CN101071502A - Image processing method - Google Patents
Image processing method Download PDFInfo
- Publication number
- CN101071502A CN101071502A CN 200610061346 CN200610061346A CN101071502A CN 101071502 A CN101071502 A CN 101071502A CN 200610061346 CN200610061346 CN 200610061346 CN 200610061346 A CN200610061346 A CN 200610061346A CN 101071502 A CN101071502 A CN 101071502A
- Authority
- CN
- China
- Prior art keywords
- pixel
- weights
- close attention
- brightness
- channel value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a method of image processing. The method is used to convert low-light images of dynamic closer to the real physical environment of high dynamic lighting effects images, including the following steps: Firt, will be concerned about pixel to pixel region around Operational value addition, the right to come to the pixel value and the pixel brightness as concerned about the extension, according to expand the scope of expansion of concern pixel brightness; Second, the concern surrounding region pixel-by-pixel sampling conducted will the channel surrounding area pixel values and concerns of the channel pixel values, it will all pixels within the region of weights under the rules of addition and subtraction and computing, according to results of operations pixel brightness adjustment concerns the precision ;Third, in accordance with step one and step-by-way deal with the images of all the pixels until the completion of the entire image processing.
Description
Technical field
The present invention relates to image processing techniques, relate in particular to and a kind ofly will hang down the method that the dynamic illumination effect image converts the high dynamic illumination effect image to.
Background technology
Along with the developing rapidly and popularizing of multimedia technology and the Internet, image processing techniques is subjected to unprecedentedly paying attention to widely, and enters into new field such as online game.Computing machine is to distinguish visual brightness with 8bit (256) level in presentation video and display display image, the promptly low dynamic illumination effect image (LDR) of common graphic file generally stores with 32 integers, each pixel in the picture has four passages of RGBA, accounting for 8 respectively, is that round values is preserved, that is: wherein any passage all is 0 to 255 round values in program, and in video card expression 0 to 1 color value, 0 expression is the darkest, and 1 expression is the brightest, and precision is 1/255.Such as: the value of R passage in program is 16, and then its value is 16/255 in video card.Nearly all video card is all supported the picture of this form.But this trivial hundreds of or several ten thousand/ precision be to reproduce the light conditions of true nature.Imagine the tonal range difference of brightness range in other words between the object of the luminous intensity of the sun and a black, considerably beyond 256 ranks.Therefore, a common scenery picture on daytime, look that it is same gray scale/brightness that the white clouds and the sun may all present, it all is pure white, but in fact the brightness between the white clouds and the sun can not be the same, difference in brightness between them is huge, and visible generic graphic file layout is very coarse, is far from recording the actual state of real world.Therefore, just having introduced HDR is that high dynamic illumination is played up technology.The principle of HDR technology is: adopt floating-point texture store a very large-scale monochrome information that exceeds 0 to 1 scope (0, Max), Max＞1.Compare with the technology of LDR,, thereby make the more approaching real image effect of the picture that is made into because the HDR technology has enlarged brightness range and luminance accuracy.
The picture of HDR effect to be produced in the prior art, the picture of the LDR of two different depth of exposures a HDR picture can be spliced into.Yet this method need be taken the photo of two same scenes with camera, by two exposure differences but the identical common photo in visual angle splice mutually, can not handle the single photo that has existed.Therefore, prior art can not satisfy the user any pictures is processed into the picture of HDR effect, lacks dirigibility.
Summary of the invention
The invention provides a kind of image processing method, in order to solve the picture that a low dynamic illumination effect picture is converted to a high dynamic illumination effect.
For addressing the above problem, the invention provides following technical scheme:
A kind of image processing method is characterized in that, comprises the steps:
Further comprising the steps of before the step 1: the maximum brightness value of setting required image.
Pay close attention to the pixel peripheral region described in step 1 and the step 2 and be square matrix shape zone arbitrarily.
Pixel weights described in the step 1 are according to being the center to pay close attention to pixel, according to from the centre to around the arbitrary value of the rule definition that successively decreases.
Channel value described in the step 2 is each channel value of pixel, respectively each channel value of pixel is calculated.
The weights of the pixel described in the step 1 and be to be the value that 255 pixel weights addition draws with channel value in the pixel on every side.
Enlarging the brightness of paying close attention to pixel described in the step 1 is meant according to the maximum brightness value of the image of setting and the weights and the brightness of each passage of expansion concern pixel respectively of each passage.
When the channel value of the channel value that is compared pixel greater than the concern pixel, then weights do not compare preceding weights and add the weights that are compared pixel with equaling in the step 2; When the channel value of the channel value that is compared pixel less than the concern pixel, then weights do not compare preceding weights and deduct the weights that are compared pixel with equaling; When the channel value that is compared pixel equals to pay close attention to the channel value of pixel, then weights and constant.
Paying close attention to pixel in the step 2 must compare one by one with all pixels in the peripheral region, the whole zone around traversal is paid close attention to pixel.
Adjusting the precision of paying close attention to pixel brightness described in the step 2 is meant according to the maximum brightness value of the image of setting and the weights and the precision that promotes each passage brightness of concern pixel respectively of each passage.
Description of drawings
Fig. 1 is the inventive method process flow diagram;
Fig. 2 is listed square matrix zone pixel weights figure;
Fig. 3 pays close attention to the pixel weights figure that pixel is supposed when the image border;
Fig. 4 pays close attention to pixel real pixel weights figure when the image border.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer,, the present invention is further elaborated below in conjunction with drawings and Examples.
With reference to figure 1, in the specific embodiment of the present invention, only processing procedure is described with the brightness adjustment of one of them passage of pixel, the brightness processed process of other passages is similar.
Step S1, set the maximum brightness value of required image earlier, be the maximum brightness value of pixel passage, common one is 100 near the maximum brightness value of the image of true effect, and the maximum brightness value that we set this image in the specific embodiment of the present invention is 100.
Step S2, the selected pixel regional pixel on every side of paying close attention to, this zone can be square matrix zone arbitrarily, and promptly the line number of this matrix area is identical with columns, and selected regional pixel be to pay close attention to the pixel pixel in 9 * 9 the square matrix zone on every side in the specific embodiment of the present invention;
Step S3 is for the pixel in the matrix area is given weights;
Generally speaking, the weights of these pixels by the position of pixel from the centre to around successively decrease, can be integer or floating number, for example, in the specific embodiment of the present invention, for the pixel in the matrix area is given weights as shown in Figure 2:
Wherein X is for paying close attention to the position of pixel, and other numerals are the weights of its position pixel.
When the position of paying close attention to pixel during in edge of image, at first according to generalized case, give not that the pixel on picture distributes weights, again with these not at the weights of the locational pixel of picture be adjusted at pixel weights on the picture.As shown in Figure 3, the position of the pixel of representing in the frame of broken lines is actually outside image, and then the rule of giving according to weights earlier gives weights for not the pixel on image.The weight table of the pixel on picture not is shown FactorRed, with all not the pixel on picture weights and be expressed as SumRed, be SumRed=∑ FactorRed, the weight table of pixel that will be on picture is shown FactorBlack, with the weights of the pixel on the picture be expressed as SumBlack, i.e. SumBlack=∑ FactorBlack.After guaranteeing that the pixel weights that picture is outer are distributed to pixel weights in the picture, the maximum weights summation of pixel remains unchanged around the concern pixel, the weights that add same ratio need for each pixel in picture, then the weights of pixel are that FactorBlack=Factorblack+Factorblack (SumRed/SumBlack) therefore pays close attention to the regional pixel weights of pixel when the image border as shown in Figure 4 in picture.
Step S41, all channel value are equaled the weights addition of 255 pixel, obtain weights and be expressed as SumFactor, the maximum of pixel weights and be expressed as MaxSumFactor in the zone, when the channel value of all pixels all equals 255 in the zone, pixel weights and be maximum SumFactor=MaxSumFactor.(scope in zone can be a square matrix arbitrarily, the weights of pixel also can for arbitrarily with the position of paying close attention to pixel from the centre to around the rule of the successively decreasing value of arranging, so SumFactor changes along with the variation of regional and weights.)
