CN102339461A - Method and equipment for enhancing image - Google Patents

Abstract

The invention discloses a method and equipment for enhancing an image. The method disclosed by the embodiment of the invention comprises the following steps of: calculating local luminance values and local contrast values of each pixel in the input digital image; determining gain factors of each pixel on the basis of the local luminance values and the local contrast values; and regulating corresponding pixel values by using the determined gain factors. According to the method and the equipment disclosed by the embodiment of the invention, the luminance and the contrast of the image can be simultaneously enhanced, and an artifact cannot be introduced. According to other embodiments, the calculation efficiency is improved, and the method and the equipment can be applied to various image processing systems such as images/video capture equipment, display equipment, printing equipment and the like.

Description

Image enchancing method and equipment

Technical field

The present invention relates to Digital Image Processing, be specifically related to a kind of image strengthened and/or dynamic range is compressed, its image based on input produces clearer and bright image, and does not have visible pseudomorphism.

Background technology

Human eyes are the scene in the snow scenes under the perception sunshine or the dim grotto easily.The human visual system can adapt to various environment, and has no the scene in the problem ground perception respective environment.But artificial various imageing sensors but do not possess such ability, and they can not the various scenes of flexible adaptation, only can catch the fixing scene of dynamic range, and it is very limited that this dynamic range is actually.For example, stand in furniture, carpet that indoor people can discern indoor corner, and can see the white clouds under the sunshine clearly simultaneously.But,, will meet difficulty if people want to take the scene that he saw with video camera.If the length that he will be provided with the time shutter is caught the scene details in the dim space, then the sky outside the window can't be taken.If he is set to short details of taking sky the time shutter, he can not see indoor any details.This all is the problem that runs in the daily life.Camera or video recorder are captured is not to be that our eyes are seen.Therefore, how the bigger dynamic range of image being compressed, is the problem that needs solution in the Digital Image Processing but do not bring pseudomorphism simultaneously.

Patent documentation 1 (US20060110052A1) discloses a kind of image processing techniques, it at first with image transformation to in the number space, eliminate colourity through subtracting each other between the different color channels then, use local mean value and variance to come basis of calculation coordinate then.This local mean value is through image and the average filter convolution such as Gaussian filter are obtained.The calculating of this local variance is carried out according to the definition of traditional standard variance, but has carried out ranking operation.Standard coordinate obtains through from average, deducting pixel value and removing local variance.Give two outer two kinds of methods of calculating the local standard coordinate in the documents 1.This method provides several method to come once more standard coordinate and original image are comprehensively obtained the enhancing image exported at last.A kind of method is that three passages with identical factor pair original pixels carry out convergent-divergent, and this factor is the ratio of standard coordinate and original brightness.Another kind method is the inverse operation of above-mentioned first step.But, do not provide the details that how to obtain desirable output image in the documents 1.

In above-mentioned patent documentation 1, need in big relatively neighborhood, calculate local mean value and local variance.This has caused the increase of computation burden.In addition, in patent documentation 1, can't guarantee not introduce pseudo-colours.In other words, the technology of patent documentation 1 can cause visible pseudomorphism.

Summary of the invention

The objective of the invention is to propose a kind of image enchancing method and equipment,, and do not have visible pseudomorphism based on the clearer and bright image of the image generation of input.

In one aspect of the invention, a kind of method that digital picture is strengthened is provided, has comprised step: the local luminance values and the local contrast value that calculate each pixel in the input digital image; Confirm the gain factor of each pixel based on said local luminance values and local contrast value; And use the gain factor of confirming to regulate the value of respective pixel.

In another aspect of this invention, a kind of equipment that digital picture is strengthened is provided, has comprised: brightness calculation unit, the local luminance values of each pixel in the calculating input digital image; The contrast computing unit, the local contrast value of each pixel in the calculating input digital image; The gain factor of each pixel is confirmed in the gain calculating unit based on said local luminance values and local contrast value; And map unit, use the gain factor of confirming to regulate the value of respective pixel.

Utilize such scheme, in time at the very large image of dynamic range, also can be under the situation of not losing details, realize the two the enhancing of brightness of image and contrast.

Description of drawings

From the detailed description below in conjunction with accompanying drawing, above-mentioned feature and advantage of the present invention will be more obvious, wherein:

Fig. 1 is the structured flowchart according to the digital picture enhancement apparatus of the embodiment of the invention;

Fig. 2 is the process flow diagram of describing according to the digital image enhancement method of the embodiment of the invention;

Fig. 3 is the synoptic diagram of description according to the calculating brightness and contrast's of the embodiment of the invention process;

Fig. 4 shows an application examples according to the equipment of the embodiment of the invention;

Fig. 5 shows an application examples according to the method and apparatus of the embodiment of the invention;

Fig. 6 shows an application examples according to the method and apparatus of the embodiment of the invention;

Fig. 7 shows an application examples according to the method and apparatus of the embodiment of the invention.

Embodiment

Below, specify preferred implementation of the present invention with reference to accompanying drawing.In the accompanying drawings, though be shown in the different drawings, identical Reference numeral is used to represent identical or similar assembly.For clear and simple and clear, the known function and the detailed description of structure that are included in here will be omitted, otherwise they will make theme of the present invention unclear.

Embodiments of the invention have proposed a kind of method and apparatus that digital picture is strengthened.In an embodiment of the present invention, consider that the contrast between the color of various piece of ambient brightness and subject in the scene of image shoot process comes captured image is strengthened.According to embodiments of the invention, can the Digital Image Processing that dynamic range is big be paired in the image in the responsive dynamic range of human eye, and the brightness and contrast who in compression of dynamic range, has improved image.

Fig. 1 shows the structured flowchart according to the equipment that digital picture is strengthened of the embodiment of the invention.The equipment 100 of the embodiment of the invention can be used for video camera; Image to taking is handled in real time; Also can from the storer 110 of having stored prior shot digital images, read digital picture; Handle according to method of the present invention, be stored in the storer then, or through 140 outputs of output devices such as printer or display.

As shown in Figure 1, according to the image-enhancing equipment 100 of the embodiment of the invention possess to color digital image carry out gray processing gray processing unit 101, to each pixel in the image calculate brightness value brightness calculation unit 105, calculate the contrast computing unit 102 of contrast, confirm the map unit 104 that gain factor that the gain calculating unit 103 and utilizing of the gain factor of each pixel calculates is handled each pixel based on the brightness and contrast who is calculated to each pixel in the image.Image through after map unit 104 processing is stored in the storer 110, or through 140 outputs of the output device such as printer or display.

Specify the concrete function and the operating process of each component units below with reference to accompanying drawing 2 and 3.Fig. 2 is the process flow diagram of describing according to the digital image enhancement method of the embodiment of the invention.

At step S11, S12 and S13, input digital image data from storer 110.According to embodiments of the invention, the color digital image that reads from storer 110 is a RGB triple channel image, and gray processing unit 101 converts thereof into gray level image.According to embodiments of the invention; Gray processing the unit 101 or pixel value of three passages averaged obtain gray-scale value; Perhaps select the pixel value of one of them passage to obtain gray-scale value, or the pixel value of three passages is carried out weighted sum obtain gray-scale value.For example, the gray-scale value of each pixel is confirmed in gray processing unit 101 according to following formula:

grey(x)＝(r(x)+g(x)+b(x))/3 ……(1)

Or

grey(x)＝max(r(x)，g(x)，b(x)) ……(2)

In above-mentioned formula (1) and (2), the position of variable x remarked pixel in input picture.It can be bivector in view of x, so can be used for the position of remarked pixel in input picture under two-dimensional coordinate.R (x), g (x) and b (x) represent the RGB component of colour element respectively.Maximal value is got in max () expression.

According to embodiments of the invention,, then need not to carry out again gray processing and handle if the image of input itself is exactly the single channel image.

Then, in step S14 and S15, calculate the brightness value and the contrast value of each pixel in brightness calculation unit 105 and the contrast computing unit 102.According to embodiments of the invention, calculate the brightness and contrast through following formula (3) and (4):

$\mathrm{lu}\min \mathrm{ance}\left(x\right)=\underset{i}{\overset{N}{\mathrm{\Σ}}}{W}_i*{\mathrm{mean}}_i\left(x\right)\·\·\·\·\·\·\left(3\right)$

$\mathrm{contrast}\left(x\right)=\underset{i}{\overset{N}{\mathrm{\Σ}}}{W}_i*\log 10\left(\frac{G\left(x\right)}{{\mathrm{mean}}_i\left(x\right)}\right)\·\·\·\·\·\·\left(4\right)$

In the above-mentioned formula, the position of x remarked pixel in input picture, the gray-scale value of G (x) statement pixel, the grey (x) in promptly above-mentioned formula (1) or (2), mean _i(x) be illustrated in i the local mean value under the yardstick, W _iRepresent i the weighting coefficient under the yardstick.According to embodiments of the invention, as shown in Figure 3 in order to calculate above-mentioned parameter effectively, at first through following formula (5) calculated product partial image:

$\mathrm{integral}\left(x\right)=\underset{\mathrm{and\; above}}{\underset{\mathrm{to\; the\; lefe}}{\underset{\mathrm{all\; pixels}}{\mathrm{\Σ}}}}G\left(x\right)......\left(5\right)$

Just the gray-scale value of all pixels in the upper left corner is sued for peace.Then, add/subtract computing through four times and calculate mean _i(x).At the local mean value mean that obtains under each yardstick _i(x) after, calculate the brightness and contrast according to above-mentioned formula (3) and (4) again.

After obtaining the brightness and contrast, increase at step S16, gain calculating unit 103 is according to the gain factor of each pixel in brightness and contrast's computed image of calculating.According to embodiments of the invention, gain calculating unit 103 calculates the gain factor of each pixel according to following formula (6):

$\mathrm{gain}\left(x\right)=\frac{\mathrm{upper}\_\mathrm{bound}\left(x\right)}{1-a*\exp (-\mathrm{strength}*(\mathrm{contrast}\left(x\right)-\mathrm{cross}\_\mathrm{point}\left(x\right)))}\·\·\·\·\·\·\left(6\right)$

A=upper_bound (x)-1 in following formula (6) makes gain (x)=1 when contrast (x)=cross_point (x).Can upper_bound (x) be set to a fixed value to entire image, perhaps be adaptive to the value of each pixel.According to embodiments of the invention, can be provided with according to following formula:

upper_bound(x)＝1/mean(x)……(7)

Or

upper_bound(x)＝1/(max(r(x)，g(x)，b(x))……(8)

Mean (x) in the above-mentioned formula (7) can be the average under any yardstick, the perhaps weighted sum of average under each yardstick.Similarly, can a fixed value be set for cross_point (x), perhaps be adaptive to the value of each pixel to entire image.According to embodiments of the invention, can be provided with according to following formula:

Cross_point (x)=1.0 ... (9) or

cross_point(x)＝1-luminance(x)……(10)

Parameter s trength in the formula (6) leaves the user for and carries out the interactive mode adjusting.The user can come the intensity of control chart image intensifying through regulating this parameter.

After the gain that has calculated each pixel, at step S17, map unit 104 is carried out enhancement process to each pixel.According to embodiments of the invention, map unit 104 is on duty with identical gain factor with each passage of each pixel, makes the tone of input picture be able to keep.In map unit 104, carry out such operation.For example map unit 104 is carried out map operation through following formula (11):

(r′，g′，b′)＝gain(x)·(r，g，b)……(11)

At step S18, the output device 140 such as printer or display will pass through the image print output of enhancement process or be presented on the display screen.

Though describe the specific embodiment of equipment of the present invention above with the functional block of separating, those of ordinary skill in the art should be able to expect above-mentioned unit is formed in the concrete device, for example among the CPU, or in a plurality of concrete parts.For example the equipment of the embodiment of the invention can be implemented in the video camera, and in the mobile phone, PC perhaps in the Medical Devices, or is realized with software, or realized with hardware.

Fig. 4 shows an application according to the equipment of the embodiment of the invention.It can be implemented in digital image capture and the display system.After ccd video camera has been caught digital picture, utilize equipment of the present invention that the digital picture of catching is carried out enhancement process, the image after obtaining to strengthen is presented on the display.

As shown in Figure 5, the method and apparatus of the embodiment of the invention also can be applied in the enhancing of video image.To the method for each frame application embodiment of the invention in the video, then in the digital picture that shows on the display device after strengthening.

As shown in Figure 6, the method and apparatus of the embodiment of the invention also can be applied in the print system, before image is printed, at first utilizes the method for embodiments of the invention this image to be carried out enhancement process, printout in printer then.

As shown in Figure 7, the method and apparatus of the process embodiment of the invention can be stored in the storer after strengthening, so that follow-up processing.

Method and apparatus according to an embodiment of the invention, the brightness and contrast that can strengthen image simultaneously, and can not introduce pseudomorphism.According to additional embodiments, submitted counting yield to, and can be applied in the various image processing systems, image/video capture equipment for example, display device, PRN device etc.

Top description only is used to realize embodiment of the present invention; It should be appreciated by those skilled in the art; In any modification that does not depart from the scope of the present invention or local replacement; All should belong to claim of the present invention and come restricted portion, therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.

Claims (12)

1. method that digital picture is strengthened comprises step:

Calculate the local luminance values and the local contrast value of each pixel in the input digital image;

Confirm the gain factor of each pixel based on said local luminance values and local contrast value; And

Use the gain factor of confirming to regulate the value of respective pixel.

2. the method for claim 1; The digital picture of wherein said input is the coloured image that comprises a plurality of Color Channels; Said method also comprises: in calculating input digital image, before the local luminance values and layout contrast value of each pixel, the digital picture of input is carried out the step of gray processing.

3. according to claim 1 or claim 2 method, the local luminance values of each pixel and the step of local contrast value comprise in the wherein said calculating input digital image:

Local luminance under the different scale is carried out weighted sum; And

Local contrast under the different scale is carried out weighted sum.

4. method as claimed in claim 3, wherein calculate brightness value and contrast value through following two formula respectively:

$\mathrm{lu}\min \mathrm{ance}\left(x\right)=\underset{i}{\overset{N}{\mathrm{\Σ}}}{W}_i*{\mathrm{mean}}_i\left(x\right)$

$\mathrm{contrast}\left(x\right)=\underset{i}{\overset{N}{\mathrm{\Σ}}}{W}_i*\log 10\left(\frac{G\left(x\right)}{{\mathrm{mean}}_i\left(x\right)}\right)$

In the above-mentioned formula, the position of x remarked pixel in input picture, the gray-scale value of G (x) statement pixel, mean _i(x) be illustrated in i the local mean value under the yardstick, W _iRepresent i the weighting coefficient under the yardstick.

5. according to claim 1 or claim 2 method, wherein confirm that based on said local luminance values and local contrast value the step of the gain factor of each pixel comprises through the computes gain factor:

$\mathrm{gain}\left(x\right)=\frac{\mathrm{upper}\_\mathrm{bound}\left(x\right)}{1-a*\exp (-\mathrm{strength}*(\mathrm{contrast}\left(x\right)-\mathrm{cross}\_\mathrm{point}\left(x\right)))}$

A=upper_bound in following formula (x)-1; Make gain (x)=1 when contrast (x)=cross_point (x); To entire image upper_bound (x) is set to a fixed value; Perhaps be adaptive to the value of each pixel, cross_point (x)=1-luminance (x), parameter s trength leave the user for and carry out the interactive mode adjusting.

6. method as claimed in claim 5, wherein

upper_bound(x)＝1/mean(x)

Or

upper_bound(x)＝1/(max(r(x)，g(x)，b(x))

Mean in the following formula (x) can be the average under any yardstick, the perhaps weighted sum of average under each yardstick.

7. equipment that digital picture is strengthened comprises:

The brightness calculation unit, the local luminance values of each pixel in the calculating input digital image;

The contrast computing unit, the local contrast value of each pixel in the calculating input digital image;

The gain factor of each pixel is confirmed in the gain calculating unit based on said local luminance values and local contrast value; And

Map unit uses the gain factor of confirming to regulate the value of respective pixel.

8. equipment as claimed in claim 7, the digital picture of wherein said input are the coloured images that comprises a plurality of Color Channels, and said equipment also comprises:

The gray processing unit before the local luminance values and layout contrast value of each pixel, carries out gray processing to the digital picture of input in calculating input digital image.

9. like claim 7 or 8 described equipment; Wherein said brightness calculation unit carries out the local luminance values that each pixel is calculated in weighted sum to the local luminance under the different scale, and said contrast computing unit carries out the local contrast value that each pixel is calculated in weighted sum to the local contrast under the different scale.

10. equipment as claimed in claim 9, wherein calculate brightness value and contrast value through following two formula respectively:

$\mathrm{lu}\min \mathrm{ance}\left(x\right)=\underset{i}{\overset{N}{\mathrm{\Σ}}}{W}_i*{\mathrm{mean}}_i\left(x\right)$

$\mathrm{contrast}\left(x\right)=\underset{i}{\overset{N}{\mathrm{\Σ}}}{W}_i*\log 10\left(\frac{G\left(x\right)}{{\mathrm{mean}}_i\left(x\right)}\right)$

In the above-mentioned formula, the position of x remarked pixel in input picture, the gray-scale value of G (x) statement pixel, mean _i(x) be illustrated in i the local mean value under the yardstick, W _iRepresent i the weighting coefficient under the yardstick.

11., wherein confirm that based on said local luminance values and local contrast value the step of the gain factor of each pixel comprises through the computes gain factor like claim 7 or 8 described equipment:

$\mathrm{gain}\left(x\right)=\frac{\mathrm{upper}\_\mathrm{bound}\left(x\right)}{1-a*\exp (-\mathrm{strength}*(\mathrm{contrast}\left(x\right)-\mathrm{cross}\_\mathrm{point}\left(x\right)))}$

A=upper_bound in following formula (x)-1; Make gain (x)=1 when contrast (x)=cross_point (x); To entire image upper_bound (x) is set to a fixed value; Perhaps be adaptive to the value of each pixel, cross_point (x)=1-luminance (x), parameter s trength leave the user for and carry out the interactive mode adjusting.

12. equipment as claimed in claim 11, wherein

upper_bound(x)＝1/mean(x)

Or

upper_bound(x)＝1/(max(r(x)，g(x)，b(x))

Mean in the following formula (x) can be the average under any yardstick, the perhaps weighted sum of average under each yardstick.

Images