CN106503707A - Licence plate recognition method and device under the conditions of a kind of infrared light filling - Google Patents
Licence plate recognition method and device under the conditions of a kind of infrared light filling Download PDFInfo
- Publication number
- CN106503707A CN106503707A CN201610920708.8A CN201610920708A CN106503707A CN 106503707 A CN106503707 A CN 106503707A CN 201610920708 A CN201610920708 A CN 201610920708A CN 106503707 A CN106503707 A CN 106503707A
- Authority
- CN
- China
- Prior art keywords
- image
- license plate
- car
- contrast
- character
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Character Input (AREA)
Abstract
The invention discloses the licence plate recognition method and device under the conditions of a kind of infrared light filling, belong to technical field of image processing, methods described obtain in car image license plate image from pending mistake first, splits license plate image by the concavo-convex line of character, obtains car plate background area therein;Then the color of adjustment car plate background area image is to default color;Finally character recognition, output identification car plate are carried out to adjusting the license plate image after color.Accordingly, the inventive system comprises background segment module, color adjustment module and character recognition module.The method of the present invention and device are retrieved a higher license plate image for meeting character recognition requirement of contrast in car plate position, are improve Car license recognition rate by the color adjustment to car plate background area in license plate image.
Description
Technical field
The invention belongs to a kind of licence plate recognition method under the conditions of technical field of image processing, more particularly to infrared light filling and
Device.
Background technology
In traffic administration, by capture vehicle video image, identify car plate, have become presently the most commonly use
A kind of traffic administration means.However as the propulsion of urban modernization, city for light pollution phenomenon control more and more strict,
In the scheme application of intelligent transportation, the stroboscopic lamp of high brightness, flashing light are gradually limited in down town section and are used.And
Light pollution can be reduced to the limit by the infrared bayonet socket camera of ball-type, and human eye is almost nil to the excitement of infrared light, while ball
Compared to gun shaped camera, advantage in terms of project installation, security monitoring becomes apparent from the structure of type camera, such as reduces external in a large number
Equipment, flexibly changes monitoring visual angle etc..Therefore, in intelligent traffic administration system, start using the infrared bayonet socket phase of ball-type more and more
Machine.
The infrared bayonet socket camera of current ball-type, has and captured the functions such as car, the identification number-plate number, its integrated colored one
Change camera, head, decoder, protective cover etc. multi-functional with one, easy for installation, using simple but powerful, extensively apply
In the monitoring of open area, different occasions can be used.The infrared bayonet socket camera of ball-type is a kind of at a relatively high product of integrated level
Product, are integrated with clouds terrace system, communication system and camera system, and clouds terrace system refers to the rotating part that motor drives, communication system
Refer to the control to motor and to image and the process part of signal, camera system refers to the all-in-one movement of employing.
But the infrared bayonet socket camera of ball-type there is also the problem on image procossing simultaneously, for the engineering of normal intelligent transportation should
With scheme, bar is high six meters, and level captures 20 meters or so of position.This cause often bright infrared light compensating lamp for the light filling effect of car plate
Not good enough, there is contrast wretched insufficiency.And current licence plate recognition method, including license plate image positioning, Character segmentation,
And character recognition, due to contrast wretched insufficiency, in car plate, font accurately can not be extracted, and cause car data Car license recognition
The low phenomenon of rate.
Content of the invention
The present invention proposes the licence plate recognition method and device under the conditions of a kind of infrared light filling, exists for solving prior art
Infrared light compensating lamp is not good enough for the light filling effect of car plate, contrast wretched insufficiency and caused cross car data Car license recognition rate low
Problem.
To achieve these goals, technical solution of the present invention is as follows:
A kind of licence plate recognition method under the conditions of infrared light filling, the licence plate recognition method include:
License plate image obtain in car image from pending mistake, license plate image is split by the concavo-convex line of character, is obtained therein
Car plate background area;
The color of adjustment car plate background area image is to default color;
Character recognition, output identification car plate are carried out to adjusting the license plate image after color.