Step S42 adjusts the brightness value BrightNess that pays close attention to pixel.With the weights of pixel on every side and SumFactor and maximum weights and MaxSumFactorm as criterion, and the value of SumFactor/MaxSumFactorm is to pay close attention to the ratio value that pixel can rise to maximum brightness value, therefore draw BrightNess=max (MaxBright * (SumFactor/MaxSumFactor), 1) promptly: maximum brightness value MaxBright be multiply by this pay close attention to the ratio value that pixel can rise to maximum brightness value, because paying close attention to the brightness value of pixel before adjusting the concern pixel brightness is 1, the value that draws and 1 is got maximal value wherein, and the brightness value of then adjusting back concern pixel is a maximal value wherein.
Step S51 compares all pixels of paying close attention in pixel and the zone one by one, in the present invention, begins till last pixel of bottom right from upper left first pixel of image.To be subjected to the influence of other pixels on every side owing to pay close attention to the bright meeting of pixel, therefore will be according to the collection of the brightness of other pixels on every side being handled the brightness of paying close attention to pixel, at first to calculate the weights and the SumFactorI that pay close attention to all pixels of pixel peripheral region, secondly the channel value that will pay close attention to all pixels in the channel value of pixel and the zone compares, according to comparative result calculate weights with.When the channel value that is compared pixel when paying close attention to the channel value of pixel, just explanation is compared the brightness of pixel greater than the brightness of paying close attention to pixel, for reduce paying close attention to the pixel and the difference of pixel brightness on every side, therefore must promote the precision of paying close attention to pixel brightness, then the pixel weights compare preceding pixel weights and add the weights that are compared pixel with becoming, and promptly SumFactorI=SumFactorI+ is compared the pixel weights; Equally when the channel value that is compared pixel during less than the channel value of concern pixel, just need to reduce the precision of paying close attention to pixel brightness, then the weights of pixel compare preceding pixel weights and deduct the weights that are compared pixel with equaling, and both SumFactorI=SumFactorI-was compared the pixel weights; When the channel value that is compared pixel equals to pay close attention to the channel value of pixel, do not need to adjust the precision of paying close attention to pixel brightness, then pixel weights and constant, i.e. SumFactorI=SumFactorI.Traveled through draw after all pixels in the zone last pixel weights and.
Step S52 is according to the precision Precision of comparative result adjustment concern pixel brightness.The weights of all pixels that begin to calculate most and maximum are expressed as MaxSumFactorI, travel through all pixel gained pixel weights at last and be SumFactorI, the ratio that gained SumFactorI obtains divided by all pixel weights and MaxSumFactorI is exactly to adjust the standard of paying close attention to the pixel brightness precision, pixel has four passages in the LDR image, each passage all accounts for 8, luminance accuracy value before the concern pixel is not adjusted is 1/256, the precision of therefore adding pixel brightness just must multiply by 1/256, promptly increase by one (0,1/256) value between the scope, because-MaxSumFactorI＜=SumFactorI＜=MaxSumFactorI, for the luminance accuracy of guarantee adding (0,1/256) between, the luminance accuracy PrecisionAdd=1/256 that need add (1/2 (SumFactorI/MaxSumFactorI+1)) then, the precision of adjusting the brightness of paying close attention to pixel at last is Precision=1/256 (1/2 (SumFactorI/MaxSumFactorI+1)).
Execute above institute in steps after, the brightness of paying close attention to a passage of pixel has been adjusted and has been finished.In the present invention, each pixel of LDR picture has four passages of ARGB, repeats above-mentioned the processing of finishing all the other each passages of pixel in steps.For converting one whole LDR picture to the HDR picture, can be according to handling all pixels the picture successively from order left to bottom right.After all adjusting and finish, the brightness of all passages that all pixels in one pictures are comprised just finishes the processing procedure of whole picture.
Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.Those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (10)
1, a kind of image processing method is characterized in that, comprises the steps:
Step 1 is carried out additive operation with the pixel weights of paying close attention to the pixel peripheral region, draws the weights of pixel and as the spreading range of the brightness of paying close attention to pixel, enlarges the brightness of paying close attention to pixel according to spreading range;
Step 2, the pixel of paying close attention to the pixel peripheral region is sampled one by one, with the channel value of peripheral region pixel and the channel value of paying close attention to pixel relatively after, carry out plus and minus calculation with the weights of all pixels in the zone with according to rule, pay close attention to the precision of pixel brightness according to the operation result adjustment;
Step 3 is according to all pixels in step 1 and the step-by-way deal image, until finishing entire image processing.
2, according to the described method of claim 1, it is characterized in that, further comprising the steps of before step 1: the maximum brightness value of setting required image.
3, according to the described method of claim 1, it is characterized in that, pay close attention to the pixel peripheral region described in step 1 and the step 2 and be square matrix shape zone arbitrarily.
According to the described method of claim 1, it is characterized in that 4, the pixel weights described in the step 1 are according to being the center to pay close attention to pixel, according to from the centre to around the arbitrary value of the rule definition that successively decreases.
According to the described method of claim 1, it is characterized in that 5, channel value described in the step 2 is each channel value of pixel, respectively each channel value of pixel is calculated.
6, according to the described method of claim 1, it is characterized in that the weights of the pixel described in the step 1 and be to be the value that 255 pixel weights addition draws with channel value in the pixel on every side.
7, according to the described method of claim 1, it is characterized in that, enlarge the brightness of paying close attention to pixel described in the step 1 and be meant according to the maximum brightness value of the image of setting and the weights and the brightness of each passage of expansion concern pixel respectively of pixel.
According to the described method of claim 1, it is characterized in that 8, when the channel value of the channel value that is compared pixel greater than the concern pixel, then weights do not compare preceding weights and add the weights that are compared pixel with equaling in the step 2; When the channel value of the channel value that is compared pixel less than the concern pixel, then weights do not compare preceding weights and deduct the weights that are compared pixel with equaling; When the channel value that is compared pixel equals to pay close attention to the channel value of pixel, then weights and constant.
9, described according to Claim 8 method is characterized in that, paying close attention to pixel in the step 2 must compare one by one with all pixels in the peripheral region, the whole zone around traversal is paid close attention to pixel.
10, according to the described method of claim 1, it is characterized in that, adjust the precision of paying close attention to pixel brightness described in the step 2 and be meant according to the maximum brightness value of the image of setting and the weights and the precision that promotes each passage brightness of concern pixel respectively of pixel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100613468A CN100552713C (en) | 2006-06-23 | 2006-06-23 | A kind of image processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100613468A CN100552713C (en) | 2006-06-23 | 2006-06-23 | A kind of image processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101071502A true CN101071502A (en) | 2007-11-14 |
| CN100552713C CN100552713C (en) | 2009-10-21 |
Family
ID=38898709
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100613468A Active CN100552713C (en) | 2006-06-23 | 2006-06-23 | A kind of image processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100552713C (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101710429B (en) * | 2009-10-12 | 2012-09-05 | 湖南大学 | Illumination algorithm of augmented reality system based on dynamic light map |
| CN102724427A (en) * | 2011-12-01 | 2012-10-10 | 新奥特（北京）视频技术有限公司 | Method for rapidly realizing video image regionalization extremum color display |
| CN103795991A (en) * | 2014-02-08 | 2014-05-14 | 厦门美图之家科技有限公司 | High-dynamic-range image generating method based on single image |
| CN107547883A (en) * | 2016-06-23 | 2018-01-05 | 精工爱普生株式会社 | Image processor, display device and image treatment method |
| CN110392205A (en) * | 2018-04-20 | 2019-10-29 | 佳能株式会社 | Image processing equipment, device for display of message, control method and storage medium |
| US11184594B2 (en) | 2018-04-20 | 2021-11-23 | Canon Kabushiki Kaisha | Image processing apparatus, information display apparatus, control method, and computer-readable storage medium for improving image visibility |
-
2006
- 2006-06-23 CN CNB2006100613468A patent/CN100552713C/en active Active
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101710429B (en) * | 2009-10-12 | 2012-09-05 | 湖南大学 | Illumination algorithm of augmented reality system based on dynamic light map |
| CN102724427A (en) * | 2011-12-01 | 2012-10-10 | 新奥特（北京）视频技术有限公司 | Method for rapidly realizing video image regionalization extremum color display |
| CN102724427B (en) * | 2011-12-01 | 2017-06-13 | 新奥特（北京）视频技术有限公司 | A kind of quick method for realizing video image region extreme value color displays |
| CN103795991A (en) * | 2014-02-08 | 2014-05-14 | 厦门美图之家科技有限公司 | High-dynamic-range image generating method based on single image |
| CN107547883A (en) * | 2016-06-23 | 2018-01-05 | 精工爱普生株式会社 | Image processor, display device and image treatment method |
| CN107547883B (en) * | 2016-06-23 | 2021-06-08 | 精工爱普生株式会社 | Image processing device, display device, and image processing method |
| CN110392205A (en) * | 2018-04-20 | 2019-10-29 | 佳能株式会社 | Image processing equipment, device for display of message, control method and storage medium |
| US10887511B2 (en) | 2018-04-20 | 2021-01-05 | Canon Kabushiki Kaisha | Image processing apparatus, information display apparatus, control method, and computer-readable storage medium that output a combined image based on brightness values of pixels in an image |
| US11184594B2 (en) | 2018-04-20 | 2021-11-23 | Canon Kabushiki Kaisha | Image processing apparatus, information display apparatus, control method, and computer-readable storage medium for improving image visibility |
| CN110392205B (en) * | 2018-04-20 | 2021-12-31 | 佳能株式会社 | Image processing apparatus, information display apparatus, control method, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100552713C (en) | 2009-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Gladnet: Low-light enhancement network with global awareness | |
| CN100552713C (en) | A kind of image processing method | |
| CN101340511B (en) | Adaptive video image enhancing method based on lightness detection | |
| Meylan et al. | The reproduction of specular highlights on high dynamic range displays | |
| EP2624204B1 (en) | Image processing apparatus and method of controlling the same | |
| CN106169182B (en) | A method of synthesizing several different exposure images | |
| CN105450939A (en) | High dynamic range image display method and mobile terminal | |
| IES20060558A2 (en) | Image blurring | |
| CN102341826A (en) | Method for converting input image data into output image data, image conversion unit for converting input image data into output image data, image processing apparatus, display device | |
| Ward et al. | High dynamic range imaging & image-based lighting | |
| Meylan et al. | Tone mapping for high-dynamic range displays | |
| WO2011031331A1 (en) | Interactive tone mapping for high dynamic range video | |
| CN102663741A (en) | Method for carrying out visual stereo perception enhancement on color digit image and system thereof | |
| Herscovitz et al. | A modified multi scale retinex algorithm with an improved global impressionof brightness for wide dynamic range pictures | |
| CN106454144B (en) | A kind of bearing calibration of pair of Google glass image overexposure | |
| CN104680518A (en) | Blue screen image matting method based on chroma overflowing processing | |
| CN109920381B (en) | Method and equipment for adjusting backlight value | |
| CN115883755A (en) | Multi-exposure image fusion method under multi-type scene | |
| CN104299213A (en) | Method for synthesizing high-dynamic image based on detail features of low-dynamic images | |
| CN109859303B (en) | Image rendering method and device, terminal equipment and readable storage medium | |
| CN105118032B (en) | A kind of wide method for dynamically processing of view-based access control model system | |
| CN105338332B (en) | A kind of information processing method and electronic equipment | |
| McNamee et al. | Live HDR video streaming on commodity hardware | |
| CN106817542A (en) | The imaging method and imaging device of microlens array | |
| Amano | Manipulation of material perception with light-field projection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |