CN1215704C

CN1215704C - Method for processing image compensation of digital camera

Info

Publication number: CN1215704C
Application number: CNB021081689A
Authority: CN
Inventors: 叶嘉佑
Original assignee: Winbond Electronics Corp
Current assignee: Winbond Electronics Corp
Priority date: 2002-03-28
Filing date: 2002-03-28
Publication date: 2005-08-17
Anticipated expiration: 2022-03-28
Also published as: CN1449200A

Abstract

The present invention relates to an image processing method. Because distortion or loss possibly exist in the process of extraction, a distorted or lost image signal is required to be properly processed and compensated by a proper device and a method so that the image can be authentically presented. The method for processing the image compensation of a digital camera of the present invention poses the image processing method. When the image extracted by a digital camera is distorted or lost, a pixel needs to be compensated is compensated by optimum coloring matters so that the display of the image has best effect.

Description

The processing method of image compensation of digital camera

Technical field

The present invention relates to a kind of method of image processing, particularly a kind of processing method of image compensation of digital camera, this method are to utilize suitable Pixel calcualting to handle, and to the distortion of image, or lose and compensate, and obtain the most real image and present, show.

Background technology

Behind people's invention camera, the mode that image energy is done to store is just constantly improved, is developed, and the presenting also of image from black-and-white photograph the earliest, the appearance of colour motion picture films by now is as for the method for image processing, also because the progress of digital processing unit, from the signal storage mode of simulation was only arranged in the past, to the selection of signal storing mode that numeral is arranged, the pattern of another kind of image processing can be provided, to now fast, easily digital society bring image treatment method more easily.

The appearance of digital camera, bring unprecedented progress for image processing technique, used traditional camera shooting entity image in the past, though image can be stored in the photograph, but for digitized living environment nowadays, equally not unique selection in the past for another example, appearance at matching computer and other E-consumer product, the requirement of image digitization is healed gradually and is become important, and how the effect of image digitization can could be reached present really, just be the emphasis that requires of technique improvement nowadays.

Carry as preceding institute, all there is difference in each family of the employed digitized video treatment technology of common technology, the signal strength signal intensity that comprises each the basic pigment that utilizes the pixel formation is wherein also arranged, cooperate the contiguous pixel of determining to make the mode of suitable processing computing, so as preceding described, the mode that each family handles computing all has difference, and is therefore also different in work, the result of its digitized video processing.

Comprise in the common technology and use a color filtering array (Color Filter Array, CFA), utilize this color filtering array that image is detected and obtain red (Red, R), green (Green, G), blue (Blue, B) three pixels that basic pigment constituted, so, in the process that the digitized video that execution continues is handled, can be easy to do calculation process.

Summary of the invention

Main purpose of the present invention is the processing method that proposes a kind of image compensation of digital camera, this method proposes a digitized video and is utilizing suitable charge coupled cell (Charge-Coupled Device, CCD) make the signal sensing of image after, color filtering array (Color Filter Array, CFA) with the mode of this digitized video with array, signal strength signal intensity according to basic pigment shows in regular turn, and because the distortion in the inner place of resulting image pixel possibility factor word camera or the friendship anchor of pixel, cause image pixel to show really, utilize the present invention suitably to handle, the pigment signal component that cooperates the neighborhood pixels of the pixel that needs the imitating element compensation, finish the work of compensation pigment, make the complete image frame that is extracted that demonstrates truly of digital camera energy.

Another object of the present invention is the processing method that proposes a kind of image compensation of digital camera, this method utilizes color filtering array that the image pixel that charge coupled cell sensed is shown with suitable aspect, this color filtering array is that a patent right licenses to the product that technology that the Bu Laisi Bel (BryceE.Bayer) of Kodak uses is produced, the color filtering array of relevant Bayer will in after describe in detail in narration, behind the image that the aspect demonstration that utilizes this color filtering array is sensed, just can utilize suitable processing computing, the basic pigment composition that the pixel that will compensate is contained, utilize the mode of compensation to be handled, obtain best real image result.

The invention of digital camera occurs, bring unprecedented change for the technology that image stores, the pattern that the image processing meeting was simulated is in the past changed into the image processing pattern of numeral, preservation for data message, one more reliable the storage and the mode of handling is provided, right in the problem of digital camera in consideration cost and usefulness, how to accomplish to reach the most dynamical performance, just the emphasis place of improveing and designing for the digital camera dealer with most economical design.

The processing method of image compensation of digital camera of the present invention just is being that former digital camera is in the work that image extracts is carried out, the inductor that a pixel color is only arranged because of the element of carrying out image processing in the employed digital camera, detection for colors of image often causes distortion or omission, cause extract the image handled and in fact image deviation or distortion, for solving the above problems, the processing method of the long-pending image compensation of teaching machine of the present invention just proposes to use the identical digital camera that electronic component structure constituted, after utilizing suitable calculation process, obtain the optimal images of image pigment compensation, and conform to real image energy.

Description of drawings

Fig. 1 uses the form schematic diagram of Bayer color filtering array for the present invention;

Fig. 2 is a circuit box schematic diagram of the present invention;

Fig. 3 is the signal strength signal intensity schematic diagram of pixel in the embodiment of the invention;

Fig. 4 is the model schematic diagram of first embodiment of the invention;

Fig. 5 is the model schematic diagram of second embodiment of the invention;

Fig. 6 is the model schematic diagram of third embodiment of the invention;

Fig. 7 is the model schematic diagram of fourth embodiment of the invention.

Embodiment

See also Fig. 1, Fig. 1 uses the form schematic diagram of Bayer color filtering array for the present invention, because the extraction of image must have the part in a pigment sense low-lying area, the present invention uses color filtering array 100 (the Color Filter Array that is delivered by Bu Laisi Bel (BryceE.Bayer), CFA) to the detection of image pigment, wherein be that a plurality of green pixels (G) are arranged in this color filtering array 100, red pixel (R) and blue pixel (B), these a plurality of green pixels (G), red pixel (R) and the basic pigment composition of blue pixel (B) for the formation image, and in the employed in the present invention Bayer color filtering array, these a plurality of green pixels (G) 100 are the arrangement of a diagonal in this colour filter battle array is ice-cold, shown in G1 and G5, then be arranged on another diagonal of this cornerwise green pixel (G), shown in R1 and B1 as for red pixel (R) and blue pixel (B).

See also Fig. 2, Fig. 2 is employed circuit box schematic diagram among the present invention, comprising there being an image to extract district's 1 and one image process area 2, and also include a charge coupled cell 11 in the part in this image extraction district 1, one image sensor 12, one analog/digital converter 13 and an image buffer 14, do the scanning of the image pigment of simulation by 11 pairs of images of this charge coupled cell, mode with electric current drives this image sensor 12, and receive the image pixel sensed current signal of being imported by this charge coupled cell 11 by this image sensor 12, and export suitable current signal to this analog-digital converter 13, the current signal of simulation is converted to the data format signal of numeral by this analog/digital converter, and utilize this image buffer 14 that the image data signal of numeral is stored, the image process area 2 that offers next stage is made the computing compensation work of suitable pixel color.

Also include a digital signal processor 15 and an erasable type holder 16 in this image process area 2 of Fig. 2, after wherein finishing by this image extraction district 1 performed image extraction king, import in this image process area 2 of suitable digital image signal, be to utilize this digital signal processor 15 to do suitable Digital Signal Processing computing, cooperate pixel compensation calculation process of the present invention to carry out, the picture element signal that will compensate is made suitable compensation to be handled, and export this erasable type holder 16 to, and cooperate in digital camera other circuit element unit by this erasable type holder, will do best demonstration through the image after the pixel color compensation and present.

See also Fig. 3, Fig. 3 is a schematic diagram of choosing the pixel extraction analog signal of a level or vertical direction in the look display 100 used in the present invention wantonly, wherein the longitudinal axis is a pixel signal intensities 10 in signal schematic representation, transverse axis then is capture position 20, and in the signal strength signal intensity that shown in this Fig. 3 is green pixel 21 (G) and red pixel 22 (R), the pixel signal intensities of being extracted in different capture positions 20 10 is also different.

See also Fig. 4, Fig. 4 pattern diagram for first embodiment of the invention, be to include a pixel A 40 in this first embodiment, this pixel A 40 is this first embodiment will carry out the pixel cell that pixel compensation is handled, and the digital image data signal of the level that these digital signal processor 15 required digitized video pixel datas in circuit must include this pixel A 40 and vertical adjacent other pixel G41, G43, G42, G44 composition, with level and vertical homochromy pixel A 41, A43, A42, the A44 of adjacent two pixels.