Further, described after license plate image is obtained from pending car image excessively, by the concavo-convex line partition cart of character
Board image, before obtaining car plate background area therein, methods described also includes:
Calculate the contrast of the license plate image;
When the contrast is less than the threshold value for setting, executes and license plate image is split by the concavo-convex line of character, obtain wherein
Car plate background area.
Further, the contrast for calculating the license plate image, including:
The length and width of license plate image are reduced by setting ratio, the central area of license plate image is obtained, is calculated the center
The contrast in domain, using the contrast of the central area as the contrast of the license plate image.
Method proposed by the invention, also includes:
When the contrast is not less than the threshold value of setting, directly execute carries out character recognition to the license plate image, defeated
Go out to recognize car plate.
Further, in the method for the invention, the width pixel quantity of the license plate image is more than the width for setting
Threshold value.
The invention allows for the license plate recognition device under the conditions of a kind of infrared light filling, the license plate recognition device includes:
Background segment module, for obtaining license plate image from pending mistake in car image, by the concavo-convex line partition cart of character
Board image, obtains car plate background area therein;
Color adjustment module, for adjusting the color of car plate background area image to default color;
Character recognition module, for carrying out character recognition, output identification car plate to adjusting the license plate image after color.
Further, the background segment module is being crossed in car image after acquisition license plate image, by character from pending
Concavo-convex line splits license plate image, before obtaining car plate background area therein, also executes following operation:
Calculate the contrast of the license plate image;
When the contrast is less than the threshold value for setting, executes and license plate image is split by the concavo-convex line of character, obtain wherein
Car plate background area.
Further, the background segment module calculates the contrast of the license plate image, executes following operation:
The length and width of license plate image are reduced by setting ratio, the central area of license plate image is obtained, is calculated the center
The contrast in domain, using the contrast of the central area as the contrast of the license plate image.
Background segment module of the present invention when the contrast is not less than the threshold value of setting, directly by the car plate figure
Character recognition, output identification car plate are carried out as sending to the character recognition module.
Licence plate recognition method and device under the conditions of a kind of infrared light filling proposed by the present invention, by car in license plate image
The color adjustment of board background area, retrieves a higher car plate for meeting character recognition requirement of contrast in car plate position
Image, improves Car license recognition rate.
Description of the drawings
Fig. 1 is the licence plate recognition method flow chart under the conditions of infrared light filling of the invention;
Fig. 2 is that embodiment of the present invention car plate background color adjusts schematic diagram;
Fig. 3 is the license plate recognition device structural representation under the conditions of infrared light filling of the invention.
Specific embodiment
With reference to the accompanying drawings and examples technical solution of the present invention is described in further details, following examples are not constituted
Limitation of the invention.
Dividend outer bayonet camera in prior art bottom is (as using the normal ball-type bayonet socket camera of bright light filling or other are infrared of low-light
The bayonet socket camera of light filling) gather night picture when, often there is the too low and caused car plate sign character of car plate contrast cannot
Extract the situation of identification, it is impossible to meet normal use.
The infrared bayonet socket camera of ball-type uses integrated low-light often bright light filling, after vehicle reaches precalculated position, ball-type phase chance
Car picture was gathered immediately, and first-selection is positioned to car plate by texture, color, length-width ratio etc., then entered the car plate of positioning
Line character is extracted, and finally the character to extracting is identified.For same style equipment is in white light light filling and infrared light filling
Under video image, the car plate difference on effect are presented by two kinds of light filling modes greatly, in the case of infrared light filling contrast compared with
Low, and contrast is higher in the case of white light light filling.
Due to the drastically decline of night-environment light, by the picture gathered by the bayonet socket camera of the normal bright light filling of infrared low-light
Characters on license plate is not clear with car plate background color contrast difference aobvious, it is impossible to accomplish that normal character separates and identification.Although it is but infrared
In the video image obtained under light filling, car plate is extremely low with character contrast, but still has in the middle of character and car plate background significantly
The profile for producing light refraction because of concavo-convex is present, and the present embodiment enters to car plate background color according to contours segmentation video image
After row is processed, then Car license recognition is carried out, to reach the effect for improving Car license recognition rate.
As shown in figure 1, the licence plate recognition method under the conditions of a kind of infrared light filling, including:
Step S1, from pending cross car image in obtain license plate image, calculate license plate image contrast.
Bayonet socket camera can shoot the video image of car, in the present embodiment when there is vehicle to pass through according to trigger condition
Car image is referred to as crossed.For car image is crossed, Car license recognition is further carried out, export the number-plate number of vehicle, for intelligent transportation
Management.
The process of Car license recognition, first has to position car plate, identifies the corresponding license plate image in car plate position, so as to
In subsequently image procossing is carried out, character therein is identified.With regard to License Plate, current method includes the car plate based on edge
Positioning, based on the License Plate of color, License Plate based on machine learning etc., repeat no more here.
This step is that the license plate image obtained by License Plate is processed, and calculates the contrast of license plate image.
Contrast refer to most bright white to light and shade region in piece image and most dark black between different brightness levels measurement, this reality
It is that such as one car plate brightness in full white screen state is 500cd/m2 for whole license plate image to apply contrast in example,
Completely black screen state luminance is 0.5cd/m2, and the contrast of such car plate is exactly 1000:1.
Impact of the contrast to visual effect is very crucial, and in general contrast is bigger, and image is more clear eye-catching, color
Also more distinct gorgeous;And contrast is little, then whole picture can be allowed all dusky.The contrast of the license plate image by calculating
Degree, it can be determined that cross whether car image is shooting at night, for the specific formula for calculation of contrast, is repeated no more here.
It should be noted that generally according to empirical value, a threshold value can be set, when contrast is not less than the threshold value,
Think that license plate image is easier to carry out character recognition, and for contrast be less than the threshold value when, it is believed that license plate image is at night
Between shoot, it is not easy to carry out character recognition, after needing to be processed by the method subsequent step of the present embodiment recognize.
Step S2, for contrast is less than the license plate image of the threshold value for setting, license plate image is split by the concavo-convex line of character,
Obtain car plate background area therein.
By the calculating of preceding step contrast, by simple threshold decision, it is possible to judge that whether license plate image is
Nighttime image, i.e., for contrast is less than the license plate image of the threshold value for setting, it is believed that be nighttime image, need to be processed, no
Then think that license plate image meets the contrast for correctly carrying out character recognition, directly carry out character recognition.I.e. for contrast
It is not less than the license plate image of the threshold value of setting, it is not necessary to through the process of step S2 and step S3, directly execute to license plate image
Carry out character recognition, output identification car plate.
Although in the video image obtained under infrared light filling, car plate is extremely low with character contrast, in character and car plate background
Centre is still with the presence of the profile for significantly producing because of concavo-convex light refraction.And the identification for profile, can pass through
In binary image, the grey scale change of pixel is being identified, it is also possible to be identified by other common methods.
For example by the concavo-convex line of character and car plate background color because the profile of refraction generation carries out the segmentation of lasso trick algorithm, segmentation
Process adopts widely used Live Wire algorithms, and the algorithm is a kind of interactive edge extracting and image partition method, its work
Be the optimal path that given point-to-point transmission in image is produced using Dynamic Programming scheme as mechanism, reasonably construct cost function and choosing
Starting point and impact point is selected, in order to extract the edge of object.The car plate for obtaining is carried out point by using Live Wire algorithms
Cut, obtain the image of car plate background parts.
The present embodiment can also carry out Image outline identification using OpenCV methods, be not discussed here.
Identify character and profile in the middle of car plate background after, it is possible to car plate figure is split by the concavo-convex line of character
Picture, obtains car plate background area therein, will character separated with car plate background area, so as to subsequently to car plate background area
Domain carries out color filling.
Step S3, the color of adjustment car plate background area image are to default color.
Car plate background color includes blueness, yellow and black, in the majority with blueness, and the present embodiment by taking blueness as an example is preset and filled out
The color for filling car plate background area is blueness.
So as to, after car plate background area is obtained, be adjusted to the color of the image of car plate background area, i.e., to this
Area filling is blue.
Car plate background background color A needed for assuming represents design parameter tone a, saturation degree b, brightness c, red x, green y, indigo plant
Color z, then most background color A is added to the car plate background parts beyond character outline at last.Car plate position retrieve one right
The license plate image for meeting character recognition requirement higher than degree.
As shown in Fig. 2 the left side is license plate image of the contrast less than given threshold, the right is to be adjusted by car plate background color
License plate image after whole, it is seen that by adjusting the color of car plate background area image to default color, car plate can be caused
The becoming of image is more prone to recognize.
Step S4, character recognition is carried out to adjusting the license plate image after color, output identification car plate.
For car plate background area is filled with the license plate image of color, the contrast of license plate image is improve, already close to
License plate image under the conditions of daytime or white light light filling, so that can correctly identify car plate according to normal character identifying method
Character in image, substantially increases the accuracy rate of Car license recognition.
It is easily understood that infrared light filling generally can be opened when ambient brightness is less than certain value, therefore the present embodiment
Contrast calculating can also be omitted and car image is crossed with the step of judgement, directly processing after infrared light filling is opened, i.e. acquiescence opening is infrared
Car image of crossing after light filling is only pending image, needs to split license plate image by the concavo-convex line of character, obtains car plate therein
Background area, carries out color adjustment.Therefore for S1 and step S2 the step of the present embodiment, it is also possible to be reduced to directly from treating
Acquisition license plate image in car image was processed, license plate image is split by the concavo-convex line of character, car plate background area therein is obtained,
Equivalent to default treatment is to open the car image excessively for spending car image, i.e. night after infrared light filling.It is preferable that this enforcement
Example execution step S1 and step S2, after being judged by contrast, weed out the car plate that contrast meets character recognition requirement
Image, can adapt to any occasion, and reducing to carry out the car image excessively of subsequent treatment.
The present embodiment step S1 calculates contrast with whole license plate image, increases a kind of preferred scheme below:
Contrast is judged to produce impact in order to exclude license plate frame and anti-theft screw, the present embodiment is identifying car plate position
After corresponding license plate image, also it is defined by the car plate position for navigating to, the length and width of license plate image is reduced by setting ratio
50% central area for obtaining license plate image, calculates the contrast of central area, and the contrast using the central area is used as car plate
The contrast of image, is compared with default contrast threshold.It is easily understood that concrete reduced scale can be adjusted,
Adjusting range relative to the original size of car plate 40%~70%, can obtain reasonable effect.Can also rule of thumb,
Centered on the central point of license plate image, take the rectangle or square area of setting range to calculate contrast, as car plate
The contrast of image, is repeated no more here.If the contrast of license plate image is directly entered line character and is carried more than the threshold value for setting
Take and identification, because even under infrared light filling, with person who is not a member of any political party's car because of its black font, contrast still meets knowledge to yellow card car
Not, it is not necessary to carry out the color adjustment of car plate background.If license plate image contrast is needed to car plate less than the threshold value for setting
Post-processed, by character and the profile of the concavo-convex phenomenon generation of car plate background, split car plate, by the car plate back of the body of font periphery
Scape resets color.
It should be noted that the car image of crossing in order to ensure meets subsequent treatment condition, certain in light exposure parameter
In the case of, it is necessary to assure car plate width pixel reaches certain requirement, is greater than the width threshold value for setting.The embodiment of the present invention is
Calculated with the sample of car plate width pixel 120, it is therefore desirable to note in actual applications capturing position, capture position
Lane width should be identical with picture width consistent, and so as to obtain suitable car plate width pixel quantity, such as pixel is unsatisfactory for
It is easily caused car plate character contour not clear enough, affects next step to process.
With said method accordingly, as shown in figure 3, license plate recognition device under the conditions of a kind of infrared light filling of the present embodiment,
Including:
Background segment module, for obtaining license plate image from pending mistake in car image, by the concavo-convex line partition cart of character
Board image, obtains car plate background area therein;
Color adjustment module, for adjusting the color of car plate background area image to default color;
Character recognition module, for carrying out character recognition, output identification car plate to adjusting the license plate image after color.
The present embodiment background segment module is being crossed in car image after acquisition license plate image, by the concavo-convex line of character from pending
Segmentation license plate image, before obtaining car plate background area therein, also executes following operation:
Calculate the contrast of the license plate image;
When the contrast is less than the threshold value for setting, executes and license plate image is split by the concavo-convex line of character, obtain wherein
Car plate background area.
