CN106851138A - A kind of image processing method based on HDR - Google Patents
A kind of image processing method based on HDR Download PDFInfo
- Publication number
- CN106851138A CN106851138A CN201710232452.6A CN201710232452A CN106851138A CN 106851138 A CN106851138 A CN 106851138A CN 201710232452 A CN201710232452 A CN 201710232452A CN 106851138 A CN106851138 A CN 106851138A
- Authority
- CN
- China
- Prior art keywords
- image
- low
- bit width
- pixel
- width
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/50—Control of the SSIS exposure
- H04N25/57—Control of the dynamic range
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/741—Circuitry for compensating brightness variation in the scene by increasing the dynamic range of the image compared to the dynamic range of the electronic image sensors
Abstract
The invention discloses a kind of image processing method based on HDR, comprise the following steps：Obtain the high-bit width view data of image capture device collection；Enter line position intercept operation to the high-bit width image of each frame of view data by setting order, generation at least three width have setting number of bits data and the different low-bit width image of brightness；The image quality parameter of each image is obtained, and calculates corresponding fusion weight factor；Down-sampled treatment carried out to pixel value and its fusion weight factor, and image to identical down-sampled dimension carries out weight fusion；Down-sampled reverse operating is carried out to weight fusion result, the high dynamic range images of a frame low-bit width are generated.High-bit width image can be mapped as low-bit width image by image processing method of the present invention based on HDR, and the dynamic range and resolution ratio of original image can be retained, and the different luminance pictures for being used derive from same exposure frame, without registration operation, have the advantages that to calculate simple and high real-time.
Description
Technical field
The present invention relates to technical field of image processing, and in particular to a kind of image processing method based on HDR.
Background technology
General CIS sensor or imageing sensor CMOS possess high-bit width sampling bit wide at present, but mesh
The display sampling bit wide of preceding most of displays is only 8 bits, and bit wide is lower than the sampling bit wide in CIS.Therefore in order to
The display bit wide of display is adapted to, the image of original high-bit width is generally carried out into cut position operation in image processing process, such as
Interception most-significant byte, generates 8 bit datas.But such processing procedure can cause damage to the dynamic range of original image.
Existing HDR algorithms, using row interlock height exposure method or using multiframe difference exposure method obtain multiframe
The initial data of different brightness, then obtains wide dynamic range image using the fusion of suitable fusion method.Staggered rows exposure
Mode can cause the loss of image resolution ratio, such as odd-numbered line exposure high (high brightness), the side of the low exposure of even number line (low-light level)
The image longitudinal frame of formula, high brightness and low-light level is the half of original resolution.The mode of the different exposures of multiframe, due to many
Order is arranged frame in time, if target is quickly moved, different exposure images needed to carry out image registration before fusion,
Image registration can reduce the real-time of shooting.
The content of the invention
The present invention is directed to above-mentioned deficiency of the prior art, there is provided high-bit width image is mapped as low-bit width figure by one kind
As, while retaining the dynamic range and resolution ratio and without the image processing method based on HDR of registration operation of original image.
In order to solve the above technical problems, present invention employs following method：
There is provided a kind of image processing method based on HDR, comprise the following steps：
Obtain the high-bit width view data of the bit wide more than or equal to 10 bits of image capture device collection；
Enter line position intercept operation, generation at least three to the high-bit width image of each frame of high-bit width view data by setting order
Width has setting number of bits data and the different low-bit width image of brightness；
The image quality parameter of every width low-bit width image is obtained, and according to the corresponding picture quality ginseng of every width low-bit width image
Number calculates corresponding fusion weight factor；
Pixel value and its fusion weight factor to each pixel of every width low-bit width image carry out down-sampled treatment, obtain
To the image of the identical down-sampled dimensions of multiple, and image to identical down-sampled dimension carries out weight fusion；
Weight fusion result to the image of identical down-sampled dimension carries out down-sampled reverse operating, generates a frame low-bit width
High dynamic range images.