This image extracts the district and obtains this pixel A 40 and adjacent pixel G41 on every side, behind G43, G42, G44, A41, A43, A42, the A44 digital data signal, utilize this image process area 2 to make suitable pigment compensation operation and handle, wherein more utilize the parameter of a vertical green pixel aberration of horizontal green pixel aberration (GHd) (GVd) to provide this digital signal processor 15 to make the suitably judgement parameter of work executing rule with one; This horizontal green pixel aberration (GHd) be the green absolute difference of a level of adjacent horizontal green pixel (G41, G43) (| G41-G43|), and this pixel of twice level two of deducting adjacent two pixels with color pixel and the total value (| G41-G43|+|2 * A40-(A41+A43) |/4) of 1/4th (| 2 * A40-(A41+A43) |/4) addition of absolute difference, its formula is:

GHd＝|G41-G43|+|2×A40-(A41+A43)|/4；

In addition this vertical green pixel aberration (GVD) for a vertical green absolute difference of adjacent vertical green pixel (G42, G44) (| G42-G44|) and this pixel of twice deduct adjacent two pixels vertical two with color pixel and the total value (| G42-G44|+|2 * A40-(A42+A44) |/4) of 1/4th (| 2 * A40-(A42+A44) |/4) addition of absolute difference, its formula is:

GVd＝|G42-G44|+|2×A40-(A42+A44)|/4；

Performed pixel compensation work more must include the green offset (GH) of a level, a vertical green offset (GV) and pixel compensation value parameter values such as (GA) to this image process area 2 to this first embodiment; Wherein the green offset of this level (GH) is 1/2nd additive values ((G41+G43)/2) of adjacent level green pixel (G41, G43), the level two that deducts adjacent two pixels with this pixel of twice with color pixel and the total value ((G41-G43)/2+A40-(A41+A43) |/4) of 1/4th (| 2 * A40-(A41+A43) |)/4 additions of absolute difference, its formula is:

GH＝(G41-G43)/2+|2×A40-(A41+A43)|/4；

/ 2nd additive values ((G42+G44)/2) that this vertical green offset (GV) then is adjacent vertical green pixel (G42, G44), with this pixel of twice deduct adjacent two pixels vertical two with color pixel and the total value ((G42-G42)/2+A40-(A42+A44) |/4) of 1/4th (| 2 * A40-(A42+A44) |/4) addition of absolute difference, its formula is:

GV＝(G42-G44)/2+|2×A40-(A42+A44)|/4；

Horizontal green pixel (G41, G43) that this pixel compensation value (GA) then is adjacent this pixel and 1/4th of the total value of vertical green pixel (G42, G44), its formula is: GA=(G41+G42+G43+G44)/4.

Utilize above-mentioned resulting horizontal green pixel aberration and vertical green pixel aberration and the green offset of level, vertical green offset and pixel compensation value, cooperate suitable computing executing rule, include:

A. this vertical green pixel aberration is set critical value smaller or equal to one, and this horizontal green pixel aberration also sets critical value smaller or equal to this, and the green pixel values that this pixel utilizes this image process area to compensate is this pixel compensation value;

B. this vertical green pixel aberration is less than this horizontal green pixel aberration, and the green pixel values of this pixel is the green offset of this level; Otherwise,

C. this horizontal green pixel aberration is less than this vertical green pixel aberration, and the green pixel values of this pixel is this vertical green offset.

We can do the best pixel compensation work of this pixel do of pixel compensation to need to utilize above-mentioned computing executing rule.

Above-mentioned is after this image process area 2 utilizes resulting each pixel parameter to make suitable calculation process in the first embodiment of the invention, this pixel is carried out the explanation of the compensation work of pixel, now please consult Fig. 5 and Fig. 6, in the model schematic diagram that shown in Fig. 5 and Fig. 6 is the of the present invention second and the 3rd embodiment, the difference of the two is that Fig. 5 is the neighbor composition of consideration level, the horizontal pixel part that comprises adjacent two pixels, Fig. 6 is then similar with second embodiment shown in 5 figure, difference is that the 3rd embodiment shown in Figure 6 considers vertical pixel, so in other calculation process part, then all identical with first embodiment, must consider that the pixel that will compensate why, and utilize the adjacent pixels composition to do suitable computing, obtain best offset, this pixel is done best pixel compensation work.

Wherein, second embodiment among Fig. 5 is a part of considering horizontal pixel, performed calculation process pattern among preceding first embodiment that applies mechanically, horizontal green pixel aberration required among second embodiment is (GHd=|C51-C52|+|2 * G50-(G51+G52) |/4), horizontal green pixel offset then is (GH=|C51-C52|/2+|2 * G50-(G51+G52) |/4), and this pixel compensation value then is (HA=(A51+A52+C51+C52)/4); Part as for the consideration of the 3rd embodiment among Fig. 6, vertical pixel, required vertical green pixel aberration is (GVd=|C61-C62|+|2 * G60-(G51+G52) |/4), horizontal green pixel offset then is (GV=|C61-C62|/2+|2 * G60-(G61+G62) |/4), and this pixel compensation value then is (GA=(A61+A62+C61+C62)/4).

Above-mentionedly be required parameter value in the required related pixel compensation work of image process area among the present invention second and the 3rd embodiment 2, yet compensation effect by the pixel compensation operational pattern gained that in this second and second embodiment, is used, because the condition of institute's reference is extensive inadequately, therefore be not technology emphasis of the present invention, but under special status, we still can use this kind compute mode to reach the work purpose of pixel compensation.

For making pixel compensation work of the present invention obtain better treatment effect, the present invention also proposes another embodiment, please consult Fig. 7, Fig. 7 is the model schematic diagram of fourth embodiment of the invention, it is same that we still must have relevant work execution parameter in this 4th embodiment, before obtaining required parameter value, must utilize this image to extract processing and storage that the 1 pair of pixel in district is made video digital signal earlier, comprising a pixel A 70 is arranged, reach the upper right of this pixel (A70), upper left, the lower-left, heterochromatic pixel (the C71 that the bottom right is adjacent, C72, C73, C74) and with this pixel next but two pixel upper right mutually, upper left, the lower-left, same color pixel (the A71 that the bottom right is adjacent, A72, A73, A74).

Utilize above-mentioned each pixel, we can obtain the parameter value of one first pixel compensation aberration (Ddd1) and one second pixel compensation aberration (Ddd2), wherein this first pixel compensation aberration be this upper right heterochromatic pixel absolute difference adjacent with the lower-left (| C71-C73|), deduct the total value (| C71-C73|+|2 * A70-(A71+A73) |/4) of 1/4th (| 2 * A70-(A71+A73) |/4) addition of the absolute difference of the adjacent same color pixel sum in the upper right and lower-left of two pixels of being separated by with this pixel of twice, its formula is:

Ddd1＝|C71-C73|+|2×A70-(A71+A73)|/4；

This second pixel compensation aberration for this upper left heterochromatic pixel absolute difference adjacent with the bottom right (| C72-C74|) and this pixel of twice deduct the total value (| C72-C74|+|2 * A70-(A72+A74) |/4) of 1/4th (| 2 * A70-(A72+A74) |/4) addition of the absolute difference of the upper left same color pixel sum adjacent of two pixels of being separated by with the bottom right, its formula is:

Ddd2＝|C72-C74|+|2×A70-(A72+A74)|/4；

Utilize neighbor in addition, we also can obtain the first pigment offset (Cdd1), the second pigment offset (Cdd2) and pigment offset parameter values such as (CA); Wherein this first pigment offset is these upper right heterochromatic pixel 1/2nd additive values ((C71+C73)/2) adjacent with the lower-left, and this pixel of twice deducts the total value ((C71-C73)/2+A70-(A71+A73) |/4) of 1/4th (| 2 * A70-(A71+A73) |/4) addition of the absolute difference of the adjacent same color pixel sum in the upper right and lower-left of two pixels of being separated by, and its formula is:

Cdd1＝(C71-C73)/2+|2×A70-(A71+A73)|/4；

This second pigment offset is these upper left heterochromatic pixel 1/2nd additive values ((G72+G74)/2) adjacent with the bottom right, deduct the total value ((C72-C72)/2+A70-(A72+A74) |/4) of 1/4th (| 2 * A70-(A72+A74) |/4) addition of the absolute difference of the adjacent same color pixel sum in the upper left and bottom right of two pixels of being separated by with this pixel of twice, its formula is:

Cdd2＝(C72-C74)/2+|2×A70-(A72+A74)|/4；

And this pigment offset be this pixel upper right (C7), upper left (C72), lower-left (C73), the heterochromatic pixel total value in bottom right (C74) 1/4th, its formula is: CA=(C71+C72+C73+C74)/4.

Utilize the above-mentioned resulting first pixel compensation aberration and the second pixel compensation aberration and the first pigment offset, the second pigment offset and pigment offset, cooperate suitable computing executing rule, include:

A. this first pixel compensation aberration is set critical value smaller or equal to one, and this second pixel compensation aberration also sets critical value smaller or equal to this, and this pixel utilizes the pixel compensation value of this image process area compensation to be this pigment offset;

B. this first pixel compensation aberration is less than this second pixel compensation aberration, and the pixel compensation value of this pixel is this first pigment offset; Otherwise,

C. this second pixel compensation aberration is less than this first pixel compensation aberration, and the pixel compensation value of this pixel is this second pigment offset.

Utilize our this pixel that need make pixel compensation in can the 4th embodiment of above-mentioned computing executing rule to do best pixel compensation work to this Fig. 7.

In sum, fill part and demonstrate data memory device of the present invention and data save method all dark rich progressive of implementing on purpose and effect thereof, have the value of industry, and be present new invention not seen before on the market.

The above only is preferred embodiment of the present invention, when can not with the scope implemented of qualification the present invention.Be all equal variation and modifications of being done according to the present patent application claim, all should still belong to the scope that patent of the present invention contains.

Claims

1. the processing method of an image compensation of digital camera, it is characterized in that, utilize an image to extract a plurality of pixels that the district extracts an image, these a plurality of pixels are made of different pigment, utilize this image to extract each the circuit element converting analogue pigment signal strength signal intensity that is provided with in the district and be digital pigment signal strength signal intensity, export an image process area to, utilize this image process area to carry out a pigment compensation method, finish the work of image processing; This pigment compensation method is:

Utilize a plurality of basic pigment composition of measuring a pixel and adjacent a plurality of pixel formations, wherein, the level and vertical homochromy pixel A 41, A43, A42, A44 of described adjacent a plurality of pixels level that is this pixel A 40 and vertical adjacent green pixel G41, G43, G42, G44 and adjacent two pixels with this pixel;

Utilize this pixel and the adjacent a plurality of basic pigment that a plurality of pixel constituted, carry out the computing of pixel compensation; Obtain the Essential colour prime number amount of the required compensation of pixel of desire compensation pigment;

Utilize a digital signal processor that is provided with in this image process area, the pixel of this desire compensation pigment is made the compensation work of basic pigment.

2. the processing method of image compensation of digital camera as claimed in claim 1, it is characterized in that in the work of described Digital Signal Processing execution compensation pigment, utilize the parameter of a horizontal green pixel aberration GHd, a vertical green pixel aberration GVd, the green offset GH of a level, a vertical green offset GV and a pixel compensation value GA to do the parameter computing, wherein utilize following rule this pixel to be done the compensation of pixel:

Set critical value as this vertical green pixel aberration smaller or equal to one, and this horizontal green pixel aberration is also set critical value smaller or equal to this, it is this pixel compensation value that this pixel is utilized the green pixel values of this image process area compensation, wherein said horizontal green pixel aberration GHd is the green absolute difference of a level of adjacent horizontal green pixel, and the level two that deducts adjacent two pixels with this pixel of twice with color pixel and the total value of 1/4th additions of absolute difference; Its formula is:

GHd＝|G41-G43|+|2×A40-(A41+A43)|/4；

Described this vertical green pixel aberration GVd is a vertical green absolute difference of adjacent vertical green pixel, and with this pixel of twice deduct adjacent two pixels vertical two with color pixel and the total value of 1/4th additions of absolute difference, its formula is:

GVd＝|G42-G44|+|2×A40-(A42+A44)|/4；

The green offset GH of described level is 1/2nd additive values of adjacent level green pixel, the level two that deducts adjacent two pixels with this pixel of twice with color pixel and the total value of 1/4th additions of absolute difference, its formula is:

GH＝(G41+G43)/2+|2×A40-(A41+A43)|/4；

/ 2nd additive values that described this vertical green offset GV then is adjacent vertical green pixel, with this pixel of twice deduct adjacent two pixels vertical two with color pixel and the total value of 1/4th additions of absolute difference, its formula is:

GV＝(G42+G44)/2+|2×A40-(A42+A44)|/4；

Horizontal green pixel that described pixel compensation value GA then is adjacent this pixel and 1/4th of the total value of vertical green pixel, its formula is :=(G41+G42+G43+G44)/4.

3. the processing method of image compensation of digital camera as claimed in claim 2 is characterized in that also including in the regular calculation process that described digital signal processor carries out:

Less than this horizontal green pixel aberration, the green pixel values of this pixel is the green offset of this level as this vertical green pixel aberration.

4. the processing method of image compensation of digital camera as claimed in claim 2 is characterized in that also including in the regular calculation process that described digital signal processor carries out:

Less than this vertical green pixel aberration, the green pixel values of this pixel is this vertical green offset as this horizontal green pixel aberration.

5. the processing method of image compensation of digital camera as claimed in claim 1, it is characterized in that described adjacent a plurality of pixels for upper right, upper left, lower-left, the bottom right of this pixel A 70 adjacent heterochromatic pixel C71, C72, C73, C74 and with this pixel upper right, upper left, lower-left, adjacent homochromy pixel A 71, A72, A73, the A74 in bottom right of next but two pixel mutually.

6. the processing method of image compensation of digital camera as claimed in claim 1, it is characterized in that in the work of described digital signal processor execution compensation pigment, utilize one first similarly to be that the parameter of compensate for chromatic aberration Ddd1, one second pixel compensation aberration Ddd2, one first pigment offset Cdd1, one second pigment offset Cdd2 and a pigment offset CA is done the parameter computing, wherein utilize following rule this pixel to be done the compensation of pixel:

Set critical value as this first pixel compensation look, difference smaller or equal to one, and this second pixel compensation aberration is also set critical value smaller or equal to this, the wherein said first pixel compensation aberration Ddd1 is this upper right heterochromatic pixel absolute difference adjacent with the lower-left, deduct the total value of 1/4th additions of the absolute difference of the adjacent same color pixel sum in the upper right and lower-left of two pixels of being separated by with this pixel of twice, its formula is:

Ddd1＝|C71-C73|+|2×A70-(A71+A73)|/4；

The described second pixel compensation aberration Ddd2 is this upper left heterochromatic pixel absolute difference adjacent with the bottom right, deduct the total value of 1/4th additions of the absolute difference of the upper left same color pixel sum adjacent with the bottom right of two pixels of being separated by with this pixel of twice, its formula is:

Ddd2＝|C72-C74|+|2×A70-(A72+A74)|/4；

The described first pigment offset Cdd1 is these upper right heterochromatic pixel 1/2nd additive values adjacent with the lower-left, deduct the total value of 1/4th additions of the absolute difference of the adjacent same color pixel sum in the upper right and lower-left of two pixels of being separated by with this pixel of twice, its formula is:

Cdd1＝(C71+C73)/2+|2×A70-(A71+A73)|/4；

The described second pigment offset Cdd2 is these upper left heterochromatic pixel 1/2nd additive values adjacent with the bottom right, deduct the total value of 1/4th additions of the absolute difference of the adjacent same color pixel sum in the upper left and bottom right of two pixels of being separated by with this pixel of twice, its formula is:

Cdd2＝(C72+C74)/2+|2×A70-(A72+A74)|/4；

Described pigment offset CA be this pixel upper right, upper left, lower-left, the heterochromatic pixel total value in bottom right 1/4th, its formula is: CA=(C71+C72+C73+C74)/4;

This pixel utilizes the pixel compensation value of this image process area compensation to be this pigment offset;

7. the processing method of image compensation of digital camera as claimed in claim 6 is characterized in that also including in the regular calculation process that described digital signal processor carries out:

Less than this second pixel compensation aberration, the pixel compensation value of this pixel is this first pigment offset as this first pixel compensation aberration.

8. the processing method of image compensation of digital camera as claimed in claim 6 is characterized in that also including in the regular calculation process that described digital signal processor carries out:

This second pixel compensation aberration is less than this first pixel compensation aberration, and the pixel compensation value of this pixel is this second pigment offset.

9. the processing method of an image compensation of digital camera, utilizing an image that is provided with in two digital cameras to extract the district does to extract and handles an image with the circuit unit of an image process area, it is characterized in that, utilize the application of a color filtering array to detect the pigment constituent of pixel A 40 respectively; Extract the basic comprising pigment composition that measures this pixel and adjacent a plurality of pixels, a digital signal processor that utilizes this image processing to get setting in the district obtains the parameter of a horizontal green pixel aberration CHd, a vertical green pixel aberration GVd, the green offset GH of a level, a vertical green offset GV and a pixel compensation value GA as calculation process, and utilizes following rule that this pixel is made pixel compensation;

Set critical value as this vertical green pixel aberration smaller or equal to one, and this horizontal green pixel aberration also sets critical value smaller or equal to this, the green pixel values that this pixel utilizes this image process area to compensate is this pixel compensation value;

Less than this horizontal green pixel aberration, the green pixel values of this pixel is the green offset of this level as this vertical green pixel aberration; Or

Less than this vertical green pixel aberration, the green pixel values of this pixel is this vertical green offset as this horizontal green pixel aberration;

Wherein said horizontal green pixel aberration GHd is the green absolute difference of a level of adjacent horizontal green pixel, and the level two that deducts adjacent two pixels with this pixel of twice with color pixel and the total value of 1/4th additions of absolute difference, its formula is:

GHd＝|G41-G43|+|2×A40-(A41+A43)|/4；

Described vertical green pixel aberration GVd be adjacent vertical green pixel a vertical green absolute difference and with this pixel of twice deduct adjacent two pixels vertical two with color pixel and the total value of 1/4th additions of absolute difference, its formula is:

GVd＝|G42-G44|+|2×A40-(A42+A44)|/4；

GH＝(G41+G43)/2+|2×A40-(A41+A43)|/4；

/ 2nd additive values that described vertical green offset GV is adjacent vertical green pixel, with this pixel of twice deduct adjacent two pixels vertical two with color pixel and the total value of 1/4th additions of absolute difference, its formula is:

GV＝(G42+G44)/2+|2×A40-(A42+A44)|/4；

Horizontal green pixel that described pixel compensation value GA is adjacent this pixel and 1/4th of the total value of vertical green pixel, its formula is: GA=(G41+G42+G43+G44)/4.

10. the processing method of image compensation of digital camera as claimed in claim 9, it is characterized in that described adjacent a plurality of pixels are, the level and vertical homochromy pixel A 41, A43, A42, A44 of the level of this pixel A 40 and vertical adjacent pixels G41, G43, G42, G44 and adjacent two pixels with this pixel.

11. the processing method of an image compensation of digital camera, be to utilize an image that is provided with in the digital camera to extract the district one image to be done to extract and handled, wherein utilize the application of a color filtering array to detect the pigment constituent of this pixel respectively with the circuit unit of an image process area; Extract the basic comprising pigment composition that measures this pixel and adjacent a plurality of pixels, utilize this image processing to get a digital signal processor that is provided with in the district and obtain the parameter of one first pixel compensation aberration Ddd1, one second pixel compensation aberration Ddd2, one first pigment offset Cdd1, one second pigment offset Cdd2 and a pigment offset CA, and utilize following rule that this pixel is made pixel compensation as calculation process:

If this first pixel compensation aberration is set critical value smaller or equal to one, and this second pixel compensation aberration also sets critical value smaller or equal to this, and this pixel utilizes the pixel compensation value of this image process area compensation to be this pigment offset;

If this first pixel compensation aberration is less than this second pixel compensation aberration, the pixel compensation value of this pixel is this first pigment offset; Or

This second pixel compensation aberration is less than this first pixel compensation aberration, and the pixel compensation value of this pixel is this second pigment offset;

The wherein said first pixel compensation aberration Ddd1 is this upper right heterochromatic pixel absolute difference adjacent with the lower-left, deduct the total value of 1/4th additions of the absolute difference of the adjacent same color pixel sum in the upper right and lower-left of two pixels of being separated by with this pixel of twice, its formula is:

Ddd1＝|C71-C73|+|2×A70-(A71+A73)|/4；

Ddd2＝|C72-C74|+|2×A70-(A72+A74)|/4；

Cdd1＝(C71+C73)/2+|2×A70-(A71+A73)|/4；

Cdd2＝(C72+C74)/2+|2×A70-(A72+A74)|/4；

Described pigment offset CA be this pixel upper right, upper left, lower-left, the heterochromatic pixel total value in bottom right 1/4th, its formula is: CA=(C71+C72+C73+C74)/4.

12. the processing method of image compensation of digital camera as claimed in claim 11, it is characterized in that described adjacent a plurality of pixels for upper right, upper left, lower-left, the bottom right of this pixel A 70 adjacent heterochromatic pixel C71, C72, C73, C74 and with this pixel upper right, upper left, lower-left, adjacent homochromy pixel A 71, A72, A73, the A74 in bottom right of next but two pixel mutually.