It is easily understood that if the car image of crossing after the infrared light filling of acquiescence opening is only pending image, unification is all needed
Color adjustment is carried out, then the threshold value of setting need not be below to calculate contrast, i.e. acquiescence contrast.
The present embodiment background segment module calculates the contrast of the license plate image, executes following operation:
The length and width of license plate image are reduced by setting ratio, the central area of license plate image is obtained, is calculated the center
The contrast in domain, using the contrast of the central area as the contrast of the license plate image.
It is easily understood that contrast can also be calculated with view picture license plate image, zoning is reduced, is advantageously reduced
Amount of calculation.
With said method accordingly, background segment module is when the contrast is not less than the threshold value of setting, directly by institute
Stating license plate image and sending to the character recognition module carries out character recognition, output identification car plate.
Above example only in order to technical scheme to be described rather than be limited, without departing substantially from essence of the invention
In the case of god and its essence, those of ordinary skill in the art work as and can make various corresponding changes and change according to the present invention
Shape, but these corresponding changes and deformation should all belong to the protection domain of appended claims of the invention.
Claims (10)
1. the licence plate recognition method under the conditions of a kind of infrared light filling, it is characterised in that the licence plate recognition method includes:
License plate image obtain in car image from pending mistake, license plate image is split by the concavo-convex line of character, car plate therein is obtained
Background area;
The color of adjustment car plate background area image is to default color;
Character recognition, output identification car plate are carried out to adjusting the license plate image after color.
2. the licence plate recognition method under the conditions of infrared light filling according to claim 1, it is characterised in that described from from treating
Managed after license plate image being obtained in car image, license plate image is split by the concavo-convex line of character, car plate background area therein is obtained
Before, methods described also includes:
Calculate the contrast of the license plate image;
When the contrast is less than the threshold value for setting, executes and license plate image is split by the concavo-convex line of character, obtain car therein
Board background area.
3. the licence plate recognition method under the conditions of infrared light filling according to claim 2, it is characterised in that described in the calculating
The contrast of license plate image, including:
The length and width of license plate image are reduced by setting ratio, the central area of license plate image is obtained, is calculated the central area
Contrast, using the contrast of the central area as the contrast of the license plate image.
4. the licence plate recognition method under the conditions of infrared light filling according to claim 2, it is characterised in that methods described is also wrapped
Include:
When the contrast is not less than the threshold value of setting, directly execute carries out character recognition to the license plate image, and output is known
Other car plate.
5. the licence plate recognition method under the conditions of infrared light filling according to claim 1, it is characterised in that the license plate image
Width pixel quantity more than set width threshold value.
6. the license plate recognition device under the conditions of a kind of infrared light filling, it is characterised in that the license plate recognition device includes:
Background segment module, for obtain in car image license plate image from pending mistake, splits car plate figure by the concavo-convex line of character
Picture, obtains car plate background area therein;
Color adjustment module, for adjusting the color of car plate background area image to default color;
Character recognition module, for carrying out character recognition, output identification car plate to adjusting the license plate image after color.
7. the license plate recognition device under the conditions of infrared light filling according to claim 6, it is characterised in that the background segment
Module is being crossed in car image after acquisition license plate image from pending, is split license plate image by the concavo-convex line of character, is obtained therein
Before car plate background area, following operation is also executed:
Calculate the contrast of the license plate image;
When the contrast is less than the threshold value for setting, executes and license plate image is split by the concavo-convex line of character, obtain car therein
Board background area.
8. the license plate recognition device under the conditions of infrared light filling according to claim 7, it is characterised in that the background segment
Module calculates the contrast of the license plate image, executes following operation:
The length and width of license plate image are reduced by setting ratio, the central area of license plate image is obtained, is calculated the central area
Contrast, using the contrast of the central area as the contrast of the license plate image.
9. the license plate recognition device under the conditions of infrared light filling according to claim 7, it is characterised in that the background segment
When module is not less than the threshold value of setting in the contrast, directly the license plate image is sent to the character recognition module
Line character recognizes that output recognizes car plate.
