CN106454014A - Method and device for improving quality of vehicle image captured in backlighting scene - Google Patents
Method and device for improving quality of vehicle image captured in backlighting scene Download PDFInfo
- Publication number
- CN106454014A CN106454014A CN201610962047.5A CN201610962047A CN106454014A CN 106454014 A CN106454014 A CN 106454014A CN 201610962047 A CN201610962047 A CN 201610962047A CN 106454014 A CN106454014 A CN 106454014A
- Authority
- CN
- China
- Prior art keywords
- image
- component
- blc
- yuv
- component image
- 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
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/80—Camera processing pipelines; Components thereof
- H04N23/81—Camera processing pipelines; Components thereof for suppressing or minimising disturbance in the image signal generation
-
- 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
- H04N23/76—Circuitry for compensating brightness variation in the scene by influencing the image signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N9/00—Details of colour television systems
- H04N9/64—Circuits for processing colour signals
- H04N9/646—Circuits for processing colour signals for image enhancement, e.g. vertical detail restoration, cross-colour elimination, contour correction, chrominance trapping filters
Abstract
The invention discloses a method and device for improving the quality of a vehicle image captured in a backlighting scene. According to the method and the device, backlighting processing is carried out on a brightness Y component through utilization of a gray scale mapping curve; a backlighting photographing dark area of the image is adjusted; a visual effect of the dark area is improved; moreover, color components are modified through combination of backlighting processing degree of the brightness Y component; and the color saturation and color bright degree of the image are prevented from being changed. Compared with the prior art, the method and the device have the advantages that the capturing effect of the backlighting scene is self-adaptively enhanced according to image partial feature information; the details in the dark area of the backlighting image is improved, and the highlight details and a partial contrast ratio of the image can be retained well. The method and the device are simple to realize in engineering application and is well-adapted. The universality and validity of a processing method for the image captured by a traffic road camera in the backlighting scene are improved.
Description
Technical field
The present invention relates to the video image processing technology of intelligent transportation field, more particularly to a kind of raising backlight
The method and device of scene vehicle snapshot picture quality.
Background technology
As modern transportation Road Development and Intelligent traffic management systems are popularized, traffic safety is more aobvious important.Electronics is warned
Examine, intelligent high definition bayonet socket, the separated householder methods such as evidence obtaining, radar overspeed snapping of stopping play pivotal role in intelligent transportation field.
The candid camera that installs in ordinary road, due to being constrained by extraneous photoenvironment, image quality can go out
Existing decay, affects effective evidence obtaining of illegal vehicle.In China's most area, for the traffic route of east-west direction, road
Road vehicles are captured image and backlight scene often occur, i.e. the Vehicle Object that is taken causes to capture along the direction running of light
The middle vehicle characteristics details of image frame effectively can not be presented.It is primarily due in video imaging system, generally adopts
Automatization's exposal control method using image mean flow rate come with static exposure reference value, cause in backlight scene hypograph
Local overexposure or local are excessively dark.How under backlight scene, it is effectively improved and reappears the key feature information for capturing image, become
For current intelligent transportation system and image processing field extreme stubborn problem.
Image effect is captured in order to improve the road vehicle under backlight scene, traditional Main has several as follows:
Method one, modification automatization spectrum assignment strategy, that is, increase imaging aperture time or improve imaging gain parameter,
So that the mistake dark areas details under backlighting condition is presented.The method realizes simple, easily operation, and subject matter is to aggravate
The loss of high bright part details in image frame, causes effective imaging region imperfect.In addition, high speed moving vehicle is shot,
Increasing aperture time can cause motion blur, excessively improve gain also can enlarged drawing noise, affect visual effect.
Method two, increase light filling equipment, that is, imaging sensitivity is increased, the method needs candid camera and light filling equipment
Effectively synchronous, improve hardware cost and system control complexity.
Method three, using the video camera with high frame per second and concrete wide dynamic complex functionality, the method mainly passes through
The mode of high dynamic range images (High-Dynamic Range, HDR) increases the dynamic range of image, high frame per second instantaneous exposure
Simultaneously multiframe synthesizes final image for imaging, reduced bright and crossed dark areas, and dynamic range is limited, can not fully meet various multiple
Miscellaneous backlight environment, under some special occasions, often effect be not clearly as camera Costco Wholesale costliness causes very
Difficult popularization.
Method four, the method for being processed based on picture signal, realize the image gray-scale level reduction in backlight region, and such as curve draws
Stretch, the balanced and color space conversion of GTG etc., the method is Global treatment mostly, causes image level sense not strong.
Content of the invention
It is an object of the invention to improving, above-mentioned traditional method is not enough present in practical implementation, proposition one kind is carried
The method and device of high backlight scene vehicle snapshot picture quality, mainly increases according to image local feature information self-adapting ground local
Candid photograph effect under strong backlighting condition, realizes simply, and adaptivity is strong, improves under backlight scene and captures the general of image processing method
Adaptive and effectiveness.
The present invention is achieved by the following technical solutions:
The invention provides a kind of method for improving backlight scene vehicle snapshot picture quality, comprises the following steps:
Step S1:Data after obtaining the initial data for capturing two field picture and processing:
Step S101:The yuv format data flow of pending video image is obtained using image capture module, is currently grabbed
The YUV image data of frame are clapped, obtains Y component map picture, U component image and the V component image of YUV image data;
Step S102:The YUV image data of step S101 are converted to rgb image data, calculate the R of rgb image data
Component image, G component image, the maximum of B component image, corresponding for maximum component image is labeled as lightness component figure
Picture, is designated as V';
Step S103:Overall average Ymean of the Y component map picture of calculation procedure S101, according to the overall equal of Y component map picture
Value divides the brightness degree Ygrade of Y component map picture;
Step S104:The Y component map picture of step S101 is smoothed, and filtering result is calculated, obtain
Yfilter component image;
Step S2:Calculate the luminance component image after BLC is processed:
Step S201:Build brightness mapping relations curve:
Wherein, variable x represents input signal, and x is the numerical value after normalized, and scope is 0~1;Variable y represents
Mapping relations curve output valve, scope is 0~1;Variable ctrl_y is the amplitude control parameter of mapping curve, its characteristic having
For:Different ctrl_y numerical value have different response curves, and ctrl_y numerical value is less, and the lifting amplitude of mapping curve is bigger;Conversely, reflecting
The lifting amplitude for penetrating curve is less;
Step S202:The Yfilter component image of step S104 is normalized, using the brightness of step S201
Mapping relations curve calculates output valve and output valve is multiplied by image gray-scale level maximum, is as a result expressed as Ymp;Image ash herein
Rank maximum is determined by picture depth, if image is N bit depth image, then image gray-scale level maximum is 2N-1;
Step S203:The mapping result of the V' component image of calculation procedure S102, obtains Vmp component image:
Step S204:Y component map picture and Vmp component image are carried out proportion weighted fusion, will be merged through proportion weighted
Data afterwards are designated as Yout as the final luminance component image for capturing two field picture;
Step S3:Calculate the color component images after BLC is processed:
Due to carrying out backlight process to luminance Y component, i.e., the dark of image reversible-light shooting is have adjusted using GTG mapping curve
Area, improves the visual effect of dark space, but also causes the color saturation of image and bright-colored degree to change simultaneously, institute
Color component is revised to need the backlight degree for the treatment of with reference to luminance Y component, specially:
Backlight process is carried out to U component image and V component image, is obtained result and is expressed as Uout and Vout:
In formula, ctrl_uv represents the control intensity of color component;
Step S4:Export the candid photograph frame YUV image data after BLC is processed:By Yout component image, Uout
Component image and Vout component image are used as the final YUV image output for capturing frame.
Further, the entirety of Y component map picture, in step S103, is calculated using the method for point traversal statistics pixel-by-pixel
Average Ymean.
Further, in step S104, using two-dimensional convolution wave filter, Y component map picture is smoothed, then
Filtering result is calculated from Gaussian filter or mean filter, obtain Yfilter component image.
Further, in step S201, ctrl_y numerical value is determined according to brightness degree Ygrade, brightness degree coefficient
Bigger, ctrl_y value is bigger.
Further, in step S204, the weight coefficient of proportion weighted fusion passes through the GTG of respective components image
Scope carries out value, if the index variables of component image GTG are i, in Y component map picture and Vmp component image, GTG i is corresponded to
Weight coefficient be Wy (i) and Wv (i), wherein:
Wy (i)=1-Wv (i)
In formula, threshold value La and threshold value Lb are the control parameters of respective components image gray-scale level Weighted Fusion.Parameter, Δ w represents right
Answer the datum quantity of component image weight coefficient.
Further, in step S3, ctrl_uv numerical value carries out value according to brightness degree coefficient Ygrade.
Present invention also offers the device of backlight scene vehicle snapshot picture quality is improved using said method, including image
Acquisition module, image BLC processing module and image display, wherein:
Described image acquisition module is used for gathering the yuv format data flow diagram picture of moving target (vehicle), and is currently grabbed
The YUV image data is activation of frame is clapped to image BLC processing module;
Described image BLC processing module, will using the method for above-mentioned raising backlight scene vehicle snapshot picture quality
The current YUV image data for capturing frame carry out BLC process, the enhanced YUV image number of the BLC after acquisition process
According to being then forwarded to image display;
Described image display module is used for the candid photograph image after showing BLC process.
The present invention has advantages below compared to existing technology:The invention provides a kind of improve backlight scene vehicle snapshot figure
As the method and device of quality, the method strengthens the candid photograph effect of backlight scene according to image local feature information self-adapting,
While backlight dark picture areas details is lifted, can preferably retain highlighted details and Image Warping.In engineer applied
Upper realization is simple, and adaptivity is strong, improves traffic route video camera and captures the pervasive of image processing method under backlight scene
Property and effectiveness.
Description of the drawings
The step of Fig. 1 is the method for improving backlight scene vehicle snapshot picture quality flow chart;
The step of Fig. 2 is for obtaining data after raw image data and process flow chart;
Fig. 3 is the step of calculating the luminance component image after BLC is processed flow chart;
Fig. 4 is the structural representation of the device for improving backlight scene vehicle snapshot picture quality.
Specific embodiment
Below embodiments of the invention are elaborated, the present embodiment is carried out under premised on technical solution of the present invention
Implement, detailed embodiment and specific operating process is given, but protection scope of the present invention is not limited to following enforcements
Example.
Embodiment 1
A kind of method and apparatus for improving backlight scene vehicle snapshot picture quality is present embodiments provided, as Fig. 1-3 institute
Show, methods described step includes:
Step S1:Data after obtaining the initial data for capturing two field picture and processing:
Step S101:The yuv format data flow of pending video image is obtained using image capture module, is currently grabbed
The YUV image data of frame are clapped, obtains Y component map picture, U component image and the V component image of YUV image data;
In the present embodiment, the YUV image data of candid photograph are 8bit depth image, and the YUV image data after candid photograph are stored in slow
Deposit operation interval.
Step S102:Designed image color space converter, the YUV image data of step S101 are converted to RGB image
Data, and caching operation interval is stored in, the matrix coefficient configuration of color of image space convertor is as follows:
As YUV image data are that 8bit, the rgb image data after changing is also 8bit.
Rgb image data is read from caching operation interval, calculate R component image, G component image, the maximum of B component image
Value, and the component corresponding to maximum is labeled as lightness component image V', i.e.,:
V'=max (R, G, B)
For 8bit depth image, the span of V' is 0~255, and V' component image is stored in caching operation interval.
Step S103:Y component map picture is read from caching operation interval, by the way of point traversal statistics pixel-by-pixel, calculate Y
The overall average of component image, is labeled as Ymean.As YUV image is 8bit depth image, the therefore span of Ymean
For 0~225.The brightness degree Ygrade of Y component map picture is divided according to the overall average of Y component map picture:
When the span of Ymean is 0~16 interval, brightness degree Ygrade value 0;
When the span of Ymean is 17~64 interval, brightness degree Ygrade value 1;
When the span of Ymean is 64~90 interval, brightness degree Ygrade value 2;
When the span of Ymean is 91~255 interval, brightness degree Ygrade value 3.
Step S104:Y component map picture is read from caching working area, using two-dimensional convolution wave filter, Y component map picture is carried out
Smoothing processing, then calculate filtering result.In order to the spatial neighborhood relation of image pixel is introduced, realization is reached to Y component map
The purpose of the smooth effect of picture, can select Gaussian filter, or mean filter is filtered the calculating of sharpening result.This
In embodiment, select the two-dimentional mean filter of 3 × 3 Size of Neighborhoods to calculate the filtering result of Y component map picture, and will put down
Slipped Clove Hitch fruit is expressed as Yfilter.Finally filtering result Yfilter is deposited into caching operation interval.
Step S2:Calculate the luminance component image after BLC is processed:
Step S201:Build brightness mapping relations curve:
Wherein, variable x represents input signal, and x is the numerical value after normalized, and scope is 0~1;Variable y represents
Mapping relations curve output valve, scope is 0~1;Variable ctrl_y is the amplitude control parameter of mapping curve, its characteristic having
For:Different ctrl_y numerical value have different response curves, and ctrl_y numerical value is less, and the lifting amplitude of mapping curve is bigger;Conversely, reflecting
The lifting amplitude for penetrating curve is less.
In the present embodiment, the value of ctrl_y numerical value is determined according to the brightness degree Ygrade for capturing two field picture, its scope
Interval is 0.1~2.0:
During current brightness level coefficient Ygrade value 0, ctrl_y span is 0.1~0.2;
During current brightness level coefficient Ygrade value 1, ctrl_y span is 0.2~0.5;
During current brightness level coefficient Ygrade value 2, ctrl_y span is 0.5~1.0;
During current brightness level coefficient Ygrade value 3, ctrl_y span is 1.0~2.0.
Step S202:The Yfilter component image of step S104 is normalized, using the brightness of step S201
Mapping relations curve calculates output valve and output valve is multiplied by image gray-scale level maximum, and result is expressed as Ymp;The present embodiment
Image is 8bit depth image, and corresponding GTG maximum is 255;
Step S203:Mapping process is carried out to the V' component image of step S102, Vmp component image is obtained, and is stored in slow
Deposit operation interval:
Step S204:Y component map picture and Vmp component image are carried out proportion weighted fusion:
Yout=Y*Wy+Vmp*Wv
In formula, Wy is the weight coefficient of Y component map picture, and span is the weighting system of 0~1, Wv for Vmp component image
Number, span is 0~1, and Wy+Wv=1, as long as so determine Wy and Wv parameter one of those.Due to Y-component and
Vmp component is all 8bit data, and grey-scale range is 0~255.So weight coefficient (Wy and Wv) can be according to different grey-scale range
Value.
In the present embodiment, it is assumed that index variables of the GTG between scope 0~255 are i, then grey inside Vmp component image
Corresponding weight coefficient Wv (i) calculating formula of rank i is as follows:
In formula, parameter i represents the index value of 8bit image gray-scale level, span 0~255;Threshold value La and threshold value Lb are figures
As the control parameter of GTG Weighted Fusion, threshold value La span 0~64, threshold value Lb span 192~255;Parameter, Δ w table
Show the datum quantity of weight coefficient, span 0~0.2.
After aforesaid operations, Y-component and Vmp component can be carried out proportion weighted fusion, and obtain final brightness dividing
Amount is expressed as Yout, then Yout is deposited into caching operation interval.
Step S3:Calculate the color component images after BLC is processed:
The color component for capturing two field picture is read from caching operation interval, and color component is comprising U component image and V
Component image.Due to carrying out backlight process to luminance Y component, i.e., the dark of image reversible-light shooting is have adjusted using GTG mapping curve
Area, improves the visual effect of dark space, also causes the color saturation of image and bright-colored degree to change.So needing
Color component is revised in conjunction with the backlight degree for the treatment of of luminance Y component.
Here, to color component:The backlight processing procedure of U component image and V component image is as follows:
Ctrl_uv represents the control intensity of color component, as follows according to brightness degree Ygrade value:
During current brightness level coefficient Ygrade value 0 or 1, ctrl_uv span is 1.0~1.2;
During current brightness level coefficient Ygrade value 2 or 3, ctrl_uv value 1.0.
The backlight of U component and V component is processed output result Uout component and Vout component is deposited into caching operation interval.
Step S4:Yout component, Uout component and Vout component are read from caching operation interval, as final candid photograph frame
Image backlight result is exported.
The present embodiment additionally provides the device for improving backlight scene vehicle snapshot picture quality using said method, with such as
Structure shown in Fig. 4, including image capture module, image BLC processing module and image display, wherein:
Described image acquisition module is used for gathering the yuv format data flow diagram picture of moving target, and is currently captured frame
YUV image data is activation is to image BLC processing module;
Described image BLC processing module, will using the method for above-mentioned raising backlight scene vehicle snapshot picture quality
The current YUV image data for capturing frame carry out BLC process, the enhanced YUV image number of the BLC after acquisition process
According to being then forwarded to image display;
Described image display module is used for the candid photograph image after showing BLC process.
