CN1905626A - Method and apparatus for removing dither process of image signal - Google Patents
Method and apparatus for removing dither process of image signal Download PDFInfo
- Publication number
- CN1905626A CN1905626A CN 200610029648 CN200610029648A CN1905626A CN 1905626 A CN1905626 A CN 1905626A CN 200610029648 CN200610029648 CN 200610029648 CN 200610029648 A CN200610029648 A CN 200610029648A CN 1905626 A CN1905626 A CN 1905626A
- Authority
- CN
- China
- Prior art keywords
- error
- pixel
- signal
- picture signal
- dithering process
- 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.)
- Pending
Links
Images
Landscapes
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The invention relates to a method and device used for de-jitter handling about the format-transformed image signals. The existing methods have the defects on wasting resource of the buffer and hardware complexity. The device has an error handling module, buffer and image generation module. The image signals are input into the error computing unit and the image generation unit respectively, in which, each pixel carries out error-diffusion in the direction of right and down, and the diffused errors are determined by the value of the pixel and the jitter-coefficient in each direction. The error computing unit saves the sum value calculated from the diffused errors of three adjacent pixels in the same row into the buffer as a signal ERRO. When the next row of image signals is input, the image generation unit computes new de-dithered image signals by the input signal and the ERRO to be output.
Description
Technical field
The present invention relates to a kind of picture signal and remove the dithering process method and apparatus, particularly a kind of being used for removed the dithering process method and apparatus to what the picture signal after the format conversion was carried out error diffusion.
Background technology
Because the pixel bit depth of LCDs (LCD) is generally less than 24, so need cut out the Pixel Information of low level when being shown to 24 image on the LCDs.With the RGB565 LCDs is example, and the figure place of 565 expressions red (R), green (G) and blue (B) passage is respectively 5,6,5, and the figure place addition of three color channels of RGB equals 16, represents that promptly this signal has 16 (bit).If 24 RGB888 signals are shown on 16 the RGB565 LCDs, because 8 R channel signals can only show 5, so need cut out low 3 signal, same, the G passage will cut out low 2, and the B passage will cut out and hang down 3.If only adopt simple cut position method, can make image produce fixing error color lump, therefore, adopt usually in the prior art and remove the error diffusion method shaken, make the cut position quantization error be distributed to entire image, thereby avoid the appearance of error color lump.
Shown in Fig. 1 (a), the basic principle of going dither image to handle be the cut position quantization error with each pixel in the image reach to the right respectively the bottom right, just down, lower left (being referred to as the below) diffusion, the propagated error of generation is determined by the value of this pixel and the shake coefficient a on all directions, b, c, d.The general computing formula of this cut position quantization error is:
ERROR_C (i, j)=C_IN (i, j)-(((C_IN (i, j)＞＞n)+2
n)＜＜n) (1) wherein, C_IN (i, j) denotation coordination is that (C can replace with R, G, B for i, the input value of a certain color component of pixel j); N represents that this color component needs the figure place of cut position;＞＞n n the position of representing to move to right,＜＜the n n position of representing to move to left.
With the RGB888 conversion of signals is that the RGB565 signal is an example, then in the picture signal coordinate be (i, the cut position quantization error that each color component produced of pixel j) is respectively:
R component: ERROR_R (i, j)=R_IN (i, j)-(((R_IN (i, j)＞＞3)+8)＜＜3);
G component: ERROR_G (i, j)=G_IN (i, j)-(((G_IN (i, j)＞＞2)+4)＜＜2);
B component: ERROR_B (i, j)=B_IN (i, j)-(((B_IN (i, j)＞＞3)+8)＜＜3).