10. the license plate recognition device under the conditions of infrared light filling according to claim 6, it is characterised in that the car plate figure
The width pixel quantity of picture is more than the width threshold value for setting.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610920708.8A CN106503707B (en) | 2016-10-21 | 2016-10-21 | Licence plate recognition method and device under a kind of infrared lighting condition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610920708.8A CN106503707B (en) | 2016-10-21 | 2016-10-21 | Licence plate recognition method and device under a kind of infrared lighting condition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106503707A true CN106503707A (en) | 2017-03-15 |
CN106503707B CN106503707B (en) | 2019-08-02 |
Family
ID=58319430
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610920708.8A Active CN106503707B (en) | 2016-10-21 | 2016-10-21 | Licence plate recognition method and device under a kind of infrared lighting condition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106503707B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108108746A (en) * | 2017-09-13 | 2018-06-01 | 湖南理工学院 | License plate character recognition method based on Caffe deep learning frames |
CN108710857A (en) * | 2018-05-22 | 2018-10-26 | 深圳前海华夏智信数据科技有限公司 | People's vehicle recognition methods based on infrared light filling and device |
CN109785251A (en) * | 2018-12-25 | 2019-05-21 | 北京中科慧眼科技有限公司 | A kind of picture quality method for improving, device and automated driving system |
CN111563402A (en) * | 2019-02-14 | 2020-08-21 | 杭州海康威视数字技术股份有限公司 | License plate recognition method and device, terminal and storage medium |
CN111582268A (en) * | 2020-04-13 | 2020-08-25 | 浙江大华技术股份有限公司 | License plate image processing method and device and computer storage medium |
CN111666788A (en) * | 2019-03-07 | 2020-09-15 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment and storage medium |
CN115050082A (en) * | 2022-08-12 | 2022-09-13 | 深圳市三千米光电科技有限公司 | Light filling monitoring system based on infrared laser lamp |
CN116129157A (en) * | 2023-04-13 | 2023-05-16 | 深圳市夜行人科技有限公司 | Intelligent image processing method and system for warning camera based on extreme low light level |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101901479A (en) * | 2010-07-26 | 2010-12-01 | 公安部交通管理科学研究所 | Method for optimizing license plate image with character color dropout |
CN102147860A (en) * | 2011-05-16 | 2011-08-10 | 杭州华三通信技术有限公司 | License plate recognition method based on white balance and device |
CN102446348A (en) * | 2010-10-15 | 2012-05-09 | 武汉上博科技有限公司 | Method for adjusting and enhancing chromatic color of vehicle snap-shot image |
CN103020930A (en) * | 2012-11-26 | 2013-04-03 | 天津大学 | Nighttime monitoring video enhancing method |
CN103137103A (en) * | 2011-12-05 | 2013-06-05 | 中兴通讯股份有限公司 | Character display method and device |
-
2016
- 2016-10-21 CN CN201610920708.8A patent/CN106503707B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101901479A (en) * | 2010-07-26 | 2010-12-01 | 公安部交通管理科学研究所 | Method for optimizing license plate image with character color dropout |
CN102446348A (en) * | 2010-10-15 | 2012-05-09 | 武汉上博科技有限公司 | Method for adjusting and enhancing chromatic color of vehicle snap-shot image |
CN102147860A (en) * | 2011-05-16 | 2011-08-10 | 杭州华三通信技术有限公司 | License plate recognition method based on white balance and device |
CN103137103A (en) * | 2011-12-05 | 2013-06-05 | 中兴通讯股份有限公司 | Character display method and device |
CN103020930A (en) * | 2012-11-26 | 2013-04-03 | 天津大学 | Nighttime monitoring video enhancing method |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108108746A (en) * | 2017-09-13 | 2018-06-01 | 湖南理工学院 | License plate character recognition method based on Caffe deep learning frames |
CN108108746B (en) * | 2017-09-13 | 2021-04-09 | 湖南理工学院 | License plate character recognition method based on Caffe deep learning framework |
CN108710857A (en) * | 2018-05-22 | 2018-10-26 | 深圳前海华夏智信数据科技有限公司 | People's vehicle recognition methods based on infrared light filling and device |
CN109785251A (en) * | 2018-12-25 | 2019-05-21 | 北京中科慧眼科技有限公司 | A kind of picture quality method for improving, device and automated driving system |