Be a kind of detailed embodiment of the present invention and specific operating process above, be with technical solutions of the utility model
Premised under implemented, but protection domain of the present utility model is not limited to the above embodiments.
Claims (7)
1. a kind of improve backlight scene vehicle snapshot picture quality method, it is characterised in that comprise the following steps:
Step S1:Data after obtaining the initial data for capturing two field picture and processing:
Step S101:The yuv format data flow of pending video image is obtained using image capture module, is currently captured frame
YUV image data, obtain the Y component map picture of YUV image data, U component image and V component image;
Step S102:The YUV image data of step S101 are converted to rgb image data, calculate the R component of rgb image data
Image, G component image, the maximum of B component image, corresponding for maximum component image is labeled as lightness component image, note
For V';
Step S103:Overall average Ymean of the Y component map picture of calculation procedure S101, the overall average according to Y component map picture is drawn
Divide the brightness degree Ygrade of Y component map picture;
Step S104:The Y component map picture of step S101 is smoothed, and filtering result is calculated, obtain Yfilter
Component image;
Step S2:Calculate the luminance component image after BLC is processed:
Step S201:Build brightness mapping relations curve:
Wherein, variable x represents input signal, and x is the numerical value after normalized, and scope is 0~1;Variable y represents mapping
Relation curve output valve, scope is 0~1;Variable ctrl_y is the amplitude control parameter of mapping curve;
Step S202:The Yfilter component image of step S104 is normalized, is mapped using the brightness of step S201
Relation curve calculates output valve and output valve is multiplied by image gray-scale level maximum, is as a result expressed as Ymp;
Step S203:The mapping result of the V' component image of calculation procedure S102, obtains Vmp component image:
Step S204:Y component map picture and Vmp component image are carried out proportion weighted fusion, by after proportion weighted merges
Data are designated as Yout as the final luminance component image for capturing two field picture;
Step S3:Calculate the color component images after BLC is processed:
Backlight process is carried out to U component image and V component image, is obtained result and is expressed as Uout and Vout:
In formula, ctrl_uv represents the control intensity of color component, according to brightness degree coefficient Ygrade value;
Step S4:Export the candid photograph frame YUV image data after BLC is processed:By Yout component image, Uout component
Image and Vout component image are used as the final YUV image output for capturing frame.
2. according to claim 1 a kind of improve backlight scene vehicle snapshot picture quality method, it is characterised in that institute
State in step S103, overall average Ymean of Y component map picture is calculated using the method for point traversal statistics pixel-by-pixel.
3. according to claim 1 a kind of improve backlight scene vehicle snapshot picture quality method, it is characterised in that institute
State in step S104, using two-dimensional convolution wave filter, Y component map picture is smoothed, then from Gaussian filter or average
Wave filter calculates filtering result, obtains Yfilter component image.
4. according to claim 1 a kind of improve backlight scene vehicle snapshot picture quality method, it is characterised in that institute
State in step S201, ctrl_y numerical value is determined according to brightness degree Ygrade, brightness degree coefficient is bigger, and ctrl_y value is got over
Greatly.
5. according to claim 1 a kind of improve backlight scene vehicle snapshot picture quality method, it is characterised in that institute
State in step S204, the weight coefficient of proportion weighted fusion carries out value by the grey-scale range of respective components image, if component
The index variables of image gray-scale level be i, in Y component map picture and Vmp component image, the corresponding weight coefficient of GTG i be Wy (i) and
Wv (i), wherein:
Wy (i)=1-Wv (i)
In formula, threshold value La and threshold value Lb are the control parameters of respective components image gray-scale level Weighted Fusion.Parameter, Δ w represents corresponding point
Spirogram is as the datum quantity of weight coefficient.
6. according to claim 1 a kind of improve backlight scene vehicle snapshot picture quality method, it is characterised in that institute
State in step S3, ctrl_uv numerical value is determined according to brightness degree coefficient Ygrade.