Only explain at the error diffusion of R component below, G, the rest may be inferred for the B component, then coordinate be (i, pixel j) disperse and the neighbor on every side of being added on propagated error be:
Right side: R_DIS (i, j+1)=ERROR_R (i, j) * a
Bottom right: R_DIS (i+1, j+1)=ERROR_R (i, j) * b
Just down: R_DIS (i+1, j)=ERROR_R (i, j) * c
Lower-left: R_DIS (i+1, j-1)=ERROR_R (i, j) * d (2)
Shown in Fig. 1 (b), for the arbitrary pixel in the picture signal (i, j), be positioned at its left side, upper left, just go up and top-right four pixels (i, j-1), (i-1, j-1), (i-1, j) with (i-1 j+1) has carried out error diffusion to it respectively.These propagated errors be added to this original pixels (i, j) on, promptly obtain the new pixel value after going dithering process:
R_NEW(i，j)＝R_IN(i，j)+ERROR_R(i，j-1)*a+ERROR_R(i-1，j-1)*b+ERROR_R(i-1，j)*c+ERROR_R(i-1，j+1)*d (3)
The existing dither image processing method of going is promptly according to above-mentioned principle, at first the pixel in the original RGB888 signal (24) is carried out the buffer memory of delegation, calculate the propagated error on all directions respectively according to shake coefficient a, b, c, d again, on the pixel around at last these propagated errors being added to generate new pixel value.Usually, the value of shaking coefficient is made as 7/16,3/16,5/16 and 1/16, must finishes jointly by shift unit and multiplier during according to formula (1) and (2) calculating propagated error, hardware configuration is comparatively complicated.In addition, because traditional dither image processing method of going must be carried out buffer memory to 24 original signals,, cause the wasting of resources so the spatial cache that takies is bigger.
Summary of the invention
The object of the present invention is to provide a kind of method and apparatus that is used for picture signal is gone dithering process, it not only can simplify hardware configuration, only adopts shift unit can finish the calculating of propagated error, and can save the storage resources of buffer greatly.
The object of the present invention is achieved like this: a kind of method that is used for picture signal is gone dithering process, wherein, each pixel of picture signal all to the right, the bottom right, error diffusion is carried out in just and lower left, the propagated error that produces is by the value of this pixel and the shake coefficient a on all directions, b, c, d determines, its substantive distinguishing features is, described method comprises the following steps: that (1) receives the capable image input signal of i-1, calculate respectively that coordinate is (i-1 in this row, j-1), (i-1, j) reach (i-1, the propagated error of adjacent three pixels j+1), wherein, i, j is nonnegative integer; (2) with the propagated error of adjacent three pixels in the step (1) and be stored in the buffer as an error signal ERRO; (3) receive the capable image input signal of i, coordinates computed is (i, the propagated error of pixel j-1), and be that (i, the value of pixel j) calculate pixel (i, j) the new value after going dithering process in conjunction with error signal ERRO and coordinate; And (4) get different values to i, j, repeats above-mentioned steps (1) to (3), and each pixel in this image is all till the past dithering process.
In the above-mentioned method that is used for picture signal is gone dithering process, the shake coefficient a on the described all directions, b, c, d value respectively are 1/2,1/8,1/4 and 1/8.
Another solution of the present invention is: a kind of device that is used for picture signal is gone dithering process is provided, it comprises an error calculation unit, one buffer and an image generation unit, picture signal inputs to this error calculation unit and this image generation unit respectively, each pixel in the picture signal all to the right, the bottom right, error diffusion is carried out in just and lower left, the propagated error that produces is by the value of this pixel and the shake coefficient a on all directions, b, c, d determines, the propagated error of described error calculation unit three adjacent pixels in will picture signal with delegation and be stored in the buffer as an error signal ERRO, after the input of next line picture signal, this image generation unit utilizes input signal and error signal ERRO to calculate the new picture signal after going dithering process and exports.
At the above-mentioned device that is used for picture signal is gone dithering process, described error calculation unit is made of shift unit and adder.
At the above-mentioned device that is used for picture signal is gone dithering process, the shake coefficient a on the described all directions, b, c, d value respectively are 1/2,1/8,1/4 and 1/8.
The dithering process method and apparatus that goes of the present invention makes error diffusion process only use shift unit to realize by adopting the shake coefficient after subtracting approximately, has saved multiplier, has simplified hardware configuration.Simultaneously, by with error signal ERRO but not primary signal be stored in the buffer, can save the storage resources of buffer greatly, reach the effect that reduces hardware cost.
