CN115170420A - Image contrast processing method and system - Google Patents
Image contrast processing method and system Download PDFInfo
- Publication number
- CN115170420A CN115170420A CN202210798303.7A CN202210798303A CN115170420A CN 115170420 A CN115170420 A CN 115170420A CN 202210798303 A CN202210798303 A CN 202210798303A CN 115170420 A CN115170420 A CN 115170420A
- Authority
- CN
- China
- Prior art keywords
- image
- picture
- brightness
- processing
- illumination
- 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.)
- Withdrawn
Links
- 238000003672 processing method Methods 0.000 title claims description 3
- 238000012545 processing Methods 0.000 claims abstract description 39
- 238000005286 illumination Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 25
- 238000013528 artificial neural network Methods 0.000 claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 18
- 230000001678 irradiating effect Effects 0.000 claims abstract description 5
- 238000012544 monitoring process Methods 0.000 claims abstract description 5
- 230000002999 depolarising effect Effects 0.000 claims abstract description 4
- 230000002708 enhancing effect Effects 0.000 claims abstract description 4
- 238000000605 extraction Methods 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 206010047513 Vision blurred Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000005693 optoelectronics Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
- G06T5/75—Unsharp masking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and a system for processing image contrast, which comprises the following steps: training a neural network, and identifying a low-illumination image and a high-illumination image; acquiring a monitoring video through a camera, intercepting a required picture part, and constructing a brightness histogram for an original image; analyzing the image characteristics of the brightness histogram and calculating the average value of the brightness; selecting a picture area needing contrast enhancement; identifying pixel points of the picture; irradiating a real object corresponding to the part of the required picture through a light generating device in the camera, and receiving reflected light which irradiates the real object through a CCD camera; depolarizing the picture, and enhancing the color area in the picture; and (5) performing unsharp processing on the image. The method can accurately judge the illuminance condition of the image, accurately process the image, effectively improve the classification speed and the classification precision of the image, perform unsharp processing, obviously improve the contrast and the resolution of the image, effectively reduce the noise of the image and further improve the quality of the image.
Description
Technical Field
The present invention relates to the field of image contrast processing technologies, and in particular, to a method and a system for processing image contrast.
Background
The image contains huge amount of information, and today in the development of multimedia technology, people want to see clear and vivid pictures, which has higher and higher requirements on the quality of the image, and the processing technology of the required image is also promoted. In life, people find the photos so dark that some detailed information of the photos is not visible to the human eye. During the process of collecting, processing, storing, displaying or transmitting the image, the image may be affected by distortion of the optoelectronic system, noise interference, underexposure or overexposure, relative movement, transmission error code, etc., so that the quality of the image itself is reduced, which affects the subsequent processing of the image. It is known that images serve as carriers of information, and if the information carried in the images is masked, not only is our vision blurred, but also subsequent processing of the images is hindered. Therefore, it becomes important to enhance the contrast of the image
Under low-illumination conditions such as cloudy days, nights and object shielding, the obtained images often have the problems of small dynamic range, serious loss of detail information, large noise and the like. Therefore, low-light images can severely affect or even limit the performance of the human eye or computer vision system. The low-illumination image is enhanced, the definition of the image can be improved, the texture details of a scene are highlighted, the quality of the image is greatly improved, and therefore the data quality is guaranteed for completing tasks such as target identification and tracking and image segmentation. Therefore, the research on the enhancement of the low-illumination image has important theoretical significance and practical application value.
Disclosure of Invention
Based on the above technical problem, the present invention provides a method and a system for processing image contrast, and to achieve the above object, the present invention provides the following technical solutions:
a method of processing image contrast, comprising the steps of:
s1, training a neural network, and identifying a low-illumination image and a high-illumination image;
s2, acquiring a monitoring video through a camera, intercepting a required picture part, and constructing a brightness histogram for an original image;
s3, analyzing the image characteristics of the brightness histogram and calculating a brightness average value;
s4, selecting a picture area needing contrast enhancement; identifying pixel points of the picture;
s5, calculating a brightness difference value of the image with the recognized illumination, and reducing the brightness of the pixel point when the difference value between the brightness of the pixel point and the average brightness value is larger than a preset value;
when the difference value of the brightness of the pixel point subtracted from the average brightness value is larger than a preset value, increasing the brightness of the pixel point;
s6, irradiating the real object corresponding to the part of the required picture through a light generating device in the camera, and receiving reflected light irradiated to the real object through a CCD camera;
s7, repeating the step S5 for 5 times or more, depolarizing the picture, and enhancing the color area in the picture;
and S8, performing unsharp processing on the image, and performing the final contrast enhancement step on the image.
Preferably, the training of the neural network in step S4 includes:
s11, collecting a large number of paired normal illumination images and low illumination images as training samples;
s12, initializing a neural network algorithm model;
s13, performing network training, and taking the selected normal illumination image and the selected low illumination image as training samples as training data sets;
and S14, adjusting the weight, and repeatedly calculating by adopting the adjusted weight until the error meets the requirement.
Preferably, step S5 specifically further includes:
the brightness difference value calculated for the image with recognized illumination is based on the average exercise degree of the whole image.
Preferably, the unsharp processing on the image in step S8 specifically includes:
selecting proper parameter values according to the noise level of the selected picture, calculating the local variance, the gain coefficient and the expected value of the dynamic level of the selected picture, then calculating the local variance of the output image, calculating the expected value of the local variance of the image and the square error of the calculated value through a cost function, adaptively iterating and updating the scale vector through the square error, then calculating the enhanced image again, and repeating the steps until all pixel points of the selected picture are processed.
A system for processing image contrast, comprising:
the camera is used for receiving the direct-shot pictures;
the image extraction module is used for extracting the image shot by the camera;
and the brightness extraction module is used for extracting and calculating the brightness of the picture extracted by the image extraction module and calculating the average brightness value of the picture.
Preferably, the method further comprises the following steps:
the light generating device is arranged inside the camera and used as a light source and can emit natural light outwards;
the CCD camera is used for identifying the light reflected by the object irradiated by the light source emitted by the light generating device, identifying image information and converting the image information into a digital signal;
and the data processing computer is used for processing the digital signals.
Preferably, the method comprises the following steps:
and the neural network training module is used for training the neural network, so that the neural network can identify and distinguish the low-illumination image and the high-illumination image.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention identifies the illuminance of the selected picture by setting the neural network, processes the illuminance by the data processing computer, accurately judges the illuminance condition of the selected picture, accurately processes the illuminance, improves the convergence rate, reduces error estimation, accelerates the network iteration speed, enhances the judgment accuracy and effectively improves the classification speed and the classification precision of the image.
2. According to the invention, the light generation device and the CCD camera are arranged, so that the picture is extracted for a plurality of times and unsharp, the contrast and resolution of the picture can be obviously improved, the noise of the picture can be effectively reduced, and the quality of the picture is further improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a schematic diagram of the system of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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.
Referring to fig. 1 to 2, the present invention provides a technical solution:
a method of processing image contrast, comprising the steps of:
s1, training a neural network, and identifying a low-illumination image and a high-illumination image;
s11, collecting a large number of paired normal illumination images and low illumination images as training samples;
s12, initializing a neural network algorithm model;
determining the function f (x):
s13, performing network training, and taking the selected normal illumination image and the selected low illumination image as training samples as training data sets; calculate actual output and error:
in the formula: w ij Is the connection right; theta j Is a threshold value; x i Is the ith godAn input value via the primitive; y is j Is the actual output;
in the formula: t is j Is the desired output;
and S14, adjusting the weight, and repeatedly calculating by adopting the adjusted weight until the error meets the requirement.
In the formula: eta is the learning rate.
S2, analyzing the image characteristics of the brightness histogram and calculating a brightness average value;
s3, selecting a picture area needing contrast enhancement; identifying pixel points of the picture;
s4, acquiring a monitoring video through a camera, intercepting a required picture part, and constructing a brightness histogram for an original image;
s5, calculating a brightness difference value of the image with the identified illumination, and reducing the brightness of the pixel point when the difference value of the brightness of the pixel point minus the average brightness value is larger than a preset value;
when the difference value of the brightness of the pixel point subtracted from the average brightness value is larger than a preset value, increasing the brightness of the pixel point; the brightness difference value calculated for the image with recognized illumination is based on the average exercise degree of the whole image.
S6, irradiating the real object corresponding to the part of the required picture through a light generating device in the camera, and receiving reflected light irradiated to the real object through a CCD camera;
s7, repeating the step S5 for 5 times or more, depolarizing the picture, and enhancing the color area in the picture;
s8, performing unsharp processing on the image, selecting appropriate parameter values according to the noise level of the selected image, calculating the local variance, the gain coefficient and the expected value of the dynamic level of the selected image, then calculating the local variance of the output image, calculating the expected value of the local variance of the image and the square error of the calculated value through a cost function, adaptively updating the scale vector through the square error, then calculating the enhanced image again, repeating the steps until all pixel points of the selected image are processed, and performing the final contrast enhancement step on the image.
The enhanced image y (n, m) is obtained from the original image x (n, m):
y(n，m)＝x(n，m)+λz(n，m)
where z (n, m) is expressed as the filtered output signal and λ is a positive scale factor which controls the degree of contrast enhancement of the output image.
The formula for calculating the image contrast is as follows:
where δ (i, j) represents the difference in brightness between adjacent pixels, P δ (i, j) is the pixel distribution probability that the brightness difference between adjacent pixel points is delta.
After the unsharp processing, the contrast and the resolution of the picture are improved visually, the noise of the picture is reduced, the unsharp picture is superposed on the upper end of the picture with the processed light intensity, the processed picture is finally obtained, more details in the processed picture are displayed, and the contrast of the picture is obviously improved;
and because the light generator irradiates objects and the CCD camera which utilizes light reflection receives the objects, the picture which is exposed for a plurality of times is higher in brightness, so that when the original picture is processed to be a picture with low brightness, the superposed picture is more accordant with normal brightness, and the processed picture is more comfortable to see.
A system for processing image contrast, comprising:
the camera is used for receiving the direct-shot picture;
the image extraction module is used for extracting the image shot by the camera;
and the brightness extraction module is used for extracting and calculating the brightness of the picture extracted by the image extraction module and calculating the average brightness value of the picture.
The light generating device is arranged inside the camera and used as a light source and can emit natural light outwards;
the CCD camera is used for identifying light reflected by the object irradiated by the light source emitted by the light generating device, identifying image information and converting the image information into a digital signal;
and the data processing computer is used for processing the digital signals.
And the neural network training module is used for training the neural network, so that the neural network can identify and distinguish the low-illumination image and the high-illumination image.
Before the system, a large amount of training materials, a large amount of low-illumination pictures and high-illumination pictures are output to a neural network module, and the neural network is trained until the neural network can effectively identify the illumination condition of the pictures, so that the identification accuracy reaches an expected value;
then, acquiring a monitoring video through the camera, extracting a picture shot by the camera by an image extraction module, and also identifying and extracting a directly input picture, then extracting and calculating the brightness of the picture extracted by the image extraction module by a brightness extraction module, and solving the brightness average value of the picture, wherein the picture can be amplified, the brightness of each pixel point is monitored, and the brightness is extracted and calculated; finally, the image is subjected to contrast enhancement processing through a data processing computer;
if the picture is acquired by the camera, then starting the light generating device, irradiating the real object which is acquired as the picture for multiple times, identifying the reflection light source by the CCD camera, and enabling the pictures acquired for multiple times to enter a data processing computer to perform contrast enhancement processing on the pictures; the method greatly improves the quality of the taken picture through multiple measurements, can effectively reduce image noise and reduce the burden of subsequent picture processing.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. A method for processing image contrast, comprising the steps of:
s1, training a neural network, and identifying a low-illumination image and a high-illumination image;
s2, acquiring a monitoring video through a camera, intercepting a required picture part, and constructing a brightness histogram for an original image;
s3, analyzing the image characteristics of the brightness histogram and calculating a brightness average value;
s4, selecting a picture area needing to enhance the contrast; identifying pixel points of the picture;
s5, calculating a brightness difference value of the image with the recognized illumination, and reducing the brightness of the pixel point when the difference value between the brightness of the pixel point and the average brightness value is larger than a preset value;
when the difference value of the brightness of the pixel point subtracted from the average brightness value is larger than a preset value, increasing the brightness of the pixel point;
s6, irradiating the real object corresponding to the part of the required picture through a light generating device in the camera, and receiving reflected light irradiated to the real object through a CCD camera;
s7, repeating the step S5 for 5 times or more, depolarizing the picture, and enhancing the color area in the picture;
and S8, performing unsharp processing on the image, and performing the final contrast enhancement step on the image.
2. The method for processing image contrast according to claim 1, wherein the training of the neural network in step S1 comprises:
s11, collecting a large number of paired normal illumination images and low illumination images as training samples;
s12, initializing a neural network algorithm model;
s13, performing network training, and taking the selected normal illumination image and the selected low illumination image as training samples as training data sets;
and S14, adjusting the weight, and repeatedly calculating by adopting the adjusted weight until the error meets the requirement.
3. The method for processing image contrast according to claim 1, wherein the step S5 further includes:
the brightness difference value calculated for the image with recognized illumination is based on the average exercise degree of the whole image.
4. The method for processing image contrast according to claim 1, wherein unsharp processing on the image in step S8 specifically includes:
selecting appropriate parameter values according to the noise level of the selected picture, calculating the local variance, the gain coefficient and the expected value of the dynamic level of the selected picture, then calculating the local variance of the output image, calculating the expected value of the local variance of the image and the square error of the calculated value through a cost function, adaptively updating the scale vector through the square error, then calculating the enhanced image again, and repeating the steps until all pixel points of the selected picture are processed.
5. An image contrast processing system adapted to the image contrast processing method of claims 1 to 4, comprising:
the camera is used for receiving the direct-shot pictures;
the image extraction module is used for extracting the image shot by the camera;
and the brightness extraction module is used for extracting and calculating the brightness of the picture extracted by the image extraction module and calculating the average brightness value of the picture.
6. An image contrast processing system according to claim 5, further comprising:
the light generating device is arranged inside the camera and used as a light source and can emit natural light outwards;
the CCD camera is used for identifying light reflected by the object irradiated by the light source emitted by the light generating device, identifying image information and converting the image information into a digital signal;
and the data processing computer is used for processing the digital signals.
7. An image contrast processing system according to claim 6, comprising:
and the neural network training module is used for training the neural network, so that the neural network can identify and distinguish the low-illumination image and the high-illumination image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210798303.7A CN115170420A (en) | 2022-07-06 | 2022-07-06 | Image contrast processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210798303.7A CN115170420A (en) | 2022-07-06 | 2022-07-06 | Image contrast processing method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115170420A true CN115170420A (en) | 2022-10-11 |
Family
ID=83490873
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210798303.7A Withdrawn CN115170420A (en) | 2022-07-06 | 2022-07-06 | Image contrast processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115170420A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664411A (en) * | 2022-10-14 | 2023-08-29 | 天翼数字生活科技有限公司 | Low-illumination image processing method and system |
-
2022
- 2022-07-06 CN CN202210798303.7A patent/CN115170420A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664411A (en) * | 2022-10-14 | 2023-08-29 | 天翼数字生活科技有限公司 | Low-illumination image processing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109636754B (en) | Extremely-low-illumination image enhancement method based on generation countermeasure network | |
| CN108777815B (en) | Video processing method and device, electronic equipment and computer readable storage medium | |
| CN108197546B (en) | Illumination processing method and device in face recognition, computer equipment and storage medium | |
| CN111292264A (en) | Image high dynamic range reconstruction method based on deep learning | |
| CN108900769A (en) | Image processing method, device, mobile terminal and computer readable storage medium | |
| CN108875619B (en) | Video processing method and device, electronic equipment and computer readable storage medium | |
| CN111064904A (en) | Dark light image enhancement method | |
| CN115223004A (en) | Method for generating confrontation network image enhancement based on improved multi-scale fusion | |
| CN111225162B (en) | Image exposure control method, system, readable storage medium and camera equipment | |
| CN107392879B (en) | A kind of low-light (level) monitoring image Enhancement Method based on reference frame | |
| CN115063331B (en) | Multi-scale block LBP operator-based ghost-free multi-exposure image fusion method | |
| CN114862698B (en) | Channel-guided real overexposure image correction method and device | |
| CN113643214B (en) | Image exposure correction method and system based on artificial intelligence | |
| CN112561813B (en) | Face image enhancement method and device, electronic equipment and storage medium | |
| CN111242860A (en) | Super night scene image generation method and device, electronic equipment and storage medium | |
| CN111667498A (en) | Automatic moving ship target detection method facing optical satellite video | |
| CN115170420A (en) | Image contrast processing method and system | |
| CN114429476A (en) | Image processing method, image processing apparatus, computer device, and storage medium | |
| CN117152182B (en) | Ultralow-illumination network camera image processing method and device and electronic equipment | |
| CN116258653B (en) | Low-light level image enhancement method and system based on deep learning | |
| CN112365425A (en) | Low-illumination image enhancement method and system | |
| Park et al. | Image enhancement for extremely low light conditions | |
| CN114638764B (en) | Multi-exposure image fusion method and system based on artificial intelligence | |
| CN116188930A (en) | Scene recognition method and system based on fusion event camera | |
| CN113379631B (en) | Image defogging method and device |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20221011 |
|
| WW01 | Invention patent application withdrawn after publication |