7. a kind of using the device for improving backlight scene vehicle snapshot picture quality as claim 1~6 methods described, including figure
As acquisition module, image BLC processing module and image display, it is characterised in that:
Described image acquisition module is used for gathering the yuv format data flow diagram picture of moving target, and the current YUV for capturing frame is schemed
As data is activation is to image BLC processing module;
Described image BLC processing module, will be current using the method for above-mentioned raising backlight scene vehicle snapshot picture quality
The YUV image data for capturing frame carry out BLC process, acquisition process after the enhanced YUV image data of BLC, then
It is sent to image display;
Described image display module is used for the candid photograph image after showing BLC process.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610962047.5A CN106454014B (en) | 2016-11-04 | 2016-11-04 | A kind of method and device improving backlight scene vehicle snapshot picture quality |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610962047.5A CN106454014B (en) | 2016-11-04 | 2016-11-04 | A kind of method and device improving backlight scene vehicle snapshot picture quality |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106454014A true CN106454014A (en) | 2017-02-22 |
CN106454014B CN106454014B (en) | 2019-03-08 |
Family
ID=58180386
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610962047.5A Active CN106454014B (en) | 2016-11-04 | 2016-11-04 | A kind of method and device improving backlight scene vehicle snapshot picture quality |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106454014B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107895355A (en) * | 2017-11-30 | 2018-04-10 | 天津天地基业科技有限公司 | A kind of mobile detection and picture contrast system for adaptive enhancement and its method |
CN108337448A (en) * | 2018-04-12 | 2018-07-27 | Oppo广东移动通信有限公司 | High-dynamic-range image acquisition method, device, terminal device and storage medium |
WO2019011147A1 (en) * | 2017-07-10 | 2019-01-17 | Oppo广东移动通信有限公司 | Human face region processing method and apparatus in backlight scene |
WO2019148912A1 (en) * | 2018-02-02 | 2019-08-08 | 杭州海康威视数字技术股份有限公司 | Image processing method, apparatus, electronic device and storage medium |
CN110298812A (en) * | 2019-06-25 | 2019-10-01 | 浙江大华技术股份有限公司 | A kind of method and device of image co-registration processing |
CN113792708A (en) * | 2021-11-10 | 2021-12-14 | 湖南高至科技有限公司 | ARM-based remote target clear imaging system and method |
CN114596280A (en) * | 2022-03-08 | 2022-06-07 | 常州市宏发纵横新材料科技股份有限公司 | Method and device for detecting scrap paper in production process of carbon fiber cloth cover |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101340511A (en) * | 2008-08-07 | 2009-01-07 | 中兴通讯股份有限公司 | Adaptive video image enhancing method based on lightness detection |
CN101860763A (en) * | 2009-04-10 | 2010-10-13 | 索尼公司 | Apparatus for processing of video signals and display device |
CN101951457A (en) * | 2009-07-08 | 2011-01-19 | 卡西欧计算机株式会社 | Image processing apparatus and computer-readable medium |
CN102231206A (en) * | 2011-07-14 | 2011-11-02 | 浙江理工大学 | Colorized night vision image brightness enhancement method applicable to automotive assisted driving system |
CN102447815A (en) * | 2010-10-09 | 2012-05-09 | 中兴通讯股份有限公司 | Video image processing method and device |
CN103002291A (en) * | 2012-12-06 | 2013-03-27 | 杭州藏愚科技有限公司 | Camera wide dynamic image enhancement method and device |
CN106023117A (en) * | 2016-06-01 | 2016-10-12 | 哈尔滨工业大学(威海) | Backlighting image restoration method based on non-linear brightness improving model |
-
2016
- 2016-11-04 CN CN201610962047.5A patent/CN106454014B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101340511A (en) * | 2008-08-07 | 2009-01-07 | 中兴通讯股份有限公司 | Adaptive video image enhancing method based on lightness detection |
CN101860763A (en) * | 2009-04-10 | 2010-10-13 | 索尼公司 | Apparatus for processing of video signals and display device |
CN101951457A (en) * | 2009-07-08 | 2011-01-19 | 卡西欧计算机株式会社 | Image processing apparatus and computer-readable medium |
CN102447815A (en) * | 2010-10-09 | 2012-05-09 | 中兴通讯股份有限公司 | Video image processing method and device |
CN102231206A (en) * | 2011-07-14 | 2011-11-02 | 浙江理工大学 | Colorized night vision image brightness enhancement method applicable to automotive assisted driving system |
CN103002291A (en) * | 2012-12-06 | 2013-03-27 | 杭州藏愚科技有限公司 | Camera wide dynamic image enhancement method and device |
CN106023117A (en) * | 2016-06-01 | 2016-10-12 | 哈尔滨工业大学(威海) | Backlighting image restoration method based on non-linear brightness improving model |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019011147A1 (en) * | 2017-07-10 | 2019-01-17 | Oppo广东移动通信有限公司 | Human face region processing method and apparatus in backlight scene |
CN107895355A (en) * | 2017-11-30 | 2018-04-10 | 天津天地基业科技有限公司 | A kind of mobile detection and picture contrast system for adaptive enhancement and its method |
CN107895355B (en) * | 2017-11-30 | 2021-08-20 | 天津天地基业科技有限公司 | Motion detection and image contrast self-adaptive enhancement system and method |
WO2019148912A1 (en) * | 2018-02-02 | 2019-08-08 | 杭州海康威视数字技术股份有限公司 | Image processing method, apparatus, electronic device and storage medium |
CN108337448A (en) * | 2018-04-12 | 2018-07-27 | Oppo广东移动通信有限公司 | High-dynamic-range image acquisition method, device, terminal device and storage medium |
CN108337448B (en) * | 2018-04-12 | 2020-09-25 | Oppo广东移动通信有限公司 | High dynamic range image acquisition method and device, terminal equipment and storage medium |
CN110298812A (en) * | 2019-06-25 | 2019-10-01 | 浙江大华技术股份有限公司 | A kind of method and device of image co-registration processing |
CN113792708A (en) * | 2021-11-10 | 2021-12-14 | 湖南高至科技有限公司 | ARM-based remote target clear imaging system and method |
CN113792708B (en) * | 2021-11-10 | 2022-03-18 | 湖南高至科技有限公司 | ARM-based remote target clear imaging system and method |
CN114596280A (en) * | 2022-03-08 | 2022-06-07 | 常州市宏发纵横新材料科技股份有限公司 | Method and device for detecting scrap paper in production process of carbon fiber cloth cover |
CN114596280B (en) * | 2022-03-08 | 2022-09-09 | 常州市宏发纵横新材料科技股份有限公司 | Method and device for detecting scrap paper in production process of carbon fiber cloth cover |
Also Published As
Publication number | Publication date |
---|---|
CN106454014B (en) | 2019-03-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106454014A (en) | Method and device for improving quality of vehicle image captured in backlighting scene | |
CN109636754B (en) | Extremely-low-illumination image enhancement method based on generation countermeasure network | |
CN102789635B (en) | Image enhancement method and image enhancement device | |
CN110378845B (en) | Image restoration method based on convolutional neural network under extreme conditions | |
CN101626454B (en) | Method for intensifying video visibility | |
CN106897981A (en) | A kind of enhancement method of low-illumination image based on guiding filtering | |
CN103177424A (en) | Low-luminance image reinforcing and denoising method | |
CN103034986A (en) | Night vision image enhancement method based on exposure fusion | |
CN107862672B (en) | Image defogging method and device | |
CN107240081A (en) | The denoising of night scene image and enhancing processing method | |
CN107578386A (en) | A kind of optimization defogging processing method of unmanned plane shooting image | |
CN106412448B (en) | Wide dynamic range processing method and system based on single frame image | |
CN104899836A (en) | Foggy image enhancing device and method based on near infrared multispectral imaging | |
CN110473158A (en) | A kind of processing method, device and the equipment of license plate image brightness | |
CN113592750B (en) | Infrared enhancement method based on gradient histogram | |
CN110400274A (en) | A kind of vehicle mounted infrared pedestrian detection infrared image enhancing method | |
CN113191986B (en) | Image processing method and device | |
CN111311503A (en) | Night low-brightness image enhancement system | |
CN116823628A (en) | Image processing method and image processing device | |
CN109035181A (en) | A kind of wide dynamic range image processing method based on mean picture brightness | |
CN114418874A (en) | Low-illumination image enhancement method | |
CN110246086A (en) | The image processing method and its image processing system of color image enhancement | |
JP3807266B2 (en) | Image processing device | |
CN112822343A (en) | Night video oriented sharpening method and storage medium | |
Tang et al. | Sky-preserved image dehazing and enhancement for outdoor scenes |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |