CN109447915B - Line scanning image quality improving method based on characteristic model establishment and gamma gray correction - Google Patents
Line scanning image quality improving method based on characteristic model establishment and gamma gray correction Download PDFInfo
- Publication number
- CN109447915B CN109447915B CN201811266376.1A CN201811266376A CN109447915B CN 109447915 B CN109447915 B CN 109447915B CN 201811266376 A CN201811266376 A CN 201811266376A CN 109447915 B CN109447915 B CN 109447915B
- Authority
- CN
- China
- Prior art keywords
- image
- gray
- value
- line scanning
- gamma
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012937 correction Methods 0.000 title claims abstract description 18
- 239000013598 vector Substances 0.000 claims abstract description 38
- 230000009466 transformation Effects 0.000 claims abstract description 20
- 238000007781 pre-processing Methods 0.000 claims abstract description 19
- 238000009826 distribution Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 230000006872 improvement Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 18
- 230000008569 process Effects 0.000 abstract description 7
- 238000009434 installation Methods 0.000 abstract description 4
- 238000005286 illumination Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000003705 background correction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003702 image correction Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000002087 whitening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The invention discloses a line scanning image quality improving method based on characteristic model establishment and gamma gray correction, and relates to the field of image processing. Firstly, carrying out gray level histogram processing on an original gray level line scanning image, establishing a line scanning image characteristic model, wherein 1-k times of Gaussian fitting is carried out by counting the histogram of the column pixels, and the average value of the optimal Gaussian distribution is selected as the characteristic value of the column pixels; then, acquiring an equalized image of the image according to the image characteristic vector, and performing gray correction on the image by using an improved gamma transformation function to obtain a preprocessing result image; the preprocessing result graph can meet the quality requirement by fine tuning the value of the image feature vector. The invention can eliminate the laser light source influence of the line scanning image, has simple and convenient realization process, good and stable effect, and can still eliminate the laser light source influence when the installation angle, the light source and the shooting object of the line scan camera are not greatly changed after the image characteristic vector is well adjusted.
Description
Technical Field
The invention belongs to the field of image processing, and particularly relates to a line scanning image quality improving method based on characteristic model establishment and gamma gray correction, so as to achieve the purpose of solving the problem of line scanning image quality generated by laser light source brightness.
Background
The linear array scanning camera is used for the situation that an object to be acquired and the camera move relatively, the linear array scanning camera acquires the height, moves to the next bit length after acquiring one line and continues to acquire the next line, and a two-dimensional image can be spliced, although the two-dimensional image is a two-dimensional image, the height can be infinite, and the width is only a few pixels and is in a linear shape. This "infinitely long" image is truncated to an image of a certain height, called a line scan image.
The linear array camera has serious uneven brightness in the process of acquiring an original line scanning image, and the main reasons are from two aspects, namely the brightness in a camera field of view is uneven, the reflected light at the edge of the field of view is less and the illumination is insufficient; secondly, the response of each pixel in the linear array camera to illumination is not necessarily completely consistent. Therefore, the detail part of the original line scanning image is unclear, some parts are overexposed, the brightness of some parts is seriously insufficient, and the parts are difficult to distinguish by naked eyes, and meanwhile, the line scanning image in the same view field can also have the effect that the two sides are darker and the middle is brighter. These directly acquired raw line scan images cannot be applied in subsequent image processing.
At present, for the image quality problem caused by uneven brightness of a laser light source in the process of acquiring an original line scanning image by a linear array camera, related specific effective open technologies and solutions are not many.
Reference 1 (chenzhenlong, phyllotang, songsiewan, etc.. image correction method [ J ] optical science, 2013,33(07):243-249.) applied to a color line scanning machine vision system discloses a flat field correction method for a color line scanning image, which mainly relates to image denoising and gray scale correction, and uses a wavelet denoising method and improved polynomial fitting gray scale correction. However, the technical scheme is not suitable for the illumination non-uniformity correction of the gray scale line scanning image aiming at the color line scanning image.
Reference 2 (kuxia, wushuing, huangyue mountain, automatic Gamma correction of illuminance unevenness image [ J ]. microcomputer information, 2009,25(18): 292-. However, the technical scheme does not solve the problem of image quality caused by the brightness of the laser light source, and the gamma function provided by the technical scheme is not suitable for correcting the illumination unevenness of the gray-scale line scanning image, so that the generated effect is not good.
Disclosure of Invention
The invention provides a line scanning image quality improving method based on characteristic model establishment and gamma gray correction aiming at the problem of image quality of a gray line scanning image caused by the brightness of a laser light source, adopts a multi-Gaussian fitting gray histogram to establish an equalization method of an image characteristic model, and combines an improved gamma conversion method to enhance the line scanning image.
The invention provides a line scanning image quality improving method based on characteristic model establishment and gamma gray correction, which carries out the following processing on an obtained original gray line scanning image, wherein the processing process comprises the following steps:
step 2, acquiring an equalized image of the image according to the model vector;
step 3, carrying out gray correction on the image by utilizing the improved gamma transformation function to obtain a preprocessing result graph;
step 4, when the preprocessing result graph meets the quality requirement, finishing the processing, and outputting a preprocessing result graph; otherwise, the model vector is adjusted, and then execution continues to step 2.
In the step 1, each column of pixels of the gray-scale line scanning image is taken, the following steps (1) - (4) are executed, the characteristic values of the column are obtained, and the characteristic values of all the columns form a model vector of the image;
(1) counting the k-th 256-level gray histogram H for the pixel in the taken columnkAnd obtaining the peak T of the histogramkInitial k is 1;
(2) by the wave crest TkTaking N-level gray scale as center, and respectively cutting the range R of (2N +1) gray scale levelskPerforming Gaussian fitting to obtain kth Gaussian distribution and calculating Mean Square Error (MSE)kMean value MEkStandard deviation sigmak(ii) a Wherein the value range of N is an even number which is more than 15 and less than 25;
(3) judging whether the conditions are met: sigmakGreater than 0.35 or k is less than or equal to (256/(2N) -1), and if so, setting the current column of pixels in the current gray level range RkThe gray value in the column is set to zero, then k is increased by 1, and a 256-level gray histogram H is counted for the column of pixelskAnd find HkWave peak T ofkTurning to the step (2) for execution; if not, executing the step (4);
(4) and comparing the mean square errors corresponding to the k Gaussian distributions obtained currently, and selecting the mean value of the Gaussian distributions corresponding to the minimum mean square error as the characteristic value of a column of pixels taken currently.
In step 3, the improved gamma transformation function is as follows:
wherein c represents a gain coefficient, and γ is a gamma value; x is an input pixel gray value, and s is a gamma-transformed value of x.
In step 3, the pixel gray value t of the equalized image is changed into x, and x is t/255, then x is input into an improved gamma transformation function, and the output result s is multiplied by 255 to obtain a preprocessing result graph of the line scanning image.
The invention has the advantages and positive effects that: the method can effectively solve the problem that the brightness of the laser light source affects the line scanning image quality, has simple and convenient realization process and good effect, and can still eliminate the influence of the laser light source without changing the model vector on the premise that the installation angle of the linear array camera, the light source and a shooting object do not change greatly once the model vector is adjusted and determined. Therefore, the method is a stable preprocessing method for the original line scanning image.
Drawings
FIG. 1 is a schematic overall flow chart of a line scan image quality improving method according to the present invention;
FIG. 2 is a schematic flow chart of the method for establishing a line scan image feature model;
FIG. 3 is a line scan image to be processed in an embodiment of the present invention;
FIG. 4 is a schematic representation of a model vector obtained by the method of the present invention for FIG. 3 in an embodiment of the present invention;
FIG. 5 is a graph illustrating the effect of setting the gain c to 15 and the gamma to 0.6 in the gamma transformation using the initial model according to the embodiment of the present invention;
FIG. 6 is a diagram illustrating model vectors obtained from multiple adjustments in an embodiment of the present invention;
FIG. 7 is a graph illustrating the effect of setting the gain c to 15 and the gamma to 0.6 in the gamma transformation according to the embodiment of the present invention using the model shown in FIG. 6;
FIG. 8 is a graph of the effect of setting the gain c to 25 and γ to 0.6 in the gamma transformation using the model shown in FIG. 6 according to the embodiment of the present invention;
FIG. 9 is a graph of the effect of setting the gain c to 25 and γ to 0.7 in the gamma transformation using the model shown in FIG. 6 according to the embodiment of the present invention;
fig. 10 is a line scanning original image of other parts photographed under the condition that the installation angle of the line camera and the light source do not change greatly in the embodiment of the present invention;
FIG. 11 is a graph of the effect of using the final model of FIG. 6 applied to FIG. 10 in an embodiment of the present invention to set the gain c to 15 and γ to 0.6 in a gamma transformation.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
The invention relates to a line scanning image quality improving method based on characteristic model establishment and gamma gray correction, which mainly relates to two parts of image model establishment and image gray correction and improves an applicable gamma correction function. The overall flow of the present invention is shown in FIG. 1, and the steps are described below.
And 2, carrying out gray level histogram processing on the acquired image, and establishing a line scanning image characteristic model.
As shown in FIG. 2, the process of obtaining the model vector model is as follows:
(1) taking a column of pixels of a line scanning image to be processed, setting the ith column of pixels of the image, and counting the k-th 256-level gray level histogram of the column of pixels, wherein the initial k is 1; counting 256-level gray histogram H for the row for the first time1Obtaining the peak T of the histogram1。
(2) By the wave crest T1Taking the center, respectively cutting N-level gray scale to left and right, namely (2N +1) gray scale range RkThe first obtained gray level range is R1For the gray scale range R1Performing Gaussian fitting to obtain 1 st Gaussian distribution, and calculating the Mean Square Error (MSE) of the Gaussian distribution1Mean value ME1Standard deviation sigma1(ii) a Wherein, the value range of N is an even number which is more than 15 and less than 25.
Mean value ME of Gaussian distribution1I.e. the mathematical expectation, the mean square error MSE of the Gaussian distribution1For the gray scale range R1The square sum of the difference between the predicted value and the true value of the internal parameter is averaged, and the smaller the value of the mean square error is, the better the accuracy of the prediction model in describing experimental data is.
(3) When sigma1If it is greater than 0.35, the gray in the original gray histogram will beZeroing the frequency in the range of gray levels, i.e. setting the row of pixels in the range of gray levels R1Setting the gray value in the histogram to zero, counting the 256 gray level histograms, increasing k by 1, and obtaining a new gray level histogram H2Find H2Wave peak T of2Repeating the step (2) to obtain the 2 nd Gaussian distribution, and solving the mean square error MSE2Mean value ME2Standard deviation sigma2。
(4) Continuously repeating the step (3) when the sigma iskGreater than 0.35 or k<When the current gray level range is (256/(2N) -1), the current ith column of pixels is set to be in the current gray level range RkSetting the gray value in the gray value to zero, and enabling k to increase by 1; then, 256 levels of gray level histograms are counted to obtain a new gray level histogram HkIn HkFind the peak T inkSo as to obtain the k-th Gaussian distribution and the corresponding mean square error MSEkMean value MEkStandard deviation sigmak. Wherein, the value range of k is an integer which is more than 2 and less than m, and m is a positive integer which is not more than (256/(2N) -1).
(5) The ith row of pixels are subjected to Gaussian distribution fitting for m times, and the mean square error MSE corresponding to m Gaussian distributions is comparedkK belongs to (2, m)]And obtaining the minimum mean square error MMSE, wherein the mean value ME of the Gaussian distribution corresponding to the MMSE is a characteristic value representing the pixel of the row of the line scanning image.
(6) If the image width is w, the steps (1) to (5) are executed for each column of pixels, and the characteristic value of each column of pixels is obtained, so that w columns of pixels have w ME characteristic values and form a characteristic vector model, which is called a model vector.
It can be seen from the above process that the method of the present invention uses the obtained mean value ME of the gaussian distribution as the characteristic value of a column of pixels, and when obtaining the mean value ME, performs 1-k times of gaussian fitting on the histogram, and selects the best ME by comparison.
And 3, acquiring an equalized image of the line scanning image according to the acquired model vector.
Dividing gray values of all pixels of the linear array image to be processed by ME characteristic values of columns where the pixel points are located in the model vector, multiplying the values by a gain c to obtain a processed pixel gray value t, wherein the gray value t larger than 255 is 255, and the obtained image is the equalized image. Wherein, the gain c is the coefficient of gamma transformation, and the value range is generally [10,30 ].
And 4, performing gray scale correction by using an improved gamma (gamma) conversion method.
The expression of the original gamma transformation is as follows:
s＝T(r)＝crγ
r is the normalized pixel gray value of the image, and the value range is [0,1 ]; the exponential term γ is the gamma value, c is the gain coefficient, and s is the gamma-transformed value of r.
The gamma conversion is a nonlinear conversion, when gamma is less than 1, the dynamic range of the low-gray-scale pixel is improved, and the dynamic range of the high-gray-scale pixel is compressed; when γ >1, the dynamic range of the high-gradation pixels is increased, and the dynamic range of the gradation is decreased. Because the high-gray pixels of the line scanning image are generally strong in illumination, the overexposure of partial areas is removed, and the influence of the dynamic range of the compressed line scanning image on the image quality is small; and the dynamic range of the low-gray pixel is improved, so that the area with insufficient illumination can be more easily identified, and the part which is difficult to identify in gray becomes clear, therefore, gamma in the method is less than 1. In combination with the actual image, the value of gamma is set to be 0.5-0.75.
In the experiment, when gamma is less than 1, when gamma is larger, the effect of improving the detail part of the linear array scanning image is not obvious; when the value of gamma is smaller than 0.4, although the details are greatly improved, the contrast of the low-gray part of the image is reduced more, and a 'thinning' phenomenon is generated, namely the low-gray pixel value of the original image close to the black background is improved too much, so that the image is whitened. Therefore, it is difficult to achieve a desired image quality by merely adjusting the γ value.
In the gamma transformation expression, since r is equal to 0,1]Therefore, consider a mapping to r
On the premise of not changing gamma value, the shape of gamma conversion curve is adjusted to improve detail and inhibit whitening phenomenon as much as possible. Let pass through mThe gray value of the pixel after the odel vector equalization is t, and in order to ensure that the independent variable of gamma transformation is between 0 and 1, let
The mapping f is defined as
The x-y curve of the mapping function meets the requirement of improving the image, so the improved gamma transformation function is as follows:
and (3) taking the gray value t of the image pixel after model vector equalization in the step (3) as the input of the improved gamma transformation function, and multiplying the output result by 255 to obtain the line scanning image preprocessing result.
Step 5, judging whether the preprocessing result graph obtained in the step 4 meets the quality requirement, if so, ending the processing, and outputting a preprocessing result graph; if not, the model vector is manually fine-tuned to suppress the response nonuniformity among the pixels of the columns of the line scanning image, and then the execution is continued in the step 3.
And (5) repeatedly executing the steps 3-5, and iteratively modifying the values of the model vector elements according to the obtained preprocessing result graph effect until the preprocessing result of the line scanning image meets the quality improvement requirement.
When the model vector value is manually adjusted, the principle is to look at the effect after image adjustment, and smooth the adjustment of the line scanning image instead of the appearance of bright and dark stripes, wherein each row of stripes corresponds to one value in the model vector. This step is fine-tuned again with the gain and gamma parameters determined, the actual one being the eigenvalue ME. The model vector that does not require final adjustment is very smooth, only that the final map is free of abrupt changes in light and dark. If the range of a small interval is not uniform, making a model value in the range into an arithmetic series or taking an average value to adapt to the whole image; if only one or a few rows have jumps, the jumps can only be fine-tuned one by one according to experience.
Example (b): A2K linear array camera is adopted to shoot a train image, the image width w is 2048, and a 2800-height image is intercepted from the shot infinite-length image and is used as a line scanning image to be processed, as shown in figure 3. The method of the invention is performed on the image.
When the line scan image feature model is established, in this embodiment, a column of pixels of the line scan image to be processed is sequentially taken from left to right, and after each column is processed, the next column is processed, where N is 20, and the gray level range is 41 gray levels in total. In step (4) of this embodiment, the maximum value of m is 5. The number of cycles is m and the standard deviation sigma obtained by each calculationkJointly determine that m is less than or equal to 5 and sigmakAbove 0.35 the next cycle can be performed. The dimension of the model vector obtained in this embodiment is the image width 2048, as shown in fig. 4.
And then processing the pixel gray value of the equalized image by using the improved gamma transformation function, wherein the pixel gray value is as follows:
the output result is multiplied by 255 to obtain the line scan image preprocessing result.
Wherein the gain c affects the overall brightness of the image and gamma affects the contrast of the image.
And (4) repeatedly executing the step (4) by taking the gray value t of the image pixel after the model vector equalization in the step (3) as the input of the improved gamma transformation function, and repeatedly and iteratively modifying the value of the model vector element according to the obtained preprocessing result graph effect until the preprocessing result of the line scanning image meets the quality improvement requirement.
As shown in FIG. 5, the effect of c being 15 and γ being 0.6 is set in the gamma transform using the initial model. According to the obtained effect graph, a model vector is manually adjusted, the mutation position of the characteristic value ME in the graph 4 corresponding to the dark and light mutation position in the graph 5 needs to be manually adjusted and smoothed, and after multiple times of adjustment, the model vector is as shown in the graph 6. In this embodiment, for comparison, c is 15 or 25, and γ is 0.6 or 0.7. The concrete effect diagrams after gamma conversion using the final model are shown in fig. 7, 8 and 9.
As can be seen from fig. 8, the gain c is characterized in that the overall brightness of the image can be adjusted, the larger c, the brighter the image, and the smaller c, the darker the image. As can be seen from fig. 9, the gamma value is characterized by affecting the contrast of the image.
According to the comparison between the final effect image and the original line scanning image, the train part of the processed result image is clear and distinguishable, the contrast is good, and the method can be applied to subsequent image processing, so that the problem that the laser light source influences the quality of the line scanning image can be effectively solved.
In particular, fig. 10 shows a line-scanned image at a height of 2800, which is taken from an "infinitely long" image taken, that is, a line-scanned original image of another part taken without a large change in the mounting angle of the line-scan camera or the light source. FIG. 11 is a graph of the effect obtained after vector processing using the final model obtained in FIG. 6. Therefore, once the model vector is adjusted and determined, if the linear array camera is not greatly changed in installation angle, light source and shooting object, the model vector can be not changed, and the influence of the laser light source can still be eliminated. Therefore, the invention is a stable preprocessing method for the original line scan image.
It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Claims (4)
1. A line scanning image quality improvement method based on characteristic model establishment and gamma gray correction is used for carrying out the following processing on an obtained original gray line scanning image, and is characterized by comprising the following steps:
step 1, carrying out gray level histogram processing on a gray level line scanning image, establishing a line scanning image characteristic model, and obtaining a characteristic vector of the image, namely a model vector;
taking each column of pixels of the gray-scale line scanning image, and executing the following steps (1) - (4) to obtain the characteristic value of the column, wherein the characteristic values of all the columns form a model vector of the image;
(1) counting the k-th 256-level gray histogram H for the pixel in the taken columnkAnd obtaining the peak T of the histogramkInitial k is 1;
(2) by the wave crest TkTaking N-level gray scale as center, and respectively cutting the range R of (2N +1) gray scale levelskPerforming Gaussian fitting to obtain kth Gaussian distribution and calculating Mean Square Error (MSE)kMean value MEkStandard deviation sigmak(ii) a Wherein the value range of N is an even number which is more than 15 and less than 25;
(3) judging whether the conditions are met: sigmakGreater than 0.35 or k is less than or equal to (256/2N-1), and if so, setting the current column of pixels in the current gray scale range RkThe gray value in the column is set to zero, then k is increased by 1, and a 256-level gray histogram H is counted for the column of pixelskAnd find HkWave peak T ofkTurning to the step (2) for execution; if not, executing the step (4);
(4) comparing the mean square errors corresponding to k Gaussian distributions obtained currently, and selecting the mean value of the Gaussian distributions corresponding to the minimum mean square error as the characteristic value of a column of pixels taken currently;
step 2, acquiring an equalized image of the image according to the model vector;
step 3, carrying out gray correction on the image by utilizing the improved gamma transformation function to obtain a preprocessing result graph;
the improved gamma transformation function is as follows:
wherein c represents a gain coefficient, and γ is a gamma value; x is an input pixel gray value, and s is a value obtained by gamma conversion of x;
firstly changing the gray value t of the image pixel after equalization into x, wherein x is t/255, then inputting x into an improved gamma transformation function, and multiplying an output result s by 255;
step 4, when the preprocessing result graph meets the quality requirement, finishing the processing, and outputting a preprocessing result graph; otherwise, the model vector is adjusted, and then execution continues to step 2.
2. The method according to claim 1, wherein in step 2, the gray values of all pixels in the gray line scan image are divided by the characteristic values of the columns of the pixels in the model vector, and then multiplied by the gain c to obtain the gray value t of the processed pixel, and when t is greater than 255, the value of t is set to be 255, and the finally obtained image is an equalized image.
3. The method of claim 2, wherein the gain c is in the range of [10,30 ].
4. The method according to claim 1, wherein gamma is between 0.5 and 0.75.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811266376.1A CN109447915B (en) | 2018-10-29 | 2018-10-29 | Line scanning image quality improving method based on characteristic model establishment and gamma gray correction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811266376.1A CN109447915B (en) | 2018-10-29 | 2018-10-29 | Line scanning image quality improving method based on characteristic model establishment and gamma gray correction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109447915A CN109447915A (en) | 2019-03-08 |
| CN109447915B true CN109447915B (en) | 2021-06-29 |
Family
ID=65548860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811266376.1A Active CN109447915B (en) | 2018-10-29 | 2018-10-29 | Line scanning image quality improving method based on characteristic model establishment and gamma gray correction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109447915B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111553852B (en) * | 2020-04-13 | 2023-10-27 | 中国资源卫星应用中心 | Method and device for generating optical remote sensing image fast view |
| CN111695474B (en) * | 2020-06-03 | 2021-12-10 | 福建福特科光电股份有限公司 | Method for adaptively adjusting gamma curve of iris image of human eye |
| CN113543405B (en) * | 2021-09-15 | 2021-12-24 | 广州中大中鸣科技有限公司 | Method and device for reconstructing inter-frame dynamic gray scale based on LED lamp display |
| CN114757853B (en) * | 2022-06-13 | 2022-09-09 | 武汉精立电子技术有限公司 | Method and system for acquiring flat field correction function and flat field correction method and system |
| CN117011302B (en) * | 2023-10-08 | 2024-01-09 | 山东济宁运河煤矿有限责任公司 | Intelligent dry separation system based on coal gangue identification |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004077816A2 (en) * | 2003-02-27 | 2004-09-10 | Polaroid Corporation | Digital image exposure correction |
| CN104156719A (en) * | 2014-07-26 | 2014-11-19 | 佳都新太科技股份有限公司 | Face image light processing method on basis of shape and light model |
| CN104680490A (en) * | 2015-02-13 | 2015-06-03 | 中科创达软件股份有限公司 | Method for enhancing visuality of text image |
| CN107507215A (en) * | 2017-08-07 | 2017-12-22 | 广东电网有限责任公司珠海供电局 | A kind of power equipment infrared chart dividing method based on adaptive quantizing enhancing |
-
2018
- 2018-10-29 CN CN201811266376.1A patent/CN109447915B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004077816A2 (en) * | 2003-02-27 | 2004-09-10 | Polaroid Corporation | Digital image exposure correction |
| CN104156719A (en) * | 2014-07-26 | 2014-11-19 | 佳都新太科技股份有限公司 | Face image light processing method on basis of shape and light model |
| CN104680490A (en) * | 2015-02-13 | 2015-06-03 | 中科创达软件股份有限公司 | Method for enhancing visuality of text image |
| CN107507215A (en) * | 2017-08-07 | 2017-12-22 | 广东电网有限责任公司珠海供电局 | A kind of power equipment infrared chart dividing method based on adaptive quantizing enhancing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109447915A (en) | 2019-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109447915B (en) | Line scanning image quality improving method based on characteristic model establishment and gamma gray correction | |
| Huang et al. | Shallow-water image enhancement using relative global histogram stretching based on adaptive parameter acquisition | |
| KR100771158B1 (en) | Method AND System For Enhancement Color Image Quality | |
| US8422815B2 (en) | Image processing apparatus, image processing method and image processing program | |
| US10832388B2 (en) | Image tuning device and method | |
| CN112819721B (en) | Method and system for reducing noise of image color noise | |
| WO2011016375A1 (en) | Image processing apparatus, image processing method, and computer program | |
| CN106097286B (en) | A kind of method and device of image procossing | |
| CN108090887B (en) | Video image processing method and device | |
| CN115578297A (en) | Generalized attenuation image enhancement method for self-adaptive color compensation and detail optimization | |
| KR101113483B1 (en) | Apparatus for enhancing visibility of color image | |
| CN115965544A (en) | Image enhancement method and system for self-adaptive brightness adjustment | |
| CN112911166A (en) | Method, device, chip, medium and camera equipment for adjusting image brightness | |
| CN110570384A (en) | method and device for carrying out illumination equalization processing on scene image, computer equipment and computer storage medium | |
| CN112561829B (en) | Multi-region non-uniform brightness distortion correction algorithm based on L-channel Gamma transformation | |
| WO2020107308A1 (en) | Low-light-level image rapid enhancement method and apparatus based on retinex | |
| CN112488968A (en) | Image enhancement method for balanced fusion of degree-based histograms | |
| CN111080563B (en) | Histogram equalization method based on traversal optimization | |
| CN101478690B (en) | Image irradiation correcting method based on color domain mapping | |
| CN112991240B (en) | Image self-adaptive enhancement algorithm for real-time image enhancement | |
| CN115293989A (en) | Image enhancement method integrating MsRcR and automatic color gradation | |
| CN112272293A (en) | Image processing method | |
| Wang et al. | An Improved Unsupervised Color Correction Algorithm for Underwater Image | |
| CN112330689A (en) | Photovoltaic camera exposure parameter adjusting method and device based on artificial intelligence | |
| CN112862709B (en) | Image feature enhancement method, device and readable storage medium |
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 | ||
| CP03 | Change of name, title or address |
Address after: 9th floor, No. 61 Zhichun Road, Haidian District, Beijing, 100190 Patentee after: Beijing Aerospace Shenzhou Intelligent Equipment Technology Co.,Ltd. Patentee after: BEIJING AEROSPACE CTROWELL INFORMATION TECHNOLOGY CO.,LTD. Address before: 100080 No. 61, Haidian District, Beijing, Zhichun Road Patentee before: BEIJING CTROWELL INFRARED TECHNOLOGY Co.,Ltd. Patentee before: BEIJING AEROSPACE CTROWELL INFORMATION TECHNOLOGY CO.,LTD. |
|
| CP03 | Change of name, title or address |