Description of drawings
Method flow of the present invention and apparatus structure are provided by following embodiment and accompanying drawing.
Fig. 1 (a) and Fig. 1 (b) are for removing the schematic diagram of dither image processing method;
Fig. 2 is the structure chart that removes the dithering process device of the present invention;
Fig. 3 is the workflow diagram that goes the dithering process method of the present invention;
Fig. 4 is the circuit block diagram that removes the dithering process device of a preferred embodiment of the present invention.
Embodiment
Go the method and apparatus of dithering process to be described in further detail a kind of being used for of the present invention to picture signal below with reference to accompanying drawing.
Referring to Fig. 1 (a) and Fig. 1 (b), the dithering process method of going of the present invention has identical principle with the existing dithering process method of going, be in the picture signal each pixel all to the right, the bottom right, just down and the lower left carry out error diffusion, the propagated error that produces is determined by the value of this pixel and the shake coefficient a on all directions, b, c, d, R component with a certain pixel is an example, and the new R component value of this pixel after going dithering process also meets aforesaid computing formula:
R_NEW(i，j)＝R_IN(i，j)+ERROR_R(i，j-1)*a+ERROR_R(i-1，j-1)*b+ERROR_R(i-1，j)*c+ERROR_R(i-1，j+1)*d (3)
Handle because back three expressions of equation (3) with the propagated error of adjacent three pixels of delegation, can merge it, if will be back three with count an error signal ERRO_R (i, j), then above-mentioned equation (3) becomes:
R_NEW(i，j)＝R_IN(i，j)+ERROR_R(i，j-1)*a+ERRO_R(i，j) (4)
The dithering process method of going of the present invention promptly is based on equation (4), by (i j) is stored in the buffer, again in conjunction with R_IN (i with error signal ERRO_R, j) and ERROR_R (i, j-1) value of * a calculates the new value of each pixel after going dithering process.
Referring to Fig. 2 and Fig. 3, describe workflow of removing the dithering process method and apparatus of the present invention in detail.For convenience of description, adopt the coordinate figure of pixel to indicate locations of pixels, if negative coordinate illustrates that then this pixel does not exist.
Is example with the RGB888 conversion of signals to the RGB565 signal, if will calculate arbitrary pixel (i, j) (i, j is nonnegative integer) value after going dithering process, then need pass through following each step: at first, receive the capable image input signal of i-1 (step S10) by error calculation unit (1), utilize formula (1) and (2) to calculate pixel (i, j) three adjacent pixel (i-1 of top, j-1), (i-1, j), (i-1 is j+1) to the propagated error (step S20) of its contribution, then, three propagated errors are sued for peace, and be stored in the buffer (2) (step S30) as an error signal ERRO.In present embodiment, because the displacement figure place of R, G, B component is respectively 3,2,3, so error signal ERRO_R (i of R component, j) take the 3bit space, the error signal ERRO_G of G, B component (i, j), (i j) takies 2bit and 3bit respectively to ERRO_B, for a pixel, the buffer space that it takies only is 3+2+3=8bit.
After error calculation unit (1) and image generation unit (3) receive the capable image input signal of i (step S40), (i, j) (i is j-1) to the propagated error of its contribution for the left side adjacent pixels by error calculation unit (1) calculating pixel.The error signal ERRO of storage before image generation unit (3) reads from buffer (2), and in conjunction with pixel (i, input value j) and pixel (i, propagated error j-1), utilize equation (4) to calculate pixel (i, j) the new value (step S50) after going dithering process.
Because i, j are nonnegative integers arbitrarily, can go dithering process to each pixel in the image by the value that changes i, j, general processing sequence is from the image upper left corner (0,0) pixel begins to move right, finishing behind the error signal calculation of delegation and the buffer memory line down handles again, so to the right, downwards line by line the diffusion, up to all pixels all till the past dithering process.
Referring to Fig. 4, it shows the circuit block diagram that removes the dithering process device of a preferred embodiment of the present invention.Among the figure, on behalf of arbitrary needs, open circle P5 go the pixel of dithering process; Solid rim P1-P4 representative is adjacent with P5 and to four pixels of P5 contribution propagated error, wherein, P1, P2, P3 are the pixel of P5 lastrow, and P4 is the pixel of going together with P5.In order to simplify hardware configuration, special P1, P2, P3, P4 carry out shake coefficient b, c, d, a that error diffusion adopted to P5 and are set to 1/8,1/4,1/8 and 1/2.
At first, need utilize formula (1) and (2) to calculate the propagated error of adjacent three pixels in P5 top to its contribution.Behind the value substitution formula with shake coefficient b, c, d, obtain P1, P2, P3 is respectively the propagated error of P5 contribution: ERROR (P1) * 1/8, ERROR (P2) * 1/4 and ERROR (P3) * 1/8.
For a binary number, multiply by 1/8 and can realize that multiply by 1/4 can be by 2 realization that move to right by moving to right 3.Therefore, when calculating above-mentioned three propagated errors, only need to adopt a shift unit move to right 3,2 and 3 respectively, need not to use multiplier can finish computing, simplified hardware configuration ERROR (P1), ERROR (P2) and ERROR (P3) signal.Three propagated errors through shift unit output input to the adder summation, and summed result ERRO (8bit) is stored in the buffer (Buffer), than the 24bit saving 16bit of prior art.
After next line picture signal input, from Buffer, take out ERRO, and and the pixel value of P5 and P4 to the propagated error of P5 contribution together, via adder stack output, this output signal P5 ' is the new value of P5 after going dithering process.Wherein, because shake coefficient a value is 1/2, the propagated error that P4 contributes P5 only needs ERROR (P4) signal is moved to right 1 and can obtain.
A kind of being used for of the present invention goes the method and apparatus of dithering process to be applicable to the cut position processing of multiple different-format to picture signal, if when being shown to the RGB888 signal on the RGB666 LCDs, then each color channel all will cut out low 2, only " n=2 " substitution aforementioned formula (1) to (4) can need be gone dithering process, wherein, when calculating the new value of each pixel, the required spatial cache that takies only is: 2+2+2=6bit.Similarly, when the RGB888 conversion of signals is the RGB444 signal, also can go dithering process according to same operating process, its spatial cache that takies is: 4+4+4=12bit, all the 24bit than prior art saves memory space over half, has saved memory resource greatly.
Claims (5)
1, a kind of method that is used for picture signal is gone dithering process, wherein, each pixel of picture signal all to the right, the bottom right, just down and the lower left carry out error diffusion, the propagated error that produces is determined by the value of this pixel and the shake coefficient a on all directions, b, c, d, it is characterized in that described method comprises the following steps:
(1) receive the capable image input signal of i-1, calculate respectively in this row coordinate for (i-1, j-1), (i-1, j) and (wherein, i, j are nonnegative integer for i-1, the propagated error of adjacent three pixels j+1);
(2) with the propagated error of adjacent three pixels in the step (1) and be stored in the buffer as an error signal ERRO;
(3) receive the capable image input signal of i, coordinates computed is (i, the propagated error of pixel j-1), and be that (i, the value of pixel j) calculate pixel (i, j) the new value after going dithering process in conjunction with error signal ERRO and coordinate; And
(4) i, j are got different values, repeat above-mentioned steps (1) to (3), each pixel in this image is all till the past dithering process.
2, the method that is used for picture signal is gone dithering process as claimed in claim 1 is characterized in that: the shake coefficient a on the described all directions, b, c, d value respectively are 1/2,1/8,1/4 and 1/8.
3, a kind of device that is used for picture signal is gone dithering process, comprise: an error calculation unit (1), one buffer (2) and an image generation unit (3), picture signal inputs to this error calculation unit (1) and this image generation unit (3) respectively, each pixel in the picture signal all to the right, the bottom right, error diffusion is carried out in just and lower left, the propagated error that produces is by the value of this pixel and the shake coefficient a on all directions, b, c, d determines, it is characterized in that: the propagated error of described error calculation unit (1) three adjacent pixels in will picture signal with delegation and be stored in the buffer (2) as an error signal ERRO, after the input of next line picture signal, this image generation unit (3) utilizes input signal and error signal ERRO to calculate the new picture signal after going dithering process and exports.
4, the device that is used for picture signal is gone dithering process as claimed in claim 3 is characterized in that: described error calculation unit (1) is made of shift unit and adder.
5, the device that is used for picture signal is gone dithering process as claimed in claim 3 is characterized in that: the shake coefficient a on the described all directions, b, c, d value respectively are 1/2,1/8,1/4 and 1/8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200610029648 CN1905626A (en) | 2006-08-02 | 2006-08-02 | Method and apparatus for removing dither process of image signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200610029648 CN1905626A (en) | 2006-08-02 | 2006-08-02 | Method and apparatus for removing dither process of image signal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1905626A true CN1905626A (en) | 2007-01-31 |
Family
ID=37674742
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200610029648 Pending CN1905626A (en) | 2006-08-02 | 2006-08-02 | Method and apparatus for removing dither process of image signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1905626A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036746A (en) * | 2014-06-10 | 2014-09-10 | 深圳市长江力伟股份有限公司 | Method for image debounce of LCOS |
| CN110992242A (en) * | 2019-10-26 | 2020-04-10 | 武汉光庭信息技术股份有限公司 | Method and device for eliminating water ripples of transparent picture |
| CN112185312A (en) * | 2020-09-29 | 2021-01-05 | 珠海格力电器股份有限公司 | Image data processing method and device |
-
2006
- 2006-08-02 CN CN 200610029648 patent/CN1905626A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036746A (en) * | 2014-06-10 | 2014-09-10 | 深圳市长江力伟股份有限公司 | Method for image debounce of LCOS |
| CN110992242A (en) * | 2019-10-26 | 2020-04-10 | 武汉光庭信息技术股份有限公司 | Method and device for eliminating water ripples of transparent picture |
| CN112185312A (en) * | 2020-09-29 | 2021-01-05 | 珠海格力电器股份有限公司 | Image data processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1254086C (en) | Image processor, image processing method and processing program | |
| EP0608053B1 (en) | Colour display system | |
| WO2018166152A1 (en) | Scanning card, led display screen control system and image data processing method | |
| US20060215929A1 (en) | Methods and apparatus for image convolution | |
| JPH0937214A (en) | Progressive scanning conversion method and progressive scanning converter | |
| CN1092924A (en) | The method and apparatus of field rate upconversion | |
| CN101076995A (en) | Picture processing based on error dirrusion | |
| CN1905626A (en) | Method and apparatus for removing dither process of image signal | |
| CN1655626A (en) | System and method for clipping values of pixels | |
| WO2019119572A1 (en) | Compensation table compression method | |
| CN1254087C (en) | Image processor, image processing method and processing program | |
| US7474442B2 (en) | High performance coprocessor for color error diffusion halftoning | |
| JPH06332395A (en) | Method for processing image and device therefor | |
| CN1830011A (en) | Image signal processor and image signal processing method | |
| CN1870048A (en) | Edge strengthening method and device of Bel image and color image pick-up system | |
| Trager et al. | GPU-enabled parallel processing for image halftoning applications | |
| CN101057488A (en) | System and method for dark noise reduction in pulse width modulated (PWM) displays | |
| CN201877112U (en) | Device for de-jittering image signal | |
| CN1784886A (en) | Image processing for expressing gradation | |
| TWI462057B (en) | Hadamard transform-based image compression circuit and method | |
| WO2011068429A1 (en) | Detecting artifacts in quantization noise in images compresses using discrete cosine transforms | |
| CN101286302A (en) | Hardware accomplishing image de-jittering method and apparatus | |
| JPH118765A (en) | Gradation lowering processing method, processor therefor integrated circuit for gradation lowering processing, and computer-readable recording medium recorded with gradation lowering program | |
| CN100355267C (en) | Improved system and method for error diffusion imaging processing | |
| CN1163853C (en) | Multiple-order half-tone image forming method and device from input digital image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |