CN102420997A

CN102420997A - Image processing method

Info

Publication number: CN102420997A
Application number: CN2010102981969A
Authority: CN
Inventors: 孙维廷; 黄铃琇
Original assignee: Himax Media Solutions Inc
Current assignee: Himax Media Solutions Inc
Priority date: 2010-09-28
Filing date: 2010-09-28
Publication date: 2012-04-18
Anticipated expiration: 2030-09-28
Also published as: CN102420997B

Abstract

The invention discloses an image processing method for improving a retinal rivalry phenomenon of a stereoscopic image. The image processing method comprises the following steps of: receiving a first image and a second image corresponding to the first image, wherein in the first image and the second image, each pixel contains a plurality of color brightness corresponding to different colors respectively; selecting the color brightness corresponding to the one color from the first image, and determining a plurality of proportions of the color brightness corresponding to the different colors in the first image according to the selected color brightness; multiplying the proportions by the color brightness corresponding to the different colors in the first image to obtain a first modulation image, and multiplying the color brightness in the second image by a plurality of preset proportions to obtain a second modulation image; and combining the first modulation image with the second modulation image to generate the stereoscopic image.

Description

Image processing method

Technical field

The present invention relates to a kind of image processing method, and be particularly related to a kind of image processing method of stereo-picture.

Background technology

Stereo-picture generally is made up of two picture group pictures.When using specific device to watch image, stereo-picture can produce the vision of the degree of depth.Complementary colours video picture mode is a kind of video picture mode of stereo-picture, and in the middle of this mode, different main colors can be used for appearing first image and second image.Usually, the spectrum of first image and second image can be not overlapped, therefore needs to use the filter of two kinds of complementary hues to view and admire first image and second image.For instance, when the first filter F1 was used to view and admire first image, the second filter F2 then was used to view and admire second image.The first filter F1 mainly transmits the main color of first image and stops the main color of second image, and the second filter F2 then mainly transmits the main color of second image and stops the main color of first image.

Complementary colours video picture mode often adopts three kinds of colors to present image, and wherein first image adopts two kinds of main colors usually, and second image then adopts another kind of main color.In the middle of the video picture mode of red/blue complementary colours (Red/Cyan), first image mainly appears green and blue, and second image then mainly presents redness.The complementary colours of other types possibly comprise indigo plant/Huang, and is green/the fuchsin complementary colours.These complementary colours are called as three look complementary colours.

Because cost is low and preferable compatibility, uses three look complementary colours to show stereo-picture usually.Yet three traditional look complementary colours have some well-known shortcomings.At first, if use color filter to view and admire image, the colour gamut of three traditional look complementary colours generally can be reduced; Secondly, traditional complementary colours method can produce the phenomenon of retina competition (Retinal rivalry), for example, when the red composition of image object more for a long time, can feel that through red blue glasses that object in flicker, causes the user to do not feel like oneself.

Therefore need a kind of new image processing method, can reduce the phenomenon of retina competition, let the user can be comfortable and cosily view and admire stereo-picture.

Summary of the invention

Therefore, one side of the present invention provides a kind of image processing method, can be under the prerequisite that keeps the image original color, and the composition of a certain color of adjustment image reduces the phenomenon that retina is competed.

According to one embodiment of the invention; Image processing method improves the retina that a stereo-picture caused and competes anti-phenomenon; This image processing method receives one second image of one first image and corresponding first image; Wherein in first image and second image, each pixel contains a plurality of chroma-luminancees that correspond respectively to different color.Afterwards, in first image, choose chroma-luminance corresponding to a kind of color; And according to selected chroma-luminance; Determine in first image a plurality of ratios, again these ratios are multiplied by in first image chroma-luminance corresponding to different color, to draw one first modulation image corresponding to the chroma-luminance of different color; Then chroma-luminance in second image is multiplied by a plurality of preset ratio; Drawing one second modulation image, and make up first modulation image and second modulation image, produce stereo-picture.

The image processing method of the foregoing description; Ratio according to number or a certain color and other colour components of a certain colour content in the image; Adjust the ratio of the central specific color of image, use the phenomenon that reduces the retina competition, still keep most colour component in the original image simultaneously.

Description of drawings

For letting above-mentioned and other purposes of the present invention, characteristic, advantage and the embodiment can be more obviously understandable, the explanation of accompanying drawing be following:

Fig. 1 is the flow chart that illustrates an embodiment of the present invention image processing method.

Fig. 2 is the yield value that illustrates the red brightness of one embodiment of the invention.

Fig. 3 is the yield value that illustrates the red brightness of another embodiment of the present invention.

[main element symbol description]

101～111: step

Embodiment

In the middle of the image processing method of following examples, a certain colour component in the middle of the rejection image (for example red) fully not, but certain the specific color composition in the image is done a little reservation, avoid excessively changing the color of original image; Simultaneously can adjust image quality, use the elimination retina and compete anti-phenomenon (Retinal rivalry) according to number or its proportion of specific color composition.

Please with reference to Fig. 1, it is the flow chart that illustrates an embodiment of the present invention image processing method.Image processing method improves the retina of a stereo-picture and competes anti-phenomenon; This image processing method at first receives second image (step 101) of first image and corresponding this first image; Wherein in the middle of first image and second image; Each pixel contains several chroma-luminancees corresponding to different color, for example red, green and blue brightness, and first image and second image have a parallax between the two and concern (Parallax).By between first image and second image, existing parallax, after first image and second image were received by user's left eye and right eye respectively, human brain can synthesize first image and second image that right eye and left eye receive, and produces stereo-picture.

Then in the middle of first image, choose the chroma-luminance (step 103) corresponding to a kind of color, for example red brightness, and according to selected chroma-luminance determines in first image several ratios (step 105) corresponding to the chroma-luminance of different color.After step 105, aforementioned ratio is multiplied by in first image corresponding to the chroma-luminance of different color, draw first modulation image (step 107); And chroma-luminance in second image is multiplied by several preset ratio, and drawing second modulation image (step 109), this preset ratio can be fixing constant; For example (0; 0,0), (0,1,0), and (0,0,1); Make up first modulation image and second modulation image at last, produce stereo-picture (step 111).

More detailed, the relation of chroma-luminance in the chroma-luminance of first image, second image and stereo-picture (for example red/blue images) is:

I_{o} = \begin{matrix} [\begin{matrix} R_{R} & R_{G} & R_{B} \\ 0 & 0 & 0 \\ 0 & 0 & 0 \end{matrix}] \end{matrix} [\begin{matrix} R_{L} \\ G_{L} \\ B_{L} \end{matrix}] + \begin{matrix} [\begin{matrix} 0 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{matrix}] [\begin{matrix} R_{R} \\ G_{R} \\ B_{R} \end{matrix}], \end{matrix}

Wherein, " I _o" stereo-picture of representative output, the matrix before "+" number is relevant with first modulation image, and "+" matrix number afterwards is relevant with second modulation image.R _L, G _L, B _LRepresent the red brightness in first image, green brightness and blue brightness respectively, R _R, R _G, R _BThen be respectively corresponding redness, green and blue brightness R in first image _L, G _L, B _LRed ratio, green ratio, with blue ratio, in order to producing the red brightness of first modulation image, and the green of first modulation image is (0,0,0) with blue brightness corresponding ratio in this example, so green and blue brightness are 0; R _R, G _R, B _RRepresent the red brightness in second image, green brightness respectively, and blue brightness, (0,0; 0), (0,1,0), (0,0; 1) then correspond to the preset ratio of second image, in order to green and the blue brightness that produces second modulation image, and the red brightness of second modulation image corresponding ratio in this example is (0; 0,0), so red brightness is 0.In this example, the red brightness of stereo-picture is the red brightness of first modulation image, and green and blue brightness then is the green and blue brightness of second modulation image, and first image can be left-eye image, and second image then can be eye image.

Please with reference to Fig. 2, it is the yield value that illustrates the red brightness of one embodiment of the invention, and wherein the numerical values recited of red brightness is 0 to 255.In the middle of this embodiment, according to red brightness R in first image _LSize decide redness, green and blue ratio R _R, R _G, R _B, to produce the red brightness of first modulation image.More detailed, hot colour brightness R _LWhen being less than or equal to the first critical value Th1 (for example 30), red ratio R _REqual a luminance gain higher limit G1 (for example 0.9), hot colour brightness R _LDuring more than or equal to the second critical value Th2 (for example 225), red ratio R _REqual luminance gain lower limit G2 (for example 0.3).

Hot colour brightness R _LIn the time of between the first critical value Th1 and the second critical value Th2; Then according to the first critical value Th1, the second critical value Th2, luminance gain higher limit G1 and luminance gain lower limit G2; Carry out linear interpolation and calculate (straight line of slope non-zero in the middle of Fig. 2 just), to obtain red ratio R _R, wherein linear interpolation account form is:

Try to achieve red ratio R _RAfterwards, try to achieve other color ratios according to following equality again:

\{\begin{matrix} R_{R} + R_{G} + R_{B} = 1 \\ R_{G} = (1 - R_{R}) \times W_{t} \end{matrix},

0≤W _t≤1, W wherein _tRepresent a predefined weight.

Please with reference to Fig. 3, it is the yield value that illustrates the red brightness of another embodiment of the present invention, and wherein the numerical values recited of red brightness is 0 to 255.In the middle of this embodiment, according to the red brightness R of first image _L, green brightness G _L, with blue brightness B _LRelation decide redness, green and blue ratio R _R, R _G, R _BEven the brightness of various colors is different, however the luminance difference D between the different color _IffStill maybe be identical.For instance, if red brightness R _LBe respectively 200 and 60, the greater then is respectively 150 and 10 in the middle of green brightness and the blue brightness, then the luminance difference D of redness and another color _IffBe 50 (200-150; 60-10), cause to pick out different chroma-luminancees.In this case, need according to red brightness R _LNumerical value adjust the first critical value Th1, the second critical value Th2, also be about to this two critical value and be multiplied by a parameter preset Sat (for example 0.9), in the hope of first parameter alpha and second parameter beta, try to achieve red ratio R according to this two parameter alpha, β again _R

Further, as luminance difference D _IffBe less than or equal to first parameter alpha, red ratio R _REqual luminance gain higher limit G1, as luminance difference D _IffDuring more than or equal to second parameter beta, red ratio R _REqual luminance gain lower limit G2; When luminance difference during between first parameter alpha and second parameter beta; Then carry out a linear interpolation and calculate (being the straight line of slope non-zero among Fig. 3), to obtain red ratio R according to first parameter alpha, second parameter beta, luminance gain higher limit G1 and luminance gain lower limit G2 _R, and luminance difference D _IffCan be: green brightness G _LOr blue brightness B _LThe greater among both is with red brightness R _LGap.The linear interpolation account form equals:

The image processing method of above embodiment can not got rid of the central specific color composition of original image (for example red) fully, but the specific color composition in the image is done a little reservation, avoids excessively changing the color of original image; Simultaneously adjust image, use the elimination retina and compete anti-phenomenon according to number or its proportion of specific color composition.

Though the present invention with execution mode openly as above; Right its is not that any those skilled in the art are not breaking away from the spirit and scope of the present invention in order to qualification the present invention; When can doing various changes and retouching, so protection scope of the present invention is as the criterion when looking the appended claims person of defining.

Claims

1. an image processing method is competed anti-phenomenon with the retina that improves a stereo-picture, and this image processing method comprises:

Receive one first image and to one second image that should first image, wherein in this first image and this second image, each pixel comprises a plurality of chroma-luminancees corresponding to different color;

In first image, choose chroma-luminance corresponding to a kind of color;

According to selected chroma-luminance, determine in this first image a plurality of ratios corresponding to these chroma-luminancees of different color;

These ratios are multiplied by in this first image these chroma-luminancees corresponding to different color, to draw one first modulation image;

These chroma-luminancees in this second image are multiplied by a plurality of preset ratio, to draw one second modulation image; And

Make up this first modulation image and this second modulation image, to produce this stereo-picture.

2. image processing method as claimed in claim 1, wherein these chroma-luminancees in these ratios, this first image, these chroma-luminancees in this second image and the relation of this stereo-picture are:

I_{o} = [\begin{matrix} R_{R} & R_{G} & R_{B} \\ 0 & 0 & 0 \\ 0 & 0 & 0 \end{matrix}] [\begin{matrix} R_{L} \\ G_{L} \\ B_{L} \end{matrix}] + [\begin{matrix} 0 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{matrix}] [\begin{matrix} R_{R} \\ G_{R} \\ B_{R} \end{matrix}],

Wherein, I _oRepresent this stereo-picture, R _L, G _L, B _LRepresent a red brightness, a green brightness and a blue brightness in this first image respectively, R _R, G _R, B _RRepresent a red brightness, a green brightness in this second image respectively, and a blue brightness, these ratios R _R, R _G, R _BBe respectively corresponding to should redness brightness in this first image, this green brightness and this blueness brightness R _L, G _L, B _LA red ratio, a green ratio and a blue ratio.

3. image processing method as claimed in claim 2, wherein accordinging to should redness brightness R in this first image _LSize decide this redness ratio R _R, this green ratio R _GWith this blueness ratio R _B

4. image processing method as claimed in claim 3 is wherein as this redness brightness R _LWhen being less than or equal to one first critical value Th1, this redness ratio R _REqual a luminance gain higher limit G1, as this redness brightness R _LDuring more than or equal to this second critical value Th2, this redness ratio R _REqual this luminance gain lower limit G2, as this redness brightness R _LIn the time of between this first critical value Th1 and this second critical value Th2, then carry out a linear interpolation and calculate, to obtain this redness ratio R according to this first critical value Th1, this second critical value Th2, this luminance gain higher limit G1 and this luminance gain lower limit G2 _R

5. image processing method as claimed in claim 4, wherein this linear interpolation account form is:

6. image processing method as claimed in claim 2, wherein these color ratios need satisfy following equality:

\{\begin{matrix} R_{R} + R_{G} + R_{B} = 1 \\ R_{G} = (1 - R_{R}) \times W_{t} \end{matrix},

0≤W _t≤1

Wherein, W _tRepresent a predefined weight.

7. image processing method as claimed in claim 2 is wherein according to this redness brightness R _L, this green brightness G _L, with this blueness brightness B _LLuminance difference relation decide this redness, green and blue ratio R _R, R _G, R _B

8. image processing method as claimed in claim 7 is wherein as this luminance difference D _IffBe less than or equal to one first parameter alpha, this redness ratio R _REqual luminance gain higher limit G1, as this luminance difference D _IffDuring more than or equal to one second parameter beta, this redness ratio R _REqual this luminance gain lower limit G2; When this luminance difference during between this first parameter alpha and this second parameter beta; Then carry out a linear interpolation and calculate, to obtain this redness ratio R according to this first parameter alpha, this second parameter beta, this luminance gain higher limit G1 and this luminance gain lower limit G2 _R, and this luminance difference D _IffFor:

This redness brightness R _L, with

This green brightness G _LOr should blueness brightness B _LThe gap of the greater among both.

9. image processing method as claimed in claim 8, wherein this linear interpolation account form equals:

10. image processing method as claimed in claim 1, wherein this first image and this second image have parallax relation.