CN111563402A (en) * | 2019-02-14 | 2020-08-21 | 杭州海康威视数字技术股份有限公司 | License plate recognition method and device, terminal and storage medium |
CN111563402B (en) * | 2019-02-14 | 2023-05-09 | 杭州海康威视数字技术股份有限公司 | License plate recognition method, license plate recognition device, terminal and storage medium |
CN111666788A (en) * | 2019-03-07 | 2020-09-15 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment and storage medium |
CN111666788B (en) * | 2019-03-07 | 2024-01-26 | 杭州海康威视数字技术股份有限公司 | Image processing method, device and equipment and storage medium |
CN111582268A (en) * | 2020-04-13 | 2020-08-25 | 浙江大华技术股份有限公司 | License plate image processing method and device and computer storage medium |
CN111582268B (en) * | 2020-04-13 | 2023-05-12 | 浙江大华技术股份有限公司 | License plate image processing method and device and computer storage medium |
CN115050082A (en) * | 2022-08-12 | 2022-09-13 | 深圳市三千米光电科技有限公司 | Light filling monitoring system based on infrared laser lamp |
CN116129157A (en) * | 2023-04-13 | 2023-05-16 | 深圳市夜行人科技有限公司 | Intelligent image processing method and system for warning camera based on extreme low light level |
Also Published As
Publication number | Publication date |
---|---|
CN106503707B (en) | 2019-08-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106503707B (en) | Licence plate recognition method and device under a kind of infrared lighting condition | |
CN103761529B (en) | A kind of naked light detection method and system based on multicolour model and rectangular characteristic | |
CN104766049B (en) | A kind of object color recognition methods and system | |
US10552706B2 (en) | Attachable matter detection apparatus and attachable matter detection method | |
TWI607901B (en) | Image inpainting system area and method using the same | |
CN106991707B (en) | Traffic signal lamp image strengthening method and device based on day and night imaging characteristics | |
KR101688695B1 (en) | Apparatus and method for recognizing the number of cars and a computer-readable recording program for performing the said method, the recording medium | |
WO2020238805A1 (en) | Facial recognition apparatus and door access control device | |
CN110781883A (en) | Parking space size identification system and method based on 360-degree all-round-looking camera | |
US20200151912A1 (en) | Vehicle surroundings recognition apparatus | |
CN110532903A (en) | A kind of method and apparatus of traffic lights image procossing | |
CN105812674A (en) | Signal lamp color correction method, monitoring method, and device thereof | |
CN105809131A (en) | Method and system for carrying out parking space waterlogging detection based on image processing technology | |
JP2012027773A (en) | Pseudo grayscale image generation device and program | |
US20210174143A1 (en) | Information processing method, information processing system, and information processing apparatus | |
CN110557628A (en) | Method and device for detecting shielding of camera and electronic equipment | |
CN104574377A (en) | Method for correcting yellow cast of red lamp of electronic police | |
Filatov et al. | Neural network system of traffic signs recognition | |
Mandal et al. | Real-time automotive night-vision system for drivers to inhibit headlight glare of the oncoming vehicles and enhance road visibility | |
CN112204566A (en) | Image processing method and device based on machine vision | |
CN105430352B (en) | A kind of processing method of video monitoring image | |
TWI767422B (en) | A low-light imaging system | |
CN107415806A (en) | Intelligent warning lamp based on image recognition | |
CN111489558A (en) | License plate recognition method and system | |
CN111428779B (en) | Method, device, equipment and storage medium for determining irradiation range of light supplementing lamp |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20221107 Address after: 15 / F, Zhongchuang Plaza, 385 Hangtian Middle Road, national civil aerospace industrial base, Xi'an City, Shaanxi Province 710100 Patentee after: Xi'an Yu vision Mdt InfoTech Ltd. Address before: 310051 Zhejiang Jiangling Hangzhou Road, Binjiang District, Jiangling, 88, No. 10 South Block 1-11. Patentee before: ZHEJIANG UNIVIEW TECHNOLOGIES Co.,Ltd. |