CN105868753A - Color identification method and apparatus of blue license plate - Google Patents
Color identification method and apparatus of blue license plate Download PDFInfo
- Publication number
- CN105868753A CN105868753A CN201610207212.6A CN201610207212A CN105868753A CN 105868753 A CN105868753 A CN 105868753A CN 201610207212 A CN201610207212 A CN 201610207212A CN 105868753 A CN105868753 A CN 105868753A
- Authority
- CN
- China
- Prior art keywords
- lighting sources
- infrared
- supplementary lighting
- rgb image
- infrared supplementary
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/141—Control of illumination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
Abstract
The invention discloses a color identification method and apparatus of a blue license plate. The method comprises the following steps: when illumination brightness is lower than specific brightness, starting a supplementary light source, wherein the supplementary light source comprises a near infrared light supplement light source and an infrared light supplement light source; switching a work mode of an infrared ray dual-filter IR-cut of a camera so as to enable invisible light and infrared light to pass at the same time; and simultaneously starting the near infrared light supplement light source and the infrared light supplement light source to identity an license plate color of a vehicle with the blue license plate in an image. According to the invention, the color of the blue license plate can be identified through processing image frames, and thus the problem of incapability of identifying the color of the blue license plate when an infrared light supplement lamp is used in the prior art is solved.
Description
Technical field
The application relates to color recognizing for vehicle id field, particularly relate to a kind of blue car plate color recognition methods and
Device.
Background technology
Along with the development of monitoring technology, carry out road prison based on white light light compensating lamp and video camera on road night
The application of control gets more and more.White light can make car plate well develop the color, and is therefore easy to the identification of car plate color,
But white light compensating lamp is disguised poor, and can disturb the sight line of driver, causes traffic safety hidden danger.
Therefore, the scheme of infrared light compensating lamp+video camera is arisen at the historic moment, and the program solves light pollution and disguise
Problem, presents black and white tone yet with car plate under infrared light compensating lamp, therefore uses infrared light compensating lamp meeting
The problem causing None-identified blueness car plate color.
Summary of the invention
For overcoming problem present in correlation technique, present disclose provides the recognition methods of blue car plate color
And device.
First aspect according to disclosure embodiment, it is provided that the recognition methods of a kind of blue car plate color, should
For video camera, described video camera includes multichannel filter, and right after described multichannel filter filters
Near infrared light and the photosensitive sensor of infrared light;Described multichannel filter is near infrared light printing opacity, it is desirable to
The transmitance of R passage is more than the transmitance of channel B, and multiple passages of described multichannel filter are to infrared light
Light transmittance identical, the recognition methods of described blue car plate color includes:
When illumination brightness is less than given luminance, open light filling source;Described light filling source includes near-infrared light filling
Light source and infrared supplementary lighting sources;
Switch the mode of operation of the double filter IR-cut of infrared ray of described video camera, so that visible ray and infrared
Light passes through simultaneously;
Near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened simultaneously and are identified the car having blue car plate in image
Car plate color.
Optionally, when described video camera is for capturing camera, described near-infrared supplementary lighting sources and infrared light filling
Light source is opened simultaneously, identifies that the car plate color of the vehicle having blue car plate in image includes:
When described target vehicle being detected, based on the near-infrared supplementary lighting sources in described light filling source and infrared
Supplementary lighting sources carries out quick-fried sudden strain of a muscle light filling and captures target vehicle, and the described target vehicle that candid photograph is arrived
Uncorrected data picture frame is converted to RGB image, positions the license plate area in described RGB image, and to car
Board region preserves after carrying out color enhancement;When being not detected by described target vehicle, only open described light filling
Infrared supplementary lighting sources in source.
Optionally, when described video camera is non-candid photograph camera, described near-infrared supplementary lighting sources and infrared benefit
Radiant is opened simultaneously, identifies that the car plate color of the vehicle having blue car plate in image includes:
Described near-infrared supplementary lighting sources and infrared supplementary lighting sources, with the brightest light filling, gather live uncorrected data figure
As frame and be converted to RGB image, position the license plate area in described RGB image, identify described RGB
Image has the car plate color of the vehicle of blue car plate, and RGB image described after process is converted to
YUV image, and carry out live stream transmission and preservation.Optionally, in the described RGB image in described location
License plate area include:
Resolution based on described RGB image sets up corresponding initial calculation matrix;
Maximum in tri-components of RGB of each for described RGB image pixel is deducted minima, and
Judge that whether calculated difference is more than predetermined threshold value;
When described difference more than described predetermined threshold value time, by described initial calculation matrix with described pixel pair
The value of the element answered is set to 1;When described difference is less than or equal to described predetermined threshold value, by described square
The value of the element corresponding with described pixel in Zhen is set to 0, the calculating matrix after being processed;
By three matrixes being made up of the R component of described RGB image, G component and B component respectively with
Described calculating matrix carries out dot product calculating, and positions car plate according to dot product result of calculation.
Optionally, described license plate area is carried out color enhancement, including:
Described RGB image after dot product being calculated is converted to HSV image;
Strengthen the saturation of described HSV image, to complete the color enhancement for described license plate area.
Optionally, described infrared supplementary lighting sources includes the first infrared supplementary lighting sources and the second infrared supplementary lighting sources,
Wherein, the centre wavelength of the centre wavelength of described first infrared supplementary lighting sources and the second infrared supplementary lighting sources is not
With.
Second aspect according to disclosure embodiment, it is provided that the identification device of a kind of blue car plate color, institute
Stating device and be applied to video camera, described video camera includes multichannel filter, and through described multichannel filter
Sensor photosensitive near infrared light and infrared light after optical filtering;Described multichannel filter is saturating near infrared light
Light, it is desirable to the transmitance of R passage is more than the transmitance of channel B, multiple passages of described multichannel filter
Identical to the light transmittance of infrared light, the identification device of described blue car plate color includes:
Opening module, for when illumination brightness is less than given luminance, opening light filling source;Described light filling source
Including near-infrared supplementary lighting sources and infrared supplementary lighting sources;
Handover module, for switching the mode of operation of the double filter IR-cut of the infrared ray of described video camera, with
Make visible ray and infrared light pass through simultaneously;
Identification module, opens in identification image for near-infrared supplementary lighting sources and infrared supplementary lighting sources simultaneously and has
The car plate color of the vehicle of blue car plate.
Optionally, when described video camera is for capturing camera, described identification module includes:
First detection sub-module, for when detecting described target vehicle, based in described light filling source
Near-infrared supplementary lighting sources and infrared supplementary lighting sources carry out quick-fried sudden strain of a muscle light filling and capture target vehicle, and will grab
The uncorrected data picture frame of the described target vehicle photographed is converted to RGB image, positions described RGB image
In license plate area, and carry out license plate area preserving after color enhancement;When being not detected by described target carriage
Time, only open the infrared supplementary lighting sources in described light filling source.
Optionally, when described video camera is non-candid photograph camera, described identification module includes:
Second detection sub-module, for described near-infrared supplementary lighting sources and infrared supplementary lighting sources the brightest same benefit
Light, gathers live uncorrected data picture frame and is converted to RGB image, positioning the car in described RGB image
Board region, identifies the car plate color of the vehicle having blue car plate in described RGB image, and by after process
Described RGB image is converted to YUV image, and carries out live stream transmission and preservation.
Optionally, described first detection sub-module or the second detection sub-module specifically for:
Resolution based on described RGB image sets up corresponding initial calculation matrix;
Maximum in tri-components of RGB of each for described RGB image pixel is deducted minima, and
Judge that whether calculated difference is more than predetermined threshold value;
When described difference more than described predetermined threshold value time, by described initial calculation matrix with described pixel pair
The value of the element answered is set to 1;When described difference is less than or equal to described predetermined threshold value, by described square
The value of the element corresponding with described pixel in Zhen is set to 0, the calculating matrix after being processed;
By three matrixes being made up of the R component of described RGB image, G component and B component respectively with
Described calculating matrix carries out dot product calculating, and positions car plate according to dot product result of calculation.
Optionally, described first detection sub-module specifically for:
Described RGB image after described dot product being calculated is converted to HSV image;
Strengthen the saturation of described HSV image, to complete the color enhancement for described license plate area.
Optionally, described infrared supplementary lighting sources includes the first infrared supplementary lighting sources and the second infrared supplementary lighting sources,
Wherein, the centre wavelength of the centre wavelength of described first infrared supplementary lighting sources and the second infrared supplementary lighting sources is not
With.
As seen from the above-described embodiment, by opening described benefit in the case of being less than given luminance in illumination brightness
Light source and IR-cut is switched to visible ray and mode of operation that infrared luminous energy passes through simultaneously so that described
Video camera identifies in image have blue car plate when near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened simultaneously
The car plate color of vehicle, thus None-identified blueness car when solving to use infrared light compensating lamp in correlation technique
The problem of board color.
It should be appreciated that it is only exemplary and explanatory that above general description and details hereinafter describe
, the disclosure can not be limited.
Accompanying drawing explanation
Accompanying drawing herein is merged in description and constitutes the part of this specification, it is shown that meet the disclosure
Embodiment, and for explaining the principle of the disclosure together with description.
Fig. 1 is the flow chart of the recognition methods of a kind of blue car plate color shown in the disclosure one embodiment.
Fig. 2 is the sensor sensitometric curve to different wave length of a kind of video camera.
Fig. 3 is the hardware block diagram of the device of the identification of blue car plate color in disclosure embodiment.
Fig. 4 is the disclosure identification device according to a kind of blue car plate color shown in an exemplary embodiment
Block diagram.
Detailed description of the invention
Here will illustrate exemplary embodiment in detail, its example represents in the accompanying drawings.Following retouches
Stating when relating to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represents same or analogous key element.
Embodiment described in following exemplary embodiment does not represent all embodiment party consistent with the disclosure
Formula.On the contrary, they only with describe in detail in appended claims, the disclosure some in terms of mutually one
The example of the apparatus and method caused.
In the related, generally use infrared light compensating lamp that video camera is carried out light filling, utilize infrared light not have
Light pollution and disguised strong characteristic, solve to use white light light compensating lamp that video camera is carried out the light that light filling causes
The problem polluted.Then, owing to car plate presents black and white tone under infrared light compensating lamp, infrared benefit is therefore used
The problem that light lantern festival causes None-identified blueness car plate color.
In the disclosed embodiments, can mend by opening in the case of being less than given luminance in illumination brightness
Light source and IR-cut is switched to visible ray and mode of operation that infrared luminous energy passes through simultaneously so that shooting
Machine identifies there is the car of blue car plate in image when near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened simultaneously
Car plate color, thus None-identified blueness car plate face when solving to use infrared light compensating lamp in correlation technique
The problem of color.
See Fig. 1, for the stream of the recognition methods of a kind of blue car plate color shown in the disclosure one embodiment
Cheng Tu, this embodiment is applied to video camera, comprises the following steps:
Step 101: when illumination brightness is less than given luminance, open light filling source;Described light filling source includes
Near-infrared supplementary lighting sources and infrared supplementary lighting sources.
Above-mentioned video camera can include multichannel filter, and near infrared light after multichannel filter filters
The sensor photosensitive with infrared light;Wherein, when selecting sensor for video camera, this sensing of requirement is needed
Device is capable of identify that near-infrared supplementary lighting sources and infrared supplementary lighting sources.Multiple passages of above-mentioned multichannel filter need
Different to the transmitance of near infrared light, wherein, the R passage of multichannel filter is near infrared light
Transmitance is more than the channel B transmitance near infrared light, and the purpose of do so is primarily to as far as possible
Allow R component pass through simultaneously to suppress passing through of B component, and multiple passages of described multichannel filter are to red
The light transmittance of outer light is identical.
During using video camera that road is monitored, when illumination brightness is less than given luminance, can
To open light filling source.Wherein, in this example, light filling source can include near-infrared supplementary lighting sources and infrared benefit
Radiant.In light filling source, introducing near-infrared supplementary lighting sources may insure that and blue car plate color is carried out positive general knowledge
Not;Infrared supplementary lighting sources is introduced in light filling source, can be at the base that blue car plate color is normally identified
On plinth, improve light filling effect and disguise further.Wherein, for same video camera, when light filling source
In near-infrared supplementary lighting sources and during wavelength value in infrared supplementary lighting sources fetching wavelength range, above-mentioned shooting
Machine can obtain preferable picture frame when light filling source light filling, such that it is able to preferably identify the face of blue car plate
Color.The chip used due to different video cameras is different, therefore, when the near-infrared supplementary lighting sources in light filling source
When taking the wavelength value in the wave-length coverage corresponding with above-mentioned video camera with infrared supplementary lighting sources, above-mentioned video camera can
To obtain preferable picture frame.
It should be noted that above-mentioned given luminance can be arranged depending on environment or by User Defined, example
Such as 10LUX (lux), this is not limited by the disclosure.
In this example, for the near-infrared supplementary lighting sources in light filling source and infrared supplementary lighting sources designated wavelength range
Time, in conjunction with the sensor of video camera, the sensitometric curve of different wave length can be determined that in light filling source, near-infrared is mended
The wave-length coverage of radiant.
See sensor that Fig. 2, Fig. 2 the are a kind of video camera sensitometric curve to different wave length.Can in conjunction with Fig. 2
Knowing, wave-length coverage is the near infrared light between 750 nanometers to 770 nanometers, generally insensitive to human eye, right
The interference of driver is less, and wavelength is positioned at the near infrared light in the range of this, and sensor can preferably divide
Distinguish the RGB component in color image, thus easily facilitate video camera in image processing process to blue car plate
Color reduce.
Therefore, in a kind of embodiment illustrated, 750 nanometers to 770 nanometers can be chosen as the reddest
The designated wavelength range of outer supplementary lighting sources.
In this example, when for near-infrared supplementary lighting sources specify wave-length coverage after, in order to improve light filling amount,
Infrared supplementary lighting sources can also be added in light filling source.This infrared supplementary lighting sources can include the of specified wavelength
One infrared supplementary lighting sources and the second infrared supplementary lighting sources, wherein, the specified wavelength of the first infrared supplementary lighting sources with
The specified wavelength of the second infrared supplementary lighting sources is different.
Continuing with seeing Fig. 2, centre wavelength is the infrared light of 830 nanometers, generally insensitive to human eye, to driving
The interference of the person of sailing is less, and the sensor of video camera can be preferably photosensitive to it, so that capture
Picture frame has higher brightness.Therefore, in a kind of embodiment illustrated, can be 830 by centre wavelength
The infrared light of nanometer is as the first HONGGUANG supplementary lighting sources of designated wavelength range.
But, in actual applications, centre wavelength is that the infrared light of 830 nanometers generally has slight red exposure
Phenomenon, the i.e. surrounding of light there will be the HONGGUANG that can be perceived by the human eye.Therefore, in order to improve the hidden of this infrared light
Covering property and further raising light filling effect, may be incorporated into the second infrared light filling of specified wavelength in this example
Light source.
In the embodiment illustrated, the infrared light that centre wavelength is 980 nanometers can be chosen to be
State the second infrared supplementary lighting sources.Centre wavelength is that the infrared light of 980 nanometers is insensitive to human eye, does not has red exposure
Phenomenon, therefore adds the infrared light that centre wavelength is 980 nanometers in infrared supplementary lighting sources, can improve red
The disguise of outer supplementary lighting sources, carries out light filling to whole road scene, reduces car plate and the luminance difference of background,
Car plate is made to be not easy overexposure.The infrared light simultaneously using 830 nanometers and 980 nanometers is received with only use 830
Comparing of rice, under requiring the scene sending brightness, the former is owing to there being the infrared light of 980 nano wave lengths to mix,
Red exposure phenomenon is much smaller.
It should be noted that when realizing, the three of the above light source in light filling source can be placed on same light filling
In unit, it is also possible to separated, it is not particularly limited.
Step 102: switch the mode of operation of the double filter IR-cut of infrared ray of described video camera, so that can
See that light and infrared light pass through simultaneously.
In this example, after opening above-mentioned light filling source, video camera can switch IR-cut (Infrared cut,
The double filter of infrared ray) mode of operation so that visible ray and infrared light pass through simultaneously, wherein, video camera
The filter of IR-cut can be multichannel filter.Here it should be noted that step 101 and step 102
Sequentially interchangeable, is not necessarily to the order by the present embodiment provides and performs.
When the mode of operation that the IR-cut of video camera is switched to visible ray and infrared luminous energy passes through simultaneously, on
The near infrared light stating designated wavelength range can also be by this multichannel filter.In this example, for the ease of
Reduction to blue car plate color, the multiple passages of this multichannel filter near-infrared to designated wavelength range
The transmitance of light can be different.
Continuing with seeing Fig. 2, when the wavelength of near infrared light is between 750 nanometers to 770 nanometers,
The sensor of video camera is different to the photosensitive degree of the RGB component of near infrared light.So that sensor
After photosensitive, for B component and R component, there is different gray values, it is simple to blue car plate color is gone back
Former, select this multichannel filter that the transmitance of the channel B of this near-infrared supplementary lighting sources is less than 15%;Right
The transmitance of the R passage of this near-infrared supplementary lighting sources is more than 80%;Gr to this near-infrared supplementary lighting sources
More than 20% and it is less than 50% with the component transmitance of Gb.
In this example, the infrared light of above-mentioned designated wavelength range can pass through above-mentioned multichannel filter.By scheming
2 understand, and when the wavelength of infrared light is positioned at above-mentioned designated wavelength range, the sensor of video camera is to infrared
The photosensitive degree of the RGB component of light is approximately the same.For the ease of will capture to picture frame in reflecting rate
Being Lycoperdon polymorphum Vitt less than the restore-zone of car plate, therefore the transmitance difference of this infrared light is got over by this multichannel filter
Little the best, it is advisable less than 5% with difference.
Step 103: when there being target vehicle, obtains and includes the picture frame of described target vehicle, and based on
The color of the described picture frame identification blueness car plate got.
When there being target vehicle, video camera can obtain the figure of target vehicle in the case of the light filling of light filling source
As frame, it is then based on the color of this picture frame identification blueness car plate.
In this example, video camera can be to capture camera.When video camera is for capturing camera, can first examine
Survey and whether have target vehicle.When target vehicle being detected, the near-infrared in light filling source can be opened simultaneously
Supplementary lighting sources and infrared supplementary lighting sources, video camera can be based on the near-infrared supplementary lighting sources in light filling source and red
Outer supplementary lighting sources carries out quick-fried sudden strain of a muscle light filling and captures target vehicle, and the figure of target vehicle candid photograph arrived
As frame is converted to RGB image, this RGB image is carried out demosaicing Demosaic and AWB
(Automatic white balance, AWB) processes, owing to this processing procedure is prior art,
Therefore the disclosure does not repeats them here;When being not detected by target vehicle, can only open in light filling source
Infrared supplementary lighting sources.
Capture camera to be typically when there being the target object such as car or people to enter monitoring range, by quick-fried sudden strain of a muscle light filling side
Formula moment captures the picture rich in detail of target object.The bayonet socket camera made a dash across the red light for shooting in such as crossroad.
After uncorrected data picture frame after capturing is converted to RGB image, video camera can position RGB figure
License plate area in Xiang, and license plate area is carried out preservation after color enhancement.
In one embodiment, the process positioned the license plate area in RGB image can be:
Video camera resolution based on RGB image sets up corresponding initial calculation matrix, wherein, initial calculation
The initial value of each element in matrix could be arranged to 0, it is then possible to by each pixel of RGB image
Maximum in tri-components of RGB deducts minima, and judges that whether calculated difference is more than presetting threshold
Value, when difference is more than predetermined threshold value, can be by the value of element corresponding with this pixel in initial calculation matrix
It is set to 1;When difference is less than or equal to predetermined threshold value, can be by corresponding with this pixel in initial calculation matrix
The value of element is set to 0.After the value of pixel each in initial calculation matrix is provided with, after can be processed
Calculating matrix.Then three squares can being made up of the R component of RGB image, G component and B component
Battle array carries out dot product calculating with above-mentioned calculating matrix respectively, and positions car plate according to dot product result of calculation.
In the embodiment illustrated, it is assumed that the resolution of RGB image is 2*2, its each pixel pair
The value of R component, G component and the B component answered can be as shown in table 1:
The position of RGB image | R component value | G component value | B component value |
{1,1} | 2 | 10 | 6 |
{1,2} | 1 | 1 | 1 |
{2,1} | 20 | 10 | 12 |
{2,2} | 2 | 2 | 12 |
Table 1
Now, video camera resolution based on RGB image as shown in table 1 can set up a 2*2
Initial calculation matrix mask, wherein, the initial value of each element in initial calculation matrix mask is permissible
It is 0.It is then possible to the maximum in tri-components of RGB of each pixel of RGB image is deducted
Minima.
The maximum of tri-components of RGB of each pixel of RGB image as shown in table 1 is deducted minimum
After value, difference as shown in table 2 can be obtained:
Table 2
It is then possible to judge whether calculated difference is more than predetermined threshold value, when difference is more than predetermined threshold value
Time, the value of element corresponding with this pixel in initial calculation matrix mask can be set to 1;When difference is less than
Or during equal to predetermined threshold value, the value of element corresponding with this pixel in initial calculation matrix mask can be set to
0。
Assume that predetermined threshold value is 8, then by after in initial calculation matrix mask, the value of each element is provided with,
Can obtain calculating matrix mask as shown in table 3:
The position of Mask matrix | Element value |
{1,1} | 1 |
{1,2} | 0 |
{2,1} | 1 |
{2,2} | 1 |
Table 3
It is then possible to three the matrixes difference being made up of the R component of RGB image, G component and B component
Dot product calculating is carried out with above-mentioned matrix.
As shown in Table 1, the R component of RGB image may be constructed matrix as shown in table 4, RGB image
G component may be constructed matrix as shown in table 5, and the B component of RGB image may be constructed as shown in table 6
Matrix:
The position of RGB image | R component value |
{1,1} | 2 |
{1,2} | 1 |
{2,1} | 20 |
{2,2} | 2 |
Table 4
The position of RGB image | G component value |
{1,1} | 10 |
{1,2} | 1 |
{2,1} | 10 |
{2,2} | 2 |
Table 5
The position of RGB image | B component value |
{1,1} | 6 |
{1,2} | 1 |
{2,1} | 12 |
{2,2} | 12 |
Table 6
To do a little with calculating matrix mask as shown in table 3 such as table 4, table 5 and three matrixes shown in table 6
After taking advantage of calculating, matrix as shown in table 7 can be obtained:
The position of Mask matrix | R component value | G component value | B component value |
{1,1} | 2 | 10 | 6 |
{1,2} | 0 | 0 | 0 |
{2,1} | 20 | 10 | 12 |
{2,2} | 2 | 2 | 12 |
Table 7
The RGB image that the value of R component, G component and B component that each pixel is corresponding is as shown in table 1 passes through
After above-mentioned process, the value of R component, G component and B component that its each pixel is corresponding can be as shown in table 7
Time, the license plate area of RGB image can show that slight blueness, non-license plate area can be shown as black.
At this point it is possible to car plate is positioned according to RGB image as shown in table 7.After completing the location to car plate,
The car plate color in RGB image after can calculating dot product is strengthened.
After completing the location to car plate, the car plate color in RGB image after can calculating dot product adds
By force, in one embodiment, the process that license plate area carries out color reinforcement can be:
RGB image after first dot product being calculated is converted to HSV image, then, strengthens this HSV image
Saturation, to complete the color enhancement for license plate area.Wherein it is possible to by by the S in HSV image
Component value is set to 1 to strengthen the saturation of this HSV image.
In the embodiment illustrated, the process that color is strengthened is specifically as follows:
First by dot product calculate after RGB image be converted to HSV image, by R component corresponding for each pixel,
After the value of G component and B component RGB image as shown in table 7 is converted to HSV image, the HSV after conversion
The value of H component, S component and V component that each pixel of image is corresponding can be as shown in table 8:
The position of Mask matrix | H component value | S component value | V component value |
{1,1} | 150 | 0.8 | 10 |
{1,2} | 0 | 0 | 0 |
{2,1} | 348 | 0.5 | 20 |
{2,2} | 240 | 0.83 | 12 |
Table 8
It should be noted that the process that RGB image is converted to HSV image is prior art, the disclosure
Do not repeat them here.
After RGB image is converted to HSV image, the S component in HSV image can be set to 1, when
When the S component of above-mentioned HSV image is set to 1, H component corresponding to each pixel, S component can be obtained
The HSV image as shown in table 9 with the value of V component:
The position of Mask matrix | H component value | S component value | V component value |
{1,1} | 150 | 1 | 10 |
{1,2} | 0 | 1 | 0 |
{2,1} | 348 | 1 | 20 |
{2,2} | 240 | 1 | 12 |
Table 9
After having been revised by S component in image, amended HSV image can be converted to RGB figure
Picture.HSV image as shown in table 9 for the value of H component corresponding for each pixel, S component and V component is turned
After turning to RGB image, R component, G component and the B that each pixel of the RGB image after conversion is corresponding
The value of component can be as shown in table 10:
The position of Mask matrix | R component value | G component value | B component value |
{1,1} | 0 | 10 | 5 |
{1,2} | 0 | 0 | 0 |
{2,1} | 20 | 0 | 4 |
{2,2} | 0 | 0 | 12 |
Table 10
Understanding in conjunction with table 7 and table 10, after color reinforcement processes, the B component value of RGB image is relative
Be increased in R component value or G component value, i.e. blue portion in RGB image is enhanced.
It should be noted that the process that HSV image is converted to RGB image is prior art, the disclosure
Do not repeat them here.
After above-mentioned process, the color of the license plate area in RGB image is enhanced, and can show the most clear
Clear blueness.
In this example, video camera can also be non-candid photograph camera.When video camera is non-candid photograph camera, can
To carry out normal bright light filling collection live image frame based on light filling source, at the live image frame that then will collect
Reason is RGB image, and based on yuv format, this RGB image is carried out live stream transmission.
Here non-candid photograph camera, refers mainly to public security system and is used in the shooting monitoring record of the region such as community, crossing
The video camera of picture.
Wherein it is desired to explanation, when carrying out live stream transmission based on yuv format, can be simultaneously
Retain the luminance component Y in image and color difference components UV, so that blue car plate in the image of transmission
Part can demonstrate that blueness, non-car plate part cannot demonstrate colour.
In one embodiment, the vehicle in the live image frame of non-candid photograph collected by camera is actual can be blue
Color vehicle, owing to the body portion of vehicle does not equally smear more reflectorized material just like car plate part,
Car plate part is typically due to through special handling, and reflecting rate can reach more than 90%, therefore, by this reality
Condition picture frame is in the image that above-mentioned process obtains, and it is the most clear that only blue car plate part can show
Blueness.
In the disclosed embodiments, can mend by opening in the case of being less than given luminance in illumination brightness
Light source and IR-cut is switched to visible ray and mode of operation that infrared luminous energy passes through simultaneously so that shooting
Machine identifies there is the car of blue car plate in image when near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened simultaneously
Car plate color, thus None-identified blueness car plate face when solving to use infrared light compensating lamp in correlation technique
The problem of color.
Based on the inventive concept same with said method, the embodiment of the present invention additionally provides blue car plate color
Identify device embodiment, the identification device of this blueness car plate color can apply on video camera.Its
In, the identification device of this blueness car plate color can be realized by software, it is also possible to by hardware or soft
The mode of combination of hardware realizes.As a example by implemented in software, as the device on a logical meaning, it is logical
Cross the processor at its place, computer program instructions corresponding in nonvolatile memory is read internal memory
Middle operation is formed.For application, as it is shown on figure 3, be blue car plate in disclosure embodiment
The hardware block diagram of the device of the identification of color, except the processor shown in Fig. 3, network interface, interior
Deposit and outside nonvolatile memory, it is also possible to include other hardware, such as camera head, be responsible for processing report
The forwarding chip etc. of literary composition.
As shown in Figure 4, Fig. 4 is that the disclosure is according to a kind of blue car plate face shown in an exemplary embodiment
The block diagram identifying device of color, described device includes: opening module 410, handover module 420 and knowledge
Other module 430.
Wherein, opening module 410, for when illumination brightness is less than given luminance, opening light filling source;
Described light filling source includes near-infrared supplementary lighting sources and infrared supplementary lighting sources;
Handover module 420, for switching the mode of operation of the double filter IR-cut of the infrared ray of described video camera,
So that visible ray and infrared light pass through simultaneously;
Identification module 430, opens for near-infrared supplementary lighting sources and infrared supplementary lighting sources simultaneously, identifies image
In have the car plate color of vehicle of blue car plate.
In an optional implementation, when described video camera is for capturing camera, described identification module
430 can include (not shown in Fig. 4):
First detection sub-module 431, for when detecting described target vehicle, based on described light filling source
In near-infrared supplementary lighting sources and infrared supplementary lighting sources carry out quick-fried sudden strain of a muscle light filling and target vehicle captured, and
By candid photograph to the uncorrected data picture frame of described target vehicle be converted to RGB image, position described RGB
License plate area in image, and license plate area is carried out preservation after color enhancement;When being not detected by described mesh
During mark vehicle, only open the infrared supplementary lighting sources in described light filling source.
In another optional implementation, when described video camera is non-candid photograph camera, described identification
Module 430 can include (not shown in Fig. 4):
Second detection sub-module 432, same often for described near-infrared supplementary lighting sources and infrared supplementary lighting sources
Bright light filling, gathers live uncorrected data picture frame and is converted to RGB image, positioning in described RGB image
License plate area, identify the car plate color of the vehicle having blue car plate in described RGB image, and will at
After reason, described RGB image is converted to YUV image, and carries out live stream transmission and preservation.
In another optional implementation, described detection sub-module 431 or 432 can be specifically for:
Resolution based on described RGB image sets up corresponding initial calculation matrix;
Maximum in tri-components of RGB of each for described RGB image pixel is deducted minima, and
Judge that whether calculated difference is more than predetermined threshold value;
When described difference more than described predetermined threshold value time, by described initial calculation matrix with described pixel pair
The value of the element answered is set to 1;When described difference is less than or equal to described predetermined threshold value, by described square
The value of the element corresponding with described pixel in Zhen is set to 0, the calculating matrix after being processed;
By three matrixes being made up of the R component of described RGB image, G component and B component respectively with
Described calculating matrix carries out dot product calculating, and positions car plate according to dot product result of calculation.
In another optional implementation, described first detection sub-module 431 can be specifically for:
Described RGB image after described dot product being calculated is converted to HSV image;
Strengthen the saturation of described HSV image, to complete the color enhancement for described license plate area.
In another optional implementation, described infrared supplementary lighting sources includes the first infrared supplementary lighting sources
With the second infrared supplementary lighting sources, wherein, the centre wavelength of described first infrared supplementary lighting sources is infrared with second
The centre wavelength of supplementary lighting sources is different.
As seen from the above-described embodiment, can mend by opening in the case of being less than given luminance in illumination brightness
Light source and IR-cut is switched to visible ray and mode of operation that infrared luminous energy passes through simultaneously so that shooting
Machine identifies there is the car of blue car plate in image when near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened simultaneously
Car plate color, thus None-identified blueness car plate face when solving to use infrared light compensating lamp in correlation technique
The problem of color.
Those skilled in the art, after considering description and putting into practice invention disclosed herein, will readily occur to this
Other embodiment disclosed.The disclosure is intended to any modification, purposes or the adaptability of the disclosure
Change, these modification, purposes or adaptations are followed the general principle of the disclosure and include these public affairs
Open undocumented common knowledge in the art or conventional techniques means.Description and embodiments only by
Being considered as exemplary, the true scope of the disclosure and spirit are pointed out by claim below.
It should be appreciated that the disclosure is not limited to accurate knot described above and illustrated in the accompanying drawings
Structure, and various modifications and changes can carried out without departing from the scope.The scope of the present disclosure is only by appended
Claim limits.
The foregoing is only the preferred embodiment of the disclosure, not in order to limit the disclosure, all at this
Within disclosed spirit and principle, any modification, equivalent substitution and improvement etc. done, should be included in
Within the scope of disclosure protection.
Claims (12)
1. a recognition methods for blue car plate color, is applied to video camera, and described video camera includes many
Passage filter, and sensor photosensitive near infrared light and infrared light after described multichannel filter filters;
It is characterized in that, described multichannel filter is near infrared light printing opacity, it is desirable to the transmitance of R passage is more than
The transmitance of channel B, multiple passages of described multichannel filter are identical to the light transmittance of infrared light, described
The recognition methods of blue car plate color includes:
When illumination brightness is less than given luminance, open light filling source;Described light filling source includes near-infrared light filling
Light source and infrared supplementary lighting sources;
Switch the mode of operation of the double filter IR-cut of infrared ray of described video camera, so that visible ray and infrared
Light passes through simultaneously;
Near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened simultaneously and are identified the car having blue car plate in image
Car plate color.
Method the most according to claim 1, it is characterised in that when described video camera is for capturing phase
During machine, described near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened in identification image simultaneously blue car plate
The car plate color of vehicle include:
When described target vehicle being detected, based on the near-infrared supplementary lighting sources in described light filling source and infrared
Supplementary lighting sources carries out quick-fried sudden strain of a muscle light filling and captures target vehicle, and the described target vehicle that candid photograph is arrived
Uncorrected data picture frame is converted to RGB image, positions the license plate area in described RGB image, and to car
Board region preserves after carrying out color enhancement;When being not detected by described target vehicle, only open described light filling
Infrared supplementary lighting sources in source.
Method the most according to claim 1, it is characterised in that when described video camera is non-candid photograph
During camera, described near-infrared supplementary lighting sources and infrared supplementary lighting sources are opened simultaneously, identify in image have blue car
The car plate color of the vehicle of board includes:
Described near-infrared supplementary lighting sources and infrared supplementary lighting sources, with the brightest light filling, gather live uncorrected data figure
As frame and be converted to RGB image, position the license plate area in described RGB image, identify described RGB
Image has the car plate color of the vehicle of blue car plate, and RGB image described after process is converted to
YUV image, and carry out live stream transmission and preservation.
The most according to the method in claim 2 or 3, it is characterised in that the described RGB in described location
License plate area in image includes:
Resolution based on described RGB image sets up corresponding initial calculation matrix;
Maximum in tri-components of RGB of each for described RGB image pixel is deducted minima, and
Judge that whether calculated difference is more than predetermined threshold value;
When described difference more than described predetermined threshold value time, by described initial calculation matrix with described pixel pair
The value of the element answered is set to 1;When described difference is less than or equal to described predetermined threshold value, by described square
The value of the element corresponding with described pixel in Zhen is set to 0, the calculating matrix after being processed;
By three matrixes being made up of the R component of described RGB image, G component and B component respectively with
Described calculating matrix carries out dot product calculating, and positions car plate according to dot product result of calculation.
Method the most according to claim 4, it is characterised in that described license plate area is carried out color
Color enhancing, including:
Described RGB image after dot product being calculated is converted to HSV image;
Strengthen the saturation of described HSV image, to complete the color enhancement for described license plate area.
Method the most according to claim 1, it is characterised in that described infrared supplementary lighting sources includes
First infrared supplementary lighting sources and the second infrared supplementary lighting sources, wherein, in described first infrared supplementary lighting sources
Cardiac wave length is different from the centre wavelength of the second infrared supplementary lighting sources.
7. an identification device for blue car plate color, is applied to video camera, and described video camera includes many
Passage filter, and sensor photosensitive near infrared light and infrared light after described multichannel filter filters;
It is characterized in that, described multichannel filter is near infrared light printing opacity, it is desirable to the transmitance of R passage is more than
The transmitance of channel B, multiple passages of described multichannel filter are identical to the light transmittance of infrared light, described
The identification device of blue car plate color includes:
Opening module, for when illumination brightness is less than given luminance, opening light filling source;Described light filling source
Including near-infrared supplementary lighting sources and infrared supplementary lighting sources;
Handover module, for switching the mode of operation of the double filter IR-cut of the infrared ray of described video camera, with
Make visible ray and infrared light pass through simultaneously;
Identification module, opens in identification image for near-infrared supplementary lighting sources and infrared supplementary lighting sources simultaneously and has
The car plate color of the vehicle of blue car plate.
Device the most according to claim 7, it is characterised in that when described video camera is for capturing phase
During machine, described identification module includes:
First detection sub-module, for when detecting described target vehicle, based in described light filling source
Near-infrared supplementary lighting sources and infrared supplementary lighting sources carry out quick-fried sudden strain of a muscle light filling and capture target vehicle, and will grab
The uncorrected data picture frame of the described target vehicle photographed is converted to RGB image, positions described RGB image
In license plate area, and carry out license plate area preserving after color enhancement;When being not detected by described target carriage
Time, only open the infrared supplementary lighting sources in described light filling source.
Device the most according to claim 7, it is characterised in that when described video camera is non-candid photograph
During camera, described identification module includes:
Second detection sub-module, for described near-infrared supplementary lighting sources and infrared supplementary lighting sources the brightest same benefit
Light, gathers live uncorrected data picture frame and is converted to RGB image, positioning the car in described RGB image
Board region, identifies the car plate color of the vehicle having blue car plate in described RGB image, and by after process
Described RGB image is converted to YUV image, and carries out live stream transmission and preservation.
Device the most according to claim 8 or claim 9, it is characterised in that described detection sub-module has
Body is used for:
Resolution based on described RGB image sets up corresponding initial calculation matrix;
Maximum in tri-components of RGB of each for described RGB image pixel is deducted minima, and
Judge that whether calculated difference is more than predetermined threshold value;
When described difference more than described predetermined threshold value time, by described initial calculation matrix with described pixel pair
The value of the element answered is set to 1;When described difference is less than or equal to described predetermined threshold value, by described square
The value of the element corresponding with described pixel in Zhen is set to 0, the calculating matrix after being processed;
By three matrixes being made up of the R component of described RGB image, G component and B component respectively with
Described calculating matrix carries out dot product calculating, and positions car plate according to dot product result of calculation.
11. devices according to claim 10, it is characterised in that described first detection sub-module
Specifically for:
Described RGB image after described dot product being calculated is converted to HSV image;
Strengthen the saturation of described HSV image, to complete the color enhancement for described license plate area.
12. devices according to claim 7, it is characterised in that described infrared supplementary lighting sources includes
First infrared supplementary lighting sources and the second infrared supplementary lighting sources, wherein, in described first infrared supplementary lighting sources
Cardiac wave length is different from the centre wavelength of the second infrared supplementary lighting sources.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610207212.6A CN105868753B (en) | 2016-04-05 | 2016-04-05 | The recognition methods of blue license plate color and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610207212.6A CN105868753B (en) | 2016-04-05 | 2016-04-05 | The recognition methods of blue license plate color and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105868753A true CN105868753A (en) | 2016-08-17 |
CN105868753B CN105868753B (en) | 2019-10-18 |
Family
ID=56626971
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610207212.6A Active CN105868753B (en) | 2016-04-05 | 2016-04-05 | The recognition methods of blue license plate color and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105868753B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109389111A (en) * | 2018-10-24 | 2019-02-26 | 浙江宇视科技有限公司 | Image acquisition and processing method, device and computer readable storage medium |
CN109672828A (en) * | 2019-01-04 | 2019-04-23 | 深圳英飞拓科技股份有限公司 | Low-light (level) lifting device and low-light (level) method for improving based on optical path white balance |
CN110378289A (en) * | 2019-07-19 | 2019-10-25 | 王立之 | A kind of the reading identifying system and method for Vehicle Identification Number |
CN110490042A (en) * | 2019-05-31 | 2019-11-22 | 杭州海康威视数字技术股份有限公司 | Face identification device and access control equipment |
CN111489558A (en) * | 2019-11-11 | 2020-08-04 | 深圳市海洋王照明工程有限公司 | License plate recognition method and system |
CN111770281A (en) * | 2018-07-30 | 2020-10-13 | 华为技术有限公司 | Time division multiplexing supplementary lighting imaging device and method |
CN111860539A (en) * | 2020-07-20 | 2020-10-30 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
CN112133101A (en) * | 2020-09-22 | 2020-12-25 | 杭州海康威视数字技术股份有限公司 | Method and device for enhancing license plate area, camera device, computing equipment and storage medium |
CN113498237A (en) * | 2020-03-20 | 2021-10-12 | 杭州海康威视数字技术股份有限公司 | Driving current control method and mixed light supplement control method for multispectral light supplement lamp |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202217288U (en) * | 2011-07-21 | 2012-05-09 | 江南大学 | Time sharing multicolor light complement license plate identification system |
CN202887451U (en) * | 2012-10-26 | 2013-04-17 | 青岛海信网络科技股份有限公司 | Infrared light and visible light filling blended electronic police system |
CN103645594A (en) * | 2013-08-01 | 2014-03-19 | 广东本致数码科技有限公司 | Multispectral infrared surveillance camera supplementary lighting method |
CN104966071A (en) * | 2015-07-03 | 2015-10-07 | 武汉烽火众智数字技术有限责任公司 | Infrared light supplement based night license plate detection and recognition method and apparatus |
-
2016
- 2016-04-05 CN CN201610207212.6A patent/CN105868753B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN202217288U (en) * | 2011-07-21 | 2012-05-09 | 江南大学 | Time sharing multicolor light complement license plate identification system |
CN202887451U (en) * | 2012-10-26 | 2013-04-17 | 青岛海信网络科技股份有限公司 | Infrared light and visible light filling blended electronic police system |
CN103645594A (en) * | 2013-08-01 | 2014-03-19 | 广东本致数码科技有限公司 | Multispectral infrared surveillance camera supplementary lighting method |
CN104966071A (en) * | 2015-07-03 | 2015-10-07 | 武汉烽火众智数字技术有限责任公司 | Infrared light supplement based night license plate detection and recognition method and apparatus |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111770281A (en) * | 2018-07-30 | 2020-10-13 | 华为技术有限公司 | Time division multiplexing supplementary lighting imaging device and method |
CN109389111A (en) * | 2018-10-24 | 2019-02-26 | 浙江宇视科技有限公司 | Image acquisition and processing method, device and computer readable storage medium |
CN109672828A (en) * | 2019-01-04 | 2019-04-23 | 深圳英飞拓科技股份有限公司 | Low-light (level) lifting device and low-light (level) method for improving based on optical path white balance |
CN109672828B (en) * | 2019-01-04 | 2021-09-07 | 深圳英飞拓科技股份有限公司 | Low-illumination lifting device and method based on light path white balance |
CN110490042A (en) * | 2019-05-31 | 2019-11-22 | 杭州海康威视数字技术股份有限公司 | Face identification device and access control equipment |
CN110490042B (en) * | 2019-05-31 | 2022-02-11 | 杭州海康威视数字技术股份有限公司 | Face recognition device and entrance guard's equipment |
CN110378289A (en) * | 2019-07-19 | 2019-10-25 | 王立之 | A kind of the reading identifying system and method for Vehicle Identification Number |
CN110378289B (en) * | 2019-07-19 | 2023-05-16 | 王立之 | Reading and identifying system and method for vehicle identification code |
CN111489558A (en) * | 2019-11-11 | 2020-08-04 | 深圳市海洋王照明工程有限公司 | License plate recognition method and system |
CN113498237A (en) * | 2020-03-20 | 2021-10-12 | 杭州海康威视数字技术股份有限公司 | Driving current control method and mixed light supplement control method for multispectral light supplement lamp |
CN111860539A (en) * | 2020-07-20 | 2020-10-30 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
CN112133101A (en) * | 2020-09-22 | 2020-12-25 | 杭州海康威视数字技术股份有限公司 | Method and device for enhancing license plate area, camera device, computing equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN105868753B (en) | 2019-10-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105868753A (en) | Color identification method and apparatus of blue license plate | |
KR101493064B1 (en) | In-camera based method of detecting defect eye with high accuracy | |
KR101662846B1 (en) | Apparatus and method for generating bokeh in out-of-focus shooting | |
WO2016068560A1 (en) | Translucent mark, method for synthesis and detection of translucent mark, transparent mark, and method for synthesis and detection of transparent mark | |
CN207491128U (en) | A kind of RGB+IR image capture devices | |
JP7255718B2 (en) | Information processing device, recognition support method, and computer program | |
CN104683767A (en) | Fog penetrating image generation method and device | |
CN110490187B (en) | License plate recognition device and method | |
CN107534735A (en) | Image processing method, device and the terminal of terminal | |
CN108712608A (en) | Terminal device image pickup method and device | |
CN108717691B (en) | Image fusion method and device, electronic equipment and medium | |
CN109804619A (en) | Image processing apparatus, image processing method and camera | |
CN106161974A (en) | Utilize display apparatus and the method thereof of HDR function | |
US9894255B2 (en) | Method and system for depth selective segmentation of object | |
CN106534677A (en) | Image overexposure optimization method and device | |
CN106951891A (en) | Spot detection method and apparatus | |
CN107920205B (en) | Image processing method, device, storage medium and electronic equipment | |
US8570433B1 (en) | Coloration artifact reduction | |
CN107948511A (en) | Brightness of image processing method, device, storage medium and electronic equipment | |
CN110674729A (en) | Method for identifying number of people based on heat energy estimation, computer device and computer readable storage medium | |
CN105263008A (en) | Method and device for improving color image quality under low ambient light illumination | |
RU2015118275A (en) | EXTERNAL IMAGE SYSTEM AND / OR AIMING EQUIPMENT FOR MILITARY DRY VEHICLES AND SHIPS OF THE NAVY | |
KR101862043B1 (en) | Multi camera | |
TWI592882B (en) | Method and system for detecting pretended image | |
CN106402717A (en) | AR (augmented reality) play control method and intelligent table lamp |
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 |