-2*v
3)/(2
2))，(4)
Abs (v wherein
3-v
1) be the residual quantity between the pre-color composition value of the pre-color composition value of object pixel V3 and pixel V1, abs (v
3-v
5) be the residual quantity between the pre-color composition value of the pre-color composition value of object pixel V3 and pixel V5, and abs (g
1-g
4) be the residual quantity between the pre-color composition value of the pre-color composition value of pixel G1 and pixel G4.
In step 210, after the first judgment value dh and the second judgment value dv were generated, relatively the first judgment value dh and the second judgment value dv were to produce a comparative result.Then, select the pre-color composition value of a plurality of specific pixel in the color filter battle array 300 to calculate an object color component composition value g of this object color component composition that object pixel V3 lacks according to this comparative result
33, and storage object color component composition value g
33
Further, if this level difference is when smaller, namely during dh＜dv, object color component composition value g then
33Can utilize following formula (5) to obtain:
g
33＝((g
2+g
3)/2)+((2*v
3-v
2-v
4)/4)； (5)
If this vertical differentiation is when smaller, namely during dh＞dv, object color component composition value g then
33Can utilize following formula (6) to obtain:
g
33＝((g
1+g
4)/2)+((2*v
3-v
1-v
5)/4)；(6)
If this level difference is when being same as this vertical differentiation, namely during dh=dv, object color component composition value g then
33Can utilize following formula (7) to obtain:
g
33＝((g
1+g
2+g
3+g
4)/4)+((4*v
3-v
1-v
2-v
4-v
5)/16)。(7)
Thus, the object color component composition value g of this object color component composition of lacking of object pixel V3
33Just can utilize the formula in formula (5), (6), (7) to obtain.
Please refer to Fig. 4 and Fig. 5, shown in Figure 4 is the schematic diagram of a color filter battle array 400 of one first specific image.Shown in Figure 5 is the schematic diagram of a color filter battle array 500 of one second specific image.Color filter battle array 400 has one first row 402, one secondary series 404, one the 3rd row 406, one the 4th row 408 and one the 5th row 410, wherein first row 402 include a plurality of pixel 402a, 402b, 402c, 402d, 402e, secondary series 404 includes a plurality of pixel 404a, 404b, 404c, 404d, 404e, the 3rd row 406 include a plurality of pixel 406a, 406b, 406c, 406d, 406e, the 4th row 408 include a plurality of pixel 408a, 408b, 408c, 408d, 408e, the 5th row 410 include a plurality of pixel 410a, 410b, 410c, 410d, 410e.In addition, a plurality of pixels in first row 402, the 3rd row 406 and the 5th row 408 all have same pre-color composition value, and a plurality of pixels in secondary series 404 and the 3rd row 406 all have same pre-color composition value, and the pre-color composition value of first row 402 is inequality in the pre-color composition value of secondary series 404, the vertical stripes pattern that it is spacing that color filter battle array 400 presents with a pixel.In addition, please also refer to color filter battle array 400 shown in Figure 4 and color filter battle array 300 shown in Figure 3, and divide other pre-color composition value v with color filter battle array more than 300 pixel V1, G1, V3, G4, V5
1, g
1, v
3, g
4, v
5A plurality of pixel 406a, 406b, 406c, 406d, 406e as opposite position in the color filter battle array 400 divides other pre-color composition value respectively, and divides other pre-color composition value v with color filter battle array more than 300 pixel V2, G2, V3, G3, V4
2, g
2, v
3, g
3, v
4A plurality of pixel 402c, 404c, 406c, 408c, 408c as opposite position in the color filter battle array 400 divides other pre-color composition value respectively.Therefore, when utilizing formula of the present invention (3) and formula (4) to come the first judgment value dh of calculating pixel 406c (being object pixel) respectively and the second judgment value dv, the first judgment value dh can not be 0, and the second judgment value dv can be 0.Thus, this vertical differentiation of pixel 406c (being object pixel) that just can judge color filter battle array 400 is smaller (be dh＞dv), so this object color component composition value of pixel 406c just can utilize formula (6) to calculate.Note that this object color component composition value is a green tint composition value of the hue component of green.
Please note, than pixel processing method 200 of the present invention, the first judgment value dh and the second judgment value dv that the traditional images processing method is calculated are 0, and make the traditional images processing method utilize formula (7) to come the object color component composition value of computing pixel 406c, so will obtain relatively poor, more false object color component composition value.
In like manner, color filter battle array 500 has one first row 502, one second row 504, one the third line 506, one fourth line 508 and a fifth line 510, wherein first row 502 includes a plurality of pixel 502a, 502b, 502c, 502d, 502e, second row 504 includes a plurality of pixel 504a, 504b, 504c, 504d, 504e, the third line 506 includes a plurality of pixel 506a, 506b, 506c, 506d, 506e, fourth line 508 includes a plurality of pixel 508a, 508b, 508c, 508d, 508e, fifth line 510 includes a plurality of pixel 510a, 510b, 510c, 510d, 510e.In addition, a plurality of pixels in first row 502, the third line 506 and the fifth line 508 all have same pre-color composition value, and a plurality of pixels in second row 504 and the third line 506 all have same pre-color composition value, and the pre-color composition value of first row 502 is inequality in the pre-color composition value of second row 504, the horizontal stripe pattern that it is spacing that color filter battle array 500 presents with a pixel.In addition, please also refer to color filter battle array 500 shown in Figure 5 and color filter battle array 300 shown in Figure 3, and divide other pre-color composition value v with color filter battle array more than 300 pixel V2, G2, V3, G3, V4
2, g
2, v
3, g
3, v
4A plurality of pixel 506a, 506b, 506c, 506d, 506e as opposite position in the color filter battle array 500 divides other pre-color composition value respectively, and divides other pre-color composition value v with color filter battle array more than 300 pixel V1, G1, V3, G4, V5
1, g
1, v
3, g
4, v
5A plurality of pixel 502c, 504c, 506c, 508c, 508c as opposite position in the color filter battle array 500 divides other pre-color composition value respectively.Therefore, when utilizing formula of the present invention (3) and formula (4) to come the first judgment value dh of calculating pixel 506c (being object pixel) respectively and the second judgment value dv, the first judgment value dh can be 0, and the second judgment value dv can not be 0.Thus, this level difference of pixel 506c (being object pixel) that just can judge color filter battle array 500 is smaller (be dh＜dv), so the object color component composition value of pixel 506c just can utilize formula (5) to calculate.Same, compared to pixel processing method 200 of the present invention, the first judgment value dh and the second judgment value dv that the traditional images processing method is calculated are 0, and make the traditional images processing method utilize formula (7) to come the object color component composition value of computing pixel 506c, so will obtain relatively poor, more false object color component composition value.
Note that pixel processing method 200 of the present invention is not to be subject to utilize which kind of means to implement.In other words, any step in the pixel processing method 200 all can be implemented with the combination of hardware means, firmware means, software approach or above-mentioned means, and it also is scope of the present invention place.
In sum, spirit of the present invention mainly is 2 points: the first, be used as judging one of the factor of the level difference of this object pixel according to a pre-color composition value of an object pixel and the difference value between a pre-color composition value of the surrounding pixel on the horizontal direction; Second, be used as judging one of the factor of the vertical differentiation of this object pixel according to this pre-color composition value of this object pixel and the difference value between a pre-color composition value of the surrounding pixel on the vertical direction, wherein this pre-color composition value of this pre-color composition value of this object pixel and this surrounding pixel corresponding different hue component respectively.
The above only is preferred embodiment of the present invention, and all equalizations of doing according to the present patent application claim change and modify, and all should belong to covering scope of the present invention.
Claims (5)
1. a plurality of pixels of handling a corresponding color filter array have the pixel processing method of the hue component that is lacked in each pixel of single color composition with reconstruction, it is characterized in that, include:
Decide one first residual quantity according to a pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the first direction of an object pixel, wherein this pre-color composition value of this object pixel is distinguished corresponding different hue component with this pre-color composition value of this surrounding pixel on being positioned at this first direction;
This pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the second direction according to this object pixel decide one second residual quantity, this first direction is orthogonal to this second direction, wherein this pre-color composition value of this object pixel be positioned at this second direction on this pre-color composition value corresponding different hue component respectively of this surrounding pixel, and this pre-color composition value of this surrounding pixel on this pre-color composition value of this surrounding pixel on this first direction and this second direction is identical, and the extension of the extension of this first direction and this second direction intersects at this object pixel;
Produce one first judgment value according to the pre-color composition value and this first residual quantity that are positioned at a plurality of pixels on this first direction;
Produce one second judgment value according to the pre-color composition value and this second residual quantity that are positioned at a plurality of pixels on this second direction; And
According to the comparative result of this first, second judgment value, select the pre-color composition value of a plurality of specific pixel to calculate an object color component composition value of the object color component composition that this object pixel lacks, and store this object color component composition value; Wherein,
Determine the step of this first residual quantity to include:
Read one first pre-color composition value of one first surrounding pixel that is positioned on this first direction and one second pre-color composition value of one second surrounding pixel, wherein this first, second surrounding pixel all is close to this object pixel;
Calculate one first difference value of this pre-color composition value of this first pre-color composition value and this object pixel; Calculate one second difference value of this pre-color composition value of this second pre-color composition value and this object pixel;
Add up this first, second difference value and add total value;
This is added total value produce one and add total value after adjusting divided by adjusting coefficient, and the total value that adds after this adjustment is carried out a signed magnitude arithmetic(al) and produced this first residual quantity; Or this is added total value carry out a signed magnitude arithmetic(al) and produce an operation result, and this operation result is produced this first residual quantity divided by this adjustment coefficient, wherein this adjustment coefficient equals 1/2
N, N is a positive integer;
Determine the step of this second residual quantity to include:
Read one first pre-color composition value of one first surrounding pixel that is positioned on this second direction and one second pre-color composition value of one second surrounding pixel, wherein this first, second surrounding pixel all is close to this object pixel;
Calculate one first difference value of this pre-color composition value of this first pre-color composition value and this object pixel;
Calculate one second difference value of this pre-color composition value of this second pre-color composition value and this object pixel;
Add up this first, second difference value and add total value; And
This is added total value produce one divided by this adjustment coefficient and add total value after adjusting, and the total value that adds after this adjustment is carried out a signed magnitude arithmetic(al) and produced this second residual quantity; Or this is added total value carry out a signed magnitude arithmetic(al) and produce an operation result, and this operation result is produced this second residual quantity divided by this adjustment coefficient, wherein this adjustment coefficient system equals 1/2
N, N is a positive integer;
In the step that produces one first judgment value, second judgment value, calculate this first judgment value, second judgment value according to formula 3, formula 4 respectively:
dh=abs(v
3-v2)+abs(v
3-v
4)+abs(g
2-g
3)+abs((g
2+g
3-2*v
3)/(2
N))； (3)
dv=abs(v
3-v1)+abs(v
3-v
5)+abs(g
1-g
4)+abs((g1+g
4-2*v
3)/(2
N))； (4)
Wherein, dh is first judgment value, v
3The pre-color that is this object pixel becomes score value, v
2, g
2, g
3, v
4Be respectively this object pixel and be positioned at the pre-color composition value of four surrounding pixels on this first direction, v
1, g
1, g
4, v
5Be respectively this object pixel and be positioned at the pre-color composition value of four surrounding pixels on this second direction;
According to the comparative result of this first, second judgment value, the step of selecting the pre-color composition value of a plurality of specific pixel to calculate an object color component composition value of the object color component composition that this object pixel lacks comprises:
If first judgment value is less than second judgment value, then calculate an object color component composition value g of the object color component composition that these object pixels lack according to formula 5
33:
g
33=((g
2+g
3)/2)+((2*v
3-v
2-v
4)/4)； (5)
If first judgment value is greater than second judgment value, then calculate an object color component composition value g of the object color component composition that these object pixels lack according to formula 6
33:
g
33=((g
1+g
4)/2)+((2*v
3-v
1-v
5)/4)； (6)
If first judgment value equals second judgment value, then calculate an object color component composition value g of the object color component composition that these object pixels lack according to formula 7
33:
g
33=((g
1+g
2+g
3+g
4)/4)+((4*v
3-v
1-v
2-v
4-v
5)/16) (7)。
2. pixel processing method according to claim 1 is characterized in that, this first direction is a horizontal direction, and this second direction is a vertical direction.
3. pixel processing method according to claim 1, it is characterized in that this pre-color composition value of four surrounding pixels on this object color component composition value, this first direction and this pre-color composition value of four surrounding pixels on this second direction are the hue component value of a green.
4. pixel processing method according to claim 1 is characterized in that, this pre-color composition value of this object pixel is the hue component of a redness or the hue component of a blueness.
5. an image processing system is characterized in that, comprises: an imageing sensor, produce the color filter array with a plurality of pixels, and wherein each pixel has single colour component; And an image-signal processor, in order to receiving these a plurality of pixels, and the pixel of rebuilding the hue component that is lacked in this each pixel with following method, as follows:
Decide one first residual quantity according to a pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the first direction of an object pixel, wherein this pre-color composition value of this object pixel is distinguished corresponding different hue component with this pre-color composition value of this surrounding pixel on being positioned at this first direction;
This pre-color composition value and the difference between a pre-color composition value of at least one surrounding pixel on the second direction according to this object pixel decide one second residual quantity, this first direction is orthogonal to this second direction, wherein this pre-color composition value of this object pixel be positioned at this second direction on this pre-color composition value corresponding different hue component respectively of this surrounding pixel, and this pre-color composition value of this surrounding pixel on this pre-color composition value of this surrounding pixel on this first direction and this second direction is identical, and the extension of the extension of this first direction and this second direction intersects at this object pixel;
Produce one first judgment value according to the pre-color composition value and this first residual quantity that are positioned at a plurality of pixels on this first direction;
Produce one second judgment value according to the pre-color composition value and this second residual quantity that are positioned at a plurality of pixels on this second direction; And
According to the comparative result of this first, second judgment value, select the pre-color composition value of a plurality of specific pixel to calculate an object color component composition value of the object color component composition that this object pixel lacks, and store this object color component composition value; Wherein,
Determine the step of this first residual quantity to include:
Read one first pre-color composition value of one first surrounding pixel that is positioned on this first direction and one second pre-color composition value of one second surrounding pixel, wherein this first, second surrounding pixel all is close to this object pixel;
Calculate one first difference value of this pre-color composition value of this first pre-color composition value and this object pixel;
Calculate one second difference value of this pre-color composition value of this second pre-color composition value and this object pixel;
Add up this first, second difference value and add total value;
This is added total value produce one and add total value after adjusting divided by adjusting coefficient, and the total value that adds after this adjustment is carried out a signed magnitude arithmetic(al) and produced this first residual quantity; Or this is added total value carry out a signed magnitude arithmetic(al) and produce an operation result, and this operation result is produced this first residual quantity divided by this adjustment coefficient, wherein this adjustment coefficient equals 1/2
N, N is a positive integer;
Determine the step of this second residual quantity to include:
Read one first pre-color composition value of one first surrounding pixel that is positioned on this second direction and one second pre-color composition value of one second surrounding pixel, wherein this first, second surrounding pixel all is close to this object pixel; Calculate one first difference value of this pre-color composition value of this first pre-color composition value and this object pixel; Calculate one second difference value of this pre-color composition value of this second pre-color composition value and this object pixel;
Add up this first, second difference value and add total value; And
This is added total value produce one divided by this adjustment coefficient and add total value after adjusting, and the total value that adds after this adjustment is carried out a signed magnitude arithmetic(al) and produced this second residual quantity; Or this is added total value carry out a signed magnitude arithmetic(al) and produce an operation result, and this operation result is produced this second residual quantity divided by this adjustment coefficient, wherein this adjustment coefficient system equals 1/2
N, N is a positive integer;
In the step that produces one first judgment value, second judgment value, calculate this first judgment value, second judgment value according to formula 3, formula 4 respectively:
dh=abs(v
3-v2)+abs(v3-v
4)+abs(g
2-g
3)+abs((g
2+g
3-2*v
3)/(2
N))； (3)
dv=abs(v
3-v1)+abs(v
3-v
5)+abs(g
1-g
4)+abs((g1+g
4-2*v
3)/(2
N))； (4)
Wherein, dh is first judgment value, v
3The pre-color that is this object pixel becomes score value, v
2, g
2, g
3, v
4Be respectively this object pixel and be positioned at the pre-color composition value of four surrounding pixels on this first direction, v
1, g
1, g
4, v
5Be respectively this object pixel and be positioned at the pre-color composition value of four surrounding pixels on this second direction;
According to the comparative result of this first, second judgment value, the step of selecting the pre-color composition value of a plurality of specific pixel to calculate an object color component composition value of the object color component composition that this object pixel lacks comprises:
If first judgment value is less than second judgment value, then calculate an object color component composition value g of the object color component composition that these object pixels lack according to formula 5
33:
g
33=((g
2+g
3)/2)+((2*v
3-v
2-v
4)/4)； (5)
If first judgment value is greater than second judgment value, then calculate an object color component composition value g of the object color component composition that these object pixels lack according to formula 6
33:
g
33=((g
1+g
4)/2)+((2*v
3-v
1-v
5)/4)； (6)
If first judgment value equals second judgment value, then calculate an object color component composition value g of the object color component composition that these object pixels lack according to formula 7
33:
g
33=((g
1+g
2+g
3+g
4)/4)+((4*v
3-v
1-v
2-v
4-v
5)/16) (7)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910109109 CN101969571B (en) | 2009-07-27 | 2009-07-27 | Pixel processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910109109 CN101969571B (en) | 2009-07-27 | 2009-07-27 | Pixel processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101969571A CN101969571A (en) | 2011-02-09 |
| CN101969571B true CN101969571B (en) | 2013-08-21 |
Family
ID=43548651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910109109 Active CN101969571B (en) | 2009-07-27 | 2009-07-27 | Pixel processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101969571B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170140187A (en) * | 2015-04-23 | 2017-12-20 | 오스텐도 테크놀로지스 인코포레이티드 | Method for fully parallax compression optical field synthesis using depth information |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1968422A (en) * | 2005-11-17 | 2007-05-23 | 八方科技股份有限公司 | Color interpolation apparatus and method for reconstructing missing colors utilizing image edge indicators |
| US7333678B1 (en) * | 2003-05-20 | 2008-02-19 | Micronas Usa, Inc. | Edge adaptive demosaic system and method |
| KR20090015566A (en) * | 2007-08-09 | 2009-02-12 | 연세대학교 산학협력단 | Method and apparatus for color interpolation |
| CN101399996A (en) * | 2007-09-27 | 2009-04-01 | 比亚迪股份有限公司 | Color filter lens array interpolation method |
-
2009
- 2009-07-27 CN CN 200910109109 patent/CN101969571B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7333678B1 (en) * | 2003-05-20 | 2008-02-19 | Micronas Usa, Inc. | Edge adaptive demosaic system and method |
| CN1968422A (en) * | 2005-11-17 | 2007-05-23 | 八方科技股份有限公司 | Color interpolation apparatus and method for reconstructing missing colors utilizing image edge indicators |
| KR20090015566A (en) * | 2007-08-09 | 2009-02-12 | 연세대학교 산학협력단 | Method and apparatus for color interpolation |
| CN101399996A (en) * | 2007-09-27 | 2009-04-01 | 比亚迪股份有限公司 | Color filter lens array interpolation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101969571A (en) | 2011-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1215455C (en) | Colour signal processing equipment for multi-primary colours display device | |
| CN100345181C (en) | Converson of a sub-pixel format data to another sub-pixel data format | |
| CN1302669C (en) | Device and method of colour compensation | |
| JP5037311B2 (en) | Color reproduction system and method | |
| CN1664905A (en) | Apparatus and method of converting image signal for four-color display device, and display device including the same | |
| CN101076848A (en) | Method and equipment for converting into target color space from source color space | |
| CN101901474B (en) | Change the method for at least one in the density of image and contrast | |
| CN1661664A (en) | Method for displaying an image, image display apparatus, method for driving an image display apparatus and apparatus for driving an image display panel | |
| CN102842276B (en) | Uneven check device and uneven inspection method | |
| CN108346147B (en) | Technical method for quickly, automatically and accurately identifying coal rock micro-components | |
| CN1658680A (en) | Image signal correction method and image signal correction apparatus | |
| TW200849971A (en) | Color calibrating method for setting target gamma curves of target display device | |
| WO2012116468A1 (en) | Image transmission and displaying method based on principle of chroma and visual fidelity | |
| KR20110100145A (en) | Color-unevenness inspection apparatus and method | |
| CN101969571B (en) | Pixel processing method | |
| JP4974607B2 (en) | Display device, display system, and RGB signal processing method | |
| CN1288918C (en) | Coloured signal conversion device and its method | |
| CN101251994A (en) | Chromatic correction method for setting target GAMMA curve of target display apparatus | |
| CN108053796A (en) | The driving method and display device of display panel | |
| CN110383801B (en) | Color saturation adjustment in non-uniform color spaces | |
| CN105657391B (en) | A kind of method and device of image procossing | |
| CN111179801A (en) | Color space adjustment method, device and system of display panel | |
| CN101854558A (en) | Method for correcting color domain in treating video signals of LED display screen | |
| JP2006048037A (en) | Apparatus and method for rendering image, and computer-readable recording media for storing image rendering program | |
| CA2417335A1 (en) | Image and video processing with chrominance attenuation |
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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: Chegongmiao Futian Tian'an Science and Technology Venture Park B901, B902 and B903, Futian District, Shenzhen City, Guangdong Province Co-patentee after: Silicon Motion, Inc. Patentee after: Huirong Technology (Shenzhen) Co.,Ltd. Address before: 518040 Guangdong, Shenzhen, Futian District, Che Kung Temple Futian Tian Tian science and Technology Pioneer Park B901, B902, B903 Co-patentee before: Silicon Motion, Inc. Patentee before: SILICON MOTION SHENZHEN Inc. |