Further, line position intercept operation is entered to the high-bit width image of each frame of high-bit width view data by setting order,
The step of generation at least three width have setting number of bits data and brightness different low-bit width image includes：
The pixel value of each pixel according to each vertical frame dimension bit wide image, obtains in the pixel value of the pixel from low
It is equal to the pixel value of target bit wide to height, as the low-bit width pixel value of the high brightness of the pixel, generates the low of high brightness
Bit wide image；
The high-bit width pixel value of the pixel is moved right setting value bit wide, obtain this move to right in rear pixel value from it is low to
Height is equal to the pixel value of target bit wide, as the low-bit width pixel value of the secondary high brightness of the pixel, generation time high brightness
Low-bit width image；
The high-bit width pixel value of the pixel is moved right setting value bit wide, this is obtained and is moved to right rear pixel value, if now
Pixel value bit wide after the pixel is moved to right is equal to target bit wide, moves to right end, and upper rheme interception step is repeated to all pixels point
Suddenly, the low-bit width image of low-light level is generated；
If moving to right rear pixel value more than target bit wide, the setting value bit wide step that then according to target bit wide is intercepted that moves right is repeated
Suddenly, the low-bit width pixel value that the pixel brightness is reduced on year-on-year basis is obtained, upper rheme interception step, generation is repeated to all pixels point
The low-bit width image that brightness falls on a year-on-year basis, until the pixel value bit wide after the pixel is moved to right is equal to target bit wide.
Further, the image quality parameter of every width low-bit width image is obtained, and it is corresponding according to every width low-bit width image
Image quality parameter calculate it is corresponding fusion weight factor the step of be：
Described image mass parameter includes contrast, saturation degree and exposure parameter, respectively to low-bit width image described in every width
Three image quality parameters of each pixel tested, and obtain corresponding contrast factor, the saturation degree factor and bright
, then be multiplied for three weight factors and obtain merging weight factor by degree weight factor.
Further, the fusion weight factor to each frame of the same pixel of low-bit width image described in multiframe is returned
One changes operation.
Further, the step tested the contrast of each pixel of each frame of the multiframe low-bit width image
Suddenly include：
Color-values fusion to tri- color channels of RGB of each pixel of low-bit width image described in every width obtains brightness
Parameter；
Laplace operator process of convolution is carried out to luminance parameter, contrast factor is obtained.
Further, the method to the saturation degree test of each pixel of low-bit width image described in every width is to calculate mesh
Tri- standard deviations of the color-values of color channel of RGB of pixel are marked, the saturation degree factor is obtained.
Further, the method to the exposure parameter test of each pixel of low-bit width image described in every width is to calculate
Tri- color-values of color channel of RGB of target pixel points and the distance of target light exposure degree, be then multiplied obtain luminance weights because
Son.
Further, the pixel value to the identical down-sampled dimension of each pixel of low-bit width image described in every width is carried out
The method of weight fusion to each color channel under each dimension, it is necessary to individually be merged.
Further, the method for down-sampled reverse operating being carried out to the weight fusion result of the image of identical down-sampled dimension
For, the weight fusion result of high-dimensional image is amplified 1 times, and with the weight fusion result phase of the image of adjacent low dimension
Plus, generate the weight fusion result of the new low dimensional image.
The beneficial effects of the invention are as follows：There is provided a kind of image processing method based on HDR, high-bit width image can be reflected
It is low-bit width image to penetrate, and can retain the dynamic range and resolution ratio of original image, and the different luminance pictures for being used
Same exposure frame is derived from, without registration operation, has the advantages that to calculate simple and high real-time.
Brief description of the drawings
Fig. 1 is the flow chart of image processing method of the present invention based on HDR.
Fig. 2 is the example flow that 10 bit images are mapped as image processing method of the present invention based on HDR 8 bit images
Figure.
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings：
As shown in figure 1, a kind of image processing method key step based on HDR of the present invention includes 101 to step 105.
In a step 101, the high-bit width view data of the bit wide more than or equal to 10 bits of image capture device collection is obtained,
In the present embodiment, 10 bit image sizes of image capture device collection are N*M, and each pixel index is (i, j), wherein
1≤i≤N, 1≤j≤M.
In a step 102, line position interception behaviour is entered to the high-bit width image of each frame of high-bit width view data by setting order
Make, generation at least three width have setting number of bits data and the different low-bit width image of brightness.Further, step 102 can
To be further refined as step S1 to step S3：
In step sl, the pixel value of each pixel according to each vertical frame dimension bit wide image, obtains the pixel
It is equal to the pixel value of target bit wide in pixel value from low to high, and casts out the pixel value beyond target bit wide, by acquired results
As the low-bit width pixel value of the high brightness of the pixel.Upper rheme interception step is repeated to all pixels point, generation is with frame
The low-bit width image of the high brightness of unit.
As shown in Fig. 2 intercepting the specific of the low-bit width image for obtaining high brightness by position to 10 bit original images below
Step is described in detail：
To each pixel Y10 (i, j) in original image, least-significant byte is taken as the right of high-brghtness picture images pixel YH (i, j)
Answer position pixel value：
If Y10 (i, j) >=255, then YH (i, j)=255；
Otherwise YH (i, j)=Y10 (i, j).
After so operating, YH maximums are 255, can save as 8 bits.
In step s 2, the high-bit width pixel value of the pixel is moved right setting value bit wide, setting value takes 1 here
Position, obtains during this moves to right rear pixel value and is equal to the pixel value of target bit wide from low to high, and cast out the picture beyond target bit wide
Element value, using acquired results as the secondary high brightness of the pixel low-bit width pixel value.Upper rheme is repeated to all pixels point to cut
Take step, the low-bit width image of generation time high brightness.
As shown in Fig. 2 below 10 bit original images are intercepted with the tool of the low-bit width image for obtaining time high brightness by position
Body step is described in detail：
Specifically, Y10 (i, j) is moved to right 1, generates 9 bit data Y9, take least-significant byte as the right of middle luminance picture YM
Answer position pixel value：
If Y9 (i, j) >=255, then YM (i, j)=255；
Otherwise, YM (i, j)=Y9 (i, j).
After so operating, YM maximums are 255, can save as 8 bits.
In step s3, the high-bit width pixel value of the pixel is moved to right 1, obtains this and move to right rear pixel value, now
Pixel value bit wide after the pixel is moved to right is equal to target bit wide.And upper rheme interception step, generation are repeated to all pixels point
The low-bit width image of low-light level.
As shown in Fig. 2 intercepting the specific of the low-bit width image for obtaining low-light level by position to 10 bit original images below
Step is described in detail：Specifically, Y10 (i, j) is moved to right 2, generates 8 bit datas, be defined as the correspondence of low-light level YL
Position pixel value.Repeats bits intercept operation generates low-luminosity picture YL, middle luminance picture YM and high-brghtness picture images to a frame end
YH。
In step 103, the image quality parameter of every width low-bit width image is obtained, and according to every width low-bit width image correspondence
Image quality parameter calculate corresponding fusion weight factor.Further, image quality parameter include contrast, saturation degree and
Exposure parameter.Three image quality parameters to each pixel of every width low-bit width image are tested respectively, with reference to specific
Parameter is calculated, and obtains corresponding contrast factor, the saturation degree factor and the luminance weights factor, and the luminance weights factor comes from
In exposure parameter.Then three weight factors are combined and obtains merging weight factor.Step 103 can be further refined as step
T1 to step T5：
Wrapped in step T1, the step of test the contrast of each pixel of each frame of low-bit width image
Include：
Color-values fusion to tri- color channels of RGB of each pixel of every width low-bit width image obtains brightness ginseng
Number.In the present embodiment, specifically, each pixel (i, j) to multiframe low-bit width image YL, YM and YH of different brightness enters
Row contrast measuring and calculation is obtaining contrast factor C.It is tri- colors of RGB of YL or YM or YH to define R, G, B first
Passage.
By taking YL as an example：The luminance picture parameter YLgray of YL is obtained, its calculation is：
YLgray (i, j)=0.2989*R (i, j)+0.5870*G (i, j)+0.1140*B (i, j).
Then Laplace operator process of convolution is carried out to luminance parameter, obtains contrast factor C.In the present embodiment,
Specifically, target is the contrast level parameter YL_C for obtaining YL.Here Laplace operator L_h=[0 10 is made；1-4 1；0 1
0], its calculation is：
YL_C (i, j)=YLgray (i, j) * 4+YLgray (i-1, j)+YLgray (i+1, j)+
YLgray(i,j-1)+YLgray(i,j+1)。
It is to calculate target in step T2, to the method for the saturation degree test of each pixel of every width low-bit width image
Tri- standard deviations of the color-values of color channel of RGB of pixel, obtain the saturation degree factor.In the present embodiment, specifically, mesh
It is designated as obtaining saturation degree factor S.By taking YL as an example, its calculation is：
Mu (i, j)=(R (i, j)+G (i, j)+B (i, j))/3.
It is to calculate mesh in step T3, to the method for the exposure parameter test of each pixel of every width low-bit width image
Tri- color-values of color channel of RGB of pixel and the distance of target light exposure degree are marked, being then multiplied obtains and exposure parameter pair
The luminance weights factor answered.Exposure has an ideal value for target, and this ideal value can be adjusted according to actual needs, for
Each color channel, preferable approximation should avoid being 0, i.e., under-exposed, or be 1, i.e. overexposure.
In the present embodiment, specifically, take 0.5 for ideal value, then assess each color channel of each pixel with
0.5 distance relation, they are multiplied, and obtain the corresponding luminance weights factor E of exposure parameter WE.By taking YL as an example, δ for deviate because
Son, could be arranged to 0.2 here, and its calculation is：
DR (i, j)=- exp ((R (i, j) -0.5)2/2σ2)
DG (i, j)=- exp ((G (i, j) -0.5)2/σ2)
DB (i, j)=- exp ((B (i, j) -0.5)2/2σ2)
YL_WE=dR (i, j) × dB (i, j) × dG (i, j).
In step t 4, it is that above three weight factor is multiplied to the computational methods for merging weight factor.In order to adjust
Weights influence when three image quality parameters are to final fusion, it is necessary to introduce proportion parameter para_C, para_S, para_WE,
Their default values are both configured to 1.In the present embodiment, specifically, by taking YL as an example：
YL_W (i, j)=YL_C (i, j)para_C+ YL_S (i, j)para_S+ YL_WE (i, j)para_WE。
Calculated in step T5, to the fusion weight factor of each frame of the same pixel of low-bit width image described in multiframe
After finishing, in addition it is also necessary to be normalized operation, to ensure the uniformity of fused image brightness.
In the present embodiment, specifically, our weight parameters to YH, HM, HL are normalized operation, by taking YL as an example：
At step 104, to each pixel of every width low-bit width image pixel value and its fusion weight factor is carried out
Down-sampled treatment, obtains the image of multiple identical down-sampled dimensions, and image to identical down-sampled dimension carries out weight fusion.
Further, it is necessary to individually be merged to each color channel under each dimension when carrying out weight fusion.
In the present embodiment, specifically, YL, YM, YH merge and obtain low-bit width HDR image.First to YL, YM,
The pixel value of YH and their fusion weight factor carry out down-sampled treatment, and 1, interval pixel is taken here to be carried out down-sampled, then dropped
Picture size after sampling is N/2*M/2.Then the image to identical down-sampled dimension carries out weight fusion.
Here down-sampled number of times is taken for 9 times, 9 yardsticks are generated, and by taking YL as an example, there is YLk, YL_Wk, YMk, YM_Wk, YHk,
YH_Wk, K here represents each yardstick, and its index value is natural number 0~8, wherein 0 represents original size, 1 dimension is by 0 dimension
Down-sampled generation, 2 dimensions by the down-sampled generation of 1 dimension, by that analogy.Under each yardstick, three luminance pictures each is led to
Road z is individually merged：Resultk (i, j, z)=YLk(i, j, z) × YL_Wk(i, j)+
YMk(i, j, z) × YM_Wk(i, j)+YHk(i, j, z) × YH_Wk(i, j)
In step 105, the weight fusion result to the image of identical down-sampled dimension carries out down-sampled reverse operating, raw
Into the high dynamic range images of a frame low-bit width.Further, the method for down-sampled reverse operating is, by the power of high-dimensional image
Weight fusion results amplify 1 times, as a result with the weight fusion results added of the image of adjacent low dimension, generate the new low dimensional
The weight fusion result of image.
In the present embodiment, specifically, 8 dimensions are amplified 1 times, as a result with 7 dimension image additions, generates 7 new dimensions
Image；7 new dimension images amplify 1 times, by result and 6 dimension image additions, generate 6 new dimension images；By that analogy.Most
A vertical frame dimension dynamic image is fused into eventually.
Specific embodiment of the invention is described above, this hair is understood in order to those skilled in the art
It is bright, it should be apparent that the invention is not restricted to the scope of specific embodiment, for those skilled in the art,
As long as in appended claim restriction and the spirit and scope of the present invention for determining, these changes are aobvious and easy to various change
See, all are using the innovation and creation of present inventive concept in the row of protection.
Claims (9)
1. a kind of image processing method based on HDR, it is characterised in that comprise the following steps：
Obtain the high-bit width view data of the bit wide more than or equal to 10 bits of image capture device collection；
Enter line position intercept operation, generation at least three width tool to the high-bit width image of each frame of high-bit width view data by setting order
There are setting number of bits data and the different low-bit width image of brightness；
The image quality parameter of every width low-bit width image is obtained, and according to the corresponding image quality parameter meter of every width low-bit width image
Calculate corresponding fusion weight factor；
Pixel value and its fusion weight factor to each pixel of every width low-bit width image carry out down-sampled treatment, obtain many
The image of individual identical down-sampled dimension, and image to identical down-sampled dimension carries out weight fusion；
Weight fusion result to the image of identical down-sampled dimension carries out down-sampled reverse operating, generates the height of a frame low-bit width
Dynamic image.
2. the image processing method based on HDR according to claim 1, it is characterised in that by setting order to high-bit width
The high-bit width image of each frame of view data enters line position intercept operation, and generation at least three width have setting number of bits data and bright
The step of spending different low-bit width images includes：
The pixel value of each pixel according to each vertical frame dimension bit wide image, obtains in the pixel value of the pixel from low to high
It is equal to the pixel value of target bit wide, as the low-bit width pixel value of the high brightness of the pixel, generates the low-bit width of high brightness
Image；
The high-bit width pixel value of the pixel is moved right setting value bit wide, this is obtained and is moved to right in rear pixel value from low to high etc.
The pixel value of target bit wide is same as, as the low-bit width pixel value of the secondary high brightness of the pixel, the low level of generation time high brightness
Image wide；
The high-bit width pixel value of the pixel is moved right setting value bit wide, this is obtained and is moved to right rear pixel value, if the now picture
Pixel value bit wide after vegetarian refreshments is moved to right is equal to target bit wide, moves to right end, and upper rheme interception step is repeated to all pixels point, raw
Into the low-bit width image of low-light level；
If moving to right rear pixel value more than target bit wide, repetition moves to right setting value bit wide and then the step of according to target bit wide is intercepted, and obtains
The low-bit width pixel value that the pixel brightness is reduced on year-on-year basis is obtained, upper rheme interception step is repeated to all pixels point, generate brightness
The low-bit width image for falling on a year-on-year basis, until the pixel value bit wide after the pixel is moved to right is equal to target bit wide.
3. the image processing method based on HDR according to claim 2, it is characterised in that obtain every width low-bit width image
Image quality parameter, and corresponding fusion weight factor is calculated according to the corresponding image quality parameter of every width low-bit width image
Step is：
Described image mass parameter includes contrast, saturation degree and exposure parameter, respectively to the every of low-bit width image described in every width
Three image quality parameters of individual pixel are tested, and obtain corresponding contrast factor, the saturation degree factor and brightness power
, then be multiplied for three weight factors and obtain merging weight factor by repeated factor.
4. the image processing method based on HDR according to claim 3, it is characterised in that also including to low described in multiframe
The fusion weight factor of each frame of the same pixel of bit wide image is normalized operation.
5. the image processing method based on HDR according to claim 3, it is characterised in that to the multiframe low-bit width figure
The step of contrast of each pixel of each frame of picture is tested includes：
Color-values fusion to tri- color channels of RGB of each pixel of low-bit width image described in every width obtains brightness ginseng
Number；
Laplace operator process of convolution is carried out to luminance parameter, contrast factor is obtained.
6. the image processing method based on HDR according to claim 3, it is characterised in that to low-bit width figure described in every width
The method of the saturation degree test of each pixel of picture is to calculate tri- color-values of color channel of RGB of target pixel points
Standard deviation, obtains the saturation degree factor.
7. the image processing method based on HDR according to claim 3, it is characterised in that to low-bit width figure described in every width
The method of the exposure parameter test of each pixel of picture is to calculate tri- color-values of color channel of RGB of target pixel points
With the distance of target light exposure degree, then it is multiplied and obtains the luminance weights factor.
8. the image processing method based on HDR according to claim 3, it is characterised in that to low-bit width figure described in every width
The pixel value of the identical down-sampled dimension of each pixel of picture carries out the method for weight fusion, it is necessary to every under to each dimension
Individual color channel is individually merged.
9. the image processing method based on HDR according to claim 1, it is characterised in that to identical down-sampled dimension
The method that the weight fusion result of image carries out down-sampled reverse operating is that the weight fusion result of high-dimensional image is amplified into 1
Times, and with the weight fusion results added of the image of adjacent low dimension, generate the weight fusion knot of the new low dimensional image
Really.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710232452.6A CN106851138B (en) | 2017-04-11 | 2017-04-11 | A kind of image processing method based on HDR |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710232452.6A CN106851138B (en) | 2017-04-11 | 2017-04-11 | A kind of image processing method based on HDR |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106851138A true CN106851138A (en) | 2017-06-13 |
| CN106851138B CN106851138B (en) | 2019-07-16 |
Family
ID=59147868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710232452.6A Active CN106851138B (en) | 2017-04-11 | 2017-04-11 | A kind of image processing method based on HDR |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106851138B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109348093A (en) * | 2018-09-26 | 2019-02-15 | 维沃移动通信有限公司 | A kind of image collecting device, method and terminal device |
| CN110225253A (en) * | 2019-06-12 | 2019-09-10 | 杨勇 | A kind of high dynamic range images processing method and system |
| CN110740229A (en) * | 2019-10-13 | 2020-01-31 | 中国航空工业集团公司洛阳电光设备研究所 | method for processing video image with liquid crystal screen capable of preventing afterimage and reducing display depth |
| CN111107274A (en) * | 2018-10-26 | 2020-05-05 | 北京图森智途科技有限公司 | Image brightness statistical method and imaging device |
| CN111292282A (en) * | 2020-01-21 | 2020-06-16 | 展讯通信（上海）有限公司 | Method and device for generating low-bit-width HDR image, storage medium and terminal |
| CN113409219A (en) * | 2021-06-28 | 2021-09-17 | 展讯通信（上海）有限公司 | Method and device for improving HDR image quality |
| WO2023015422A1 (en) * | 2021-08-09 | 2023-02-16 | 北京小米移动软件有限公司 | Image processing method and apparatus |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100329554A1 (en) * | 2009-06-29 | 2010-12-30 | Jiefu Zhai | Automatic exposure estimation for HDR images based on image statistics |
| CN102497490A (en) * | 2011-12-16 | 2012-06-13 | 上海富瀚微电子有限公司 | System and method for realizing image high dynamic range compression |
| CN105163047A (en) * | 2015-09-15 | 2015-12-16 | 厦门美图之家科技有限公司 | HDR (High Dynamic Range) image generation method and system based on color space conversion and shooting terminal |
| CN105184759A (en) * | 2015-09-22 | 2015-12-23 | 中国科学院西安光学精密机械研究所 | Image adaptive enhancement method based on histogram compactness transformation |
| CN105430298A (en) * | 2015-12-08 | 2016-03-23 | 天津大学 | Method for simultaneously exposing and synthesizing HDR image via stereo camera system |
| US20160125630A1 (en) * | 2014-10-30 | 2016-05-05 | PathPartner Technology Consulting Pvt. Ltd. | System and Method to Align and Merge Differently Exposed Digital Images to Create a HDR (High Dynamic Range) Image |
| CN106408547A (en) * | 2015-07-28 | 2017-02-15 | 展讯通信（上海）有限公司 | Image fusion method and apparatus, and terminal device |
| CN106506983A (en) * | 2016-12-12 | 2017-03-15 | 天津大学 | A kind of HDR video generation methods suitable for LDR videos |
| CN106530263A (en) * | 2016-10-19 | 2017-03-22 | 天津大学 | Single-exposure high-dynamic range image generation method adapted to medical image |
-
2017
- 2017-04-11 CN CN201710232452.6A patent/CN106851138B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100329554A1 (en) * | 2009-06-29 | 2010-12-30 | Jiefu Zhai | Automatic exposure estimation for HDR images based on image statistics |
| CN102497490A (en) * | 2011-12-16 | 2012-06-13 | 上海富瀚微电子有限公司 | System and method for realizing image high dynamic range compression |
| US20160125630A1 (en) * | 2014-10-30 | 2016-05-05 | PathPartner Technology Consulting Pvt. Ltd. | System and Method to Align and Merge Differently Exposed Digital Images to Create a HDR (High Dynamic Range) Image |
| CN106408547A (en) * | 2015-07-28 | 2017-02-15 | 展讯通信（上海）有限公司 | Image fusion method and apparatus, and terminal device |
| CN105163047A (en) * | 2015-09-15 | 2015-12-16 | 厦门美图之家科技有限公司 | HDR (High Dynamic Range) image generation method and system based on color space conversion and shooting terminal |
| CN105184759A (en) * | 2015-09-22 | 2015-12-23 | 中国科学院西安光学精密机械研究所 | Image adaptive enhancement method based on histogram compactness transformation |
| CN105430298A (en) * | 2015-12-08 | 2016-03-23 | 天津大学 | Method for simultaneously exposing and synthesizing HDR image via stereo camera system |
| CN106530263A (en) * | 2016-10-19 | 2017-03-22 | 天津大学 | Single-exposure high-dynamic range image generation method adapted to medical image |
| CN106506983A (en) * | 2016-12-12 | 2017-03-15 | 天津大学 | A kind of HDR video generation methods suitable for LDR videos |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109348093A (en) * | 2018-09-26 | 2019-02-15 | 维沃移动通信有限公司 | A kind of image collecting device, method and terminal device |
| CN111107274B (en) * | 2018-10-26 | 2021-01-08 | 北京图森智途科技有限公司 | Image brightness statistical method and imaging device |
| US11875486B2 (en) | 2018-10-26 | 2024-01-16 | Beijing Tusen Zhitu Technology Co., Ltd. | Image brightness statistical method and imaging device |
| CN111107274A (en) * | 2018-10-26 | 2020-05-05 | 北京图森智途科技有限公司 | Image brightness statistical method and imaging device |
| CN110225253A (en) * | 2019-06-12 | 2019-09-10 | 杨勇 | A kind of high dynamic range images processing method and system |
| CN110740229B (en) * | 2019-10-13 | 2021-09-10 | 中国航空工业集团公司洛阳电光设备研究所 | Video image processing method for preventing ghosting and reducing display depth of liquid crystal screen |
| CN110740229A (en) * | 2019-10-13 | 2020-01-31 | 中国航空工业集团公司洛阳电光设备研究所 | method for processing video image with liquid crystal screen capable of preventing afterimage and reducing display depth |
| CN111292282A (en) * | 2020-01-21 | 2020-06-16 | 展讯通信（上海）有限公司 | Method and device for generating low-bit-width HDR image, storage medium and terminal |
| CN111292282B (en) * | 2020-01-21 | 2023-02-17 | 展讯通信（上海）有限公司 | Method and device for generating low-bit-width HDR image, storage medium and terminal |
| EP4095793A4 (en) * | 2020-01-21 | 2023-10-25 | Spreadtrum Communications (Shanghai) Co., Ltd. | Method and apparatus for generating low bit width hdr image, storage medium, and terminal |
| CN113409219A (en) * | 2021-06-28 | 2021-09-17 | 展讯通信（上海）有限公司 | Method and device for improving HDR image quality |
| CN113409219B (en) * | 2021-06-28 | 2022-11-25 | 展讯通信（上海）有限公司 | Method and device for improving HDR image quality |
| WO2023015422A1 (en) * | 2021-08-09 | 2023-02-16 | 北京小米移动软件有限公司 | Image processing method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106851138B (en) | 2019-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106851138B (en) | A kind of image processing method based on HDR | |
| CN103826066B (en) | Automatic exposure adjusting method and system | |
| CN106341617B (en) | A kind of method and device of camera crosstalk compensation | |
| CN105578165B (en) | A kind of image white balancing treatment method, device and video camera | |
| JP5859080B2 (en) | Method and related apparatus for correcting color artifacts in images | |
| CN1820496B (en) | Image correcting method and imaging apparatus | |
| CN105185314A (en) | Uniformity compensation method for LED display screen | |
| WO2008148291A1 (en) | Device and method for calibrating multi-screen wall | |
| CN112669758B (en) | Display screen correction method, device, system and computer readable storage medium | |
| CN110390653B (en) | High-robustness DeMURA method for OLED screen | |
| DE102015105071A1 (en) | IMAGE DISPLAY DEVICE AND CONTROL PROCEDURE FOR THEM | |
| US20230410369A1 (en) | Method and apparatus for adjusting an image acquisition apparatus, compensation method of a display panel, device and medium | |
| CN104333707B (en) | Camera parameter method of adjustment and system | |
| CN106815873A (en) | The determination method and apparatus of camera internal reference | |
| CN110290370A (en) | Image processing method and device | |
| CN109191386A (en) | A kind of quick Gamma bearing calibration and device based on BPNN | |
| CN107113412B (en) | Display device, display correction device, display correction system and display bearing calibration | |
| CN101873504B (en) | Method for correcting automatic white balance | |
| CN101088297A (en) | Automatic white balance control | |
| CN102510502B (en) | Method and system for generating high-dynamic-range image | |
| CN110175967A (en) | Image defogging processing method, system, computer equipment and storage medium | |
| JP2010177917A (en) | White balance adjusting device, white balance adjusting method, and white balance adjusting program | |
| CN115388808B (en) | DLP projector parameter self-adjustment method for calibrating structured light sensor | |
| CN112104814B (en) | Automatic exposure adjusting method and device and LED display screen correcting method | |
| CN111489294A (en) | Image amplification processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |