CN106781521A - The recognition methods of traffic lights and device - Google Patents
The recognition methods of traffic lights and device Download PDFInfo
- Publication number
- CN106781521A CN106781521A CN201611265156.8A CN201611265156A CN106781521A CN 106781521 A CN106781521 A CN 106781521A CN 201611265156 A CN201611265156 A CN 201611265156A CN 106781521 A CN106781521 A CN 106781521A
- Authority
- CN
- China
- Prior art keywords
- area
- graphics template
- image
- matching degree
- gray
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
Recognition methods and device the present disclosure proposes a kind of traffic lights, are related to image identification technical field, this method to include：Obtain the monitored area image of picture pick-up device collection；Obtain the gray level image of monitored area image；Using the graphics template for pre-setting, the target area most matched with graphics template in gray level image is obtained；Graphics template is to be set according to the shape of the indicator lamp on traffic lights base plate configuration and traffic lights base plate；Traffic lights image using the image of target area as traffic lights in the image of monitored area.The disclosure can improve the degree of accuracy of identification traffic lights.
Description
Technical field
This disclosure relates to image identification technical field, more particularly to a kind of traffic lights recognition methods and device.
Background technology
At present, as the recoverable amount of automobile is on the increase, the pressure of road traffic is increasingly weighed.Traffic lights as should
Widest Traffic controller is used, the flow that can effectively relieve traffic congestion, raising road passage capability, reduction traffic accident.Handing over
It is to judge to be whether there is violation vehicle by detecting the state of traffic lights in logical monitoring system, in recent years, in order to
Driver attention can more be caused, traffic lights start using amber lamp base plate and shell, to the identification band of signal lamp
Interference is carried out, has caused to recognize the accuracy of traffic lights.
The content of the invention
The disclosure provides recognition methods and the device of a kind of traffic lights, is used to solve traffic lights using yellow letter
Signal lamp base plate and shell, cause the problem of the accuracy for recognizing traffic lights.
To achieve these goals, according to the first aspect of the embodiment of the present disclosure, there is provided a kind of identification of traffic lights
Method, the method includes：
Obtain the monitored area image of picture pick-up device collection；
Obtain the gray level image of the monitored area image；
Using the graphics template for pre-setting, the target area most matched with the graphics template in the gray level image is obtained
Domain；The graphics template is the shape according to the indicator lamp on traffic lights base plate configuration and the traffic lights base plate
Set；
Traffic lights figure using the image of the target area as traffic lights in the monitored area image
Picture.
Optionally, shape of the graphics template with the traffic lights base plate in the monitored area image and big
It is small identical, and N number of circle corresponding with the position of the N number of indicator lamp on traffic lights base plate is provided with the graphics template
Subregion, it is described using the graphics template for pre-setting, obtain the mesh most matched with the graphics template in the gray level image
Mark region, including：
A. n-th comparison area is selected on the gray level image using the graphics template；
B. obtain in n-th comparison area and be located in the graphics template and outside N number of circular sub-area
Region described in pixel gray value sum, as the first gray value；
C. the N number of area grayscale value for being located at N number of circular sub-area respectively in n-th comparison area is obtained,
And minimum value is selected in N number of area grayscale value as the second gray value, wherein i-th area grayscale of circular sub-area
It is the gray value sum of all pixels being located in n-th comparison area in i-th circular sub-area to be worth, wherein 1
≤i≤N；
D. according to first gray value and second gray value, n-th comparison area and the figure are obtained
First matching degree of template, wherein, n is positive integer, and 1≤n≤M, M are by the monitored area figure using the graphics template
Comparison number of times as needed for traveling through one time；
Step a to step d is performed again after taking n=n+1, until obtaining M the first matching degree；
Using the corresponding comparison area of the first matching degree maximum in M the first matching degrees as in the gray level image with institute
State the target area that graphics template is most matched.
Optionally, it is described according to first gray value and second gray value, obtain n-th comparison area with
First matching degree of the graphics template includes：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is first gray scale
Value, a is second gray value.
Optionally, shape of the graphics template with the traffic lights base plate in the monitored area image and big
It is small identical, and N number of circle corresponding with the position of the N number of indicator lamp on traffic lights base plate is provided with the graphics template
Subregion, it is described using the graphics template for pre-setting, obtain the mesh most matched with the graphics template in the gray level image
Mark region, including：
E. n-th comparison area is selected on the gray level image using the graphics template；
F. obtain and be located in n-th comparison area in each circular sub-area in N number of circular sub-area
The gray scale difference of pixel, obtains N number of gray scale difference；Wherein described i-th gray scale difference be n-th comparison area in be located at i-th
The difference of the average gray value of the outer shroud pixel of the image in circular sub-area and the average gray value of inner ring pixel, wherein 1≤i
≤N；
G. according to N number of gray scale difference, the second matching degree of n-th comparison area and the graphics template is obtained,
Wherein, n is positive integer, and 1≤n≤M, M are for needed for using the graphics template by the monitored area image traversal one time
Compare number of times；
Step e to step g is performed again after taking n=n+1, until obtaining M the second matching degree；
Using the corresponding comparison area of the second matching degree maximum in M the second matching degrees as in the gray level image with institute
State the target area that graphics template is most matched.
Optionally, it is described according to N number of gray scale difference, obtain the of n-th comparison area and the graphics template
Two matching degrees include：
Wherein, YnIt is n-th comparison area and the second matching degree of the graphics template, eiIt is i-th gray scale
Difference.
Optionally, shape of the graphics template with the traffic lights base plate in the monitored area image and big
It is small identical, and N number of circle corresponding with the position of the N number of indicator lamp on traffic lights base plate is provided with the graphics template
Subregion, it is described using the graphics template for pre-setting, obtain the mesh most matched with the graphics template in the gray level image
Mark region, including：
A. n-th comparison area is selected on the gray level image using the graphics template；
B. obtain in n-th comparison area and be located in the graphics template and outside N number of circular sub-area
Region described in pixel gray value sum, as the first gray value；
C. the N number of area grayscale value for being located at N number of circular sub-area respectively in n-th comparison area is obtained,
And minimum value is selected in N number of area grayscale value as the second gray value, wherein i-th area grayscale of circular sub-area
It is the gray value sum of all pixels being located in n-th comparison area in i-th circular sub-area to be worth, wherein 1
≤i≤N；
D. according to first gray value and second gray value, n-th comparison area and the figure are obtained
First matching degree of template；
E. obtain and be located in n-th comparison area in each circular sub-area in N number of circular sub-area
The gray scale difference of pixel, obtains N number of gray scale difference；Wherein described i-th gray scale difference be n-th comparison area in be located at i-th
The difference of the average gray value of the outer shroud pixel of the image in circular sub-area and the average gray value of inner ring pixel, wherein 1≤i
≤N；
F. according to N number of gray scale difference, the second matching degree of n-th comparison area and the graphics template is obtained,
Wherein, n is positive integer, and 1≤n≤M, M are for needed for using the graphics template by the monitored area image traversal one time
Compare number of times；
G. obtain n-th comprehensive matching degree of comparison area, the comprehensive matching degree be first matching degree with it is described
The product of the second matching degree；
Step a to step g is performed again after taking n=n+1, until obtaining the M comprehensive matching degree；
Using the corresponding comparison area of comprehensive matching degree maximum in M comprehensive matching degree as in the gray level image with institute
State the target area that graphics template is most matched.
Optionally, it is described according to first gray value and second gray value, obtain n-th comparison area with
First matching degree of the graphics template includes：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is first gray scale
Value, a is second gray value；
It is described according to N number of gray scale difference, obtain the second matching degree of n-th comparison area and the graphics template
Including：
Wherein, YnIt is n-th comparison area and the second matching degree of the graphics template, eiIt is i-th gray scale
Difference.
Optionally, methods described also includes：
The indicator lamp information in the traffic lights is obtained, the indicator lamp information is including every in the traffic lights
The position of the indicator lamp of individual color；
The brightness of each indicator lamp in the traffic lights is obtained according to the monitored area image；
The position of the indicator lamp according to each color, and the position of brightness highest indicator lamp determines current unlatching
Indicator lamp color；
The color of the indicator lamp according to the current unlatching judges whether violation vehicle.
According to the second aspect of the embodiment of the present disclosure, there is provided a kind of identifying device of traffic lights, described device includes：
Image capture module, gray level image acquisition module, matching module and identification module；
Described image acquisition module, the monitored area image for obtaining picture pick-up device collection；
The gray level image acquisition module, the gray level image for obtaining the monitored area image；
The matching module, for using the graphics template for pre-setting, with the figure in the acquisition gray level image
The target area that template is most matched；The graphics template is according to traffic lights base plate configuration and the traffic lights bottom
What the shape of the indicator lamp on plate was set；
The identification module, for using the image of the target area as traffic lights in the monitored area image
In traffic lights image.
Optionally, shape of the graphics template with the traffic lights base plate in the monitored area image and big
It is small identical, and N number of circle corresponding with the position of the N number of indicator lamp on traffic lights base plate is provided with the graphics template
Subregion, the matching module includes：Selection submodule, the first gray value acquisition submodule, the second gray value acquisition submodule
With the first matching degree acquisition submodule；
The selection submodule, for selecting n-th comparison area on the gray level image using the graphics template；
The first gray value acquisition submodule, the graphics template is located in n-th comparison area for obtaining
The gray value sum of pixel described in interior and region outside N number of circular sub-area, as the first gray value；
The second gray value acquisition submodule, for obtain in n-th comparison area be located at respectively it is described N number of
N number of area grayscale value of circular sub-area, and minimum value is selected in N number of area grayscale value as the second gray value, its
In the area grayscale value of i-th circular sub-area be to be located in i-th circular sub-area in n-th comparison area
The gray value sum of all pixels, wherein 1≤i≤N；
The first matching degree acquisition submodule, for according to first gray value and second gray value, obtaining
First matching degree of n-th comparison area and the graphics template, wherein, n is positive integer, and 1≤n≤M, M are utilization
The graphics template is by the comparison number of times needed for the monitored area image traversal one time；
The selection submodule, the first gray value acquisition submodule, second ash are performed again after taking n=n+1
Step performed by angle value acquisition submodule and the first matching degree acquisition submodule, until obtaining M the first matching degree；
Using the corresponding comparison area of the first matching degree maximum in M the first matching degrees as in the gray level image with institute
State the target area that graphics template is most matched.
Optionally, the first matching degree acquisition submodule includes：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is first gray scale
Value, a is second gray value.
Optionally, shape of the graphics template with the traffic lights base plate in the monitored area image and big
It is small identical, and N number of circle corresponding with the position of the N number of indicator lamp on traffic lights base plate is provided with the graphics template
Subregion, the matching module includes：Selection submodule, gray scale difference acquisition submodule and the second matching degree acquisition submodule；
The selection submodule, for selecting n-th comparison area on the gray level image using the graphics template；
The gray scale difference acquisition submodule, N number of circular sub-area is located in n-th comparison area for obtaining
In each circular sub-area in pixel gray scale difference, obtain N number of gray scale difference；Wherein described i-th gray scale difference is described n-th
In individual comparison area be located at i-th circular sub-area in image outer shroud pixel average gray value and inner ring pixel it is average
The difference of gray value, wherein 1≤i≤N；
The second matching degree acquisition submodule, for according to N number of gray scale difference, obtaining n-th comparison area
With the second matching degree of the graphics template, wherein, n is positive integer, and 1≤n≤M, M be will be described using the graphics template
Comparison number of times needed for monitored area image traversal one time；
The selection submodule, the gray scale difference acquisition submodule and second matching degree are performed again after taking n=n+1
Step performed by acquisition submodule, until obtaining M the second matching degree；
Using the corresponding comparison area of the second matching degree maximum in M the second matching degrees as in the gray level image with institute
State the target area that graphics template is most matched.
Optionally, the second matching degree acquisition submodule includes：
Wherein, YnIt is n-th comparison area and the second matching degree of the graphics template, eiIt is i-th gray scale
Difference.
Optionally, shape of the graphics template with the traffic lights base plate in the monitored area image and big
It is small identical, and N number of circle corresponding with the position of the N number of indicator lamp on traffic lights base plate is provided with the graphics template
Subregion, the matching module includes：Selection submodule, the first gray value acquisition submodule, the second gray value acquisition submodule,
First matching degree acquisition submodule, gray scale difference acquisition submodule, the second matching degree acquisition submodule and comprehensive matching degree obtain son
Module；
The selection submodule, for selecting n-th comparison area on the gray level image using the graphics template；
The first gray value acquisition submodule, the graphics template is located in n-th comparison area for obtaining
The gray value sum of pixel described in interior and region outside N number of circular sub-area, as the first gray value；
The second gray value acquisition submodule, for obtain in n-th comparison area be located at respectively it is described N number of
N number of area grayscale value of circular sub-area, and minimum value is selected in N number of area grayscale value as the second gray value, its
In the area grayscale value of i-th circular sub-area be to be located in i-th circular sub-area in n-th comparison area
The gray value sum of all pixels, wherein 1≤i≤N；
The first matching degree acquisition submodule, for according to first gray value and second gray value, obtaining
First matching degree of n-th comparison area and the graphics template；
The gray scale difference acquisition submodule, N number of circular sub-area is located in n-th comparison area for obtaining
In each circular sub-area in pixel gray scale difference, obtain N number of gray scale difference；Wherein described i-th gray scale difference is described n-th
In individual comparison area be located at i-th circular sub-area in image outer shroud pixel average gray value and inner ring pixel it is average
The difference of gray value, wherein 1≤i≤N；
The second matching degree acquisition submodule, for according to N number of gray scale difference, obtaining n-th comparison area
With the second matching degree of the graphics template, wherein, n is positive integer, and 1≤n≤M, M be will be described using the graphics template
Comparison number of times needed for monitored area image traversal one time；
The comprehensive matching degree acquisition submodule, for obtaining n-th comprehensive matching degree of comparison area, described comprehensive
It is first matching degree and the product of second matching degree with degree；
The selection submodule, the first gray value acquisition submodule, second ash are performed again after taking n=n+1
Angle value acquisition submodule, the first matching degree acquisition submodule, the gray scale difference acquisition submodule, second matching degree are obtained
Submodule and the step performed by the comprehensive matching degree acquisition submodule are taken, until obtaining the M comprehensive matching degree；
Using the corresponding comparison area of comprehensive matching degree maximum in M comprehensive matching degree as in the gray level image with institute
State the target area that graphics template is most matched.
Optionally, the first matching degree acquisition submodule includes：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is first gray scale
Value, a is second gray value；
The second matching degree acquisition submodule includes：
Wherein, YnIt is n-th comparison area and the second matching degree of the graphics template, eiIt is i-th gray scale
Difference.
Optionally, described device also includes：Data obtaining module, luminance acquisition module, indicator lamp identification module and judgement
Module；
Described information acquisition module, for obtaining the indicator lamp information in the traffic lights, the indicator lamp information
Position including the indicator lamp of each color in the traffic lights；
The luminance acquisition module, for according to each instruction in the monitored area image acquisition traffic lights
The brightness of lamp；
The indicator lamp identification module, for the position of the indicator lamp according to each color, and brightness highest
The position of indicator lamp determines the color of the current indicator lamp opened；
The judge module, the color for the indicator lamp according to the current unlatching judges whether violation vehicle.
By above-mentioned technical proposal, the disclosure utilizes the gray level image of monitored area image, the figure analyzed and pre-set
The gray scale of the image pixel of traffic lights base plate and indicator lamp opposite position is come in obtaining monitored area image in shape template
The region most matched with graphics template, so as to identify the position of traffic lights, can solve the problem that traffic lights use yellow
Signal lamp base plate and shell, cause the problem of the accuracy for recognizing traffic lights, and traffic lights are recognized with improving
The degree of accuracy effect.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary and explanatory, not
The disclosure can be limited.
Brief description of the drawings
Accompanying drawing is, for providing further understanding of the disclosure, and to constitute the part of specification, with following tool
Body implementation method is used to explain the disclosure together, but does not constitute limitation of this disclosure.In the accompanying drawings：
Fig. 1 is a kind of flow chart of the recognition methods of traffic lights that the embodiment of the disclosure one is provided；
Fig. 2 is the flow chart of the recognition methods of another traffic lights that the embodiment of the disclosure one is provided；
Fig. 3 is the schematic diagram of the comparison area of method identification according to Fig. 2；
Fig. 4 is the flow chart of the recognition methods of another traffic lights that the embodiment of the disclosure one is provided；
Fig. 5 is the schematic diagram of the comparison area of method identification according to Fig. 4；
Fig. 6 is the flow chart of the recognition methods of another traffic lights that the embodiment of the disclosure one is provided；
Fig. 7 is the flow chart of the recognition methods of another traffic lights that the embodiment of the disclosure one is provided；
Fig. 8 is a kind of block diagram of the identifying device of traffic lights that the embodiment of the disclosure one is provided；
Fig. 9 is the block diagram of the identifying device of another traffic lights that the embodiment of the disclosure one is provided；
Figure 10 is the block diagram of the identifying device of another traffic lights that the embodiment of the disclosure one is provided；
Figure 11 is the block diagram of the identifying device of another traffic lights that the embodiment of the disclosure one is provided；
Figure 12 is the block diagram of the identifying device of another traffic lights that the embodiment of the disclosure one is provided.
Specific embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in implementation method do not represent all implementation methods consistent with the disclosure.Conversely, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects described in detail in claims, the disclosure.
Before the recognition methods of traffic lights of disclosure offer and device is introduced, first to each implementation of the disclosure
The involved application scenarios of example are introduced.The application scenarios can, at the crossing for having traffic lights, be set near the crossing
Traffic monitoring system is equipped with, comprising the picture pick-up device being arranged near crossing in the system, can be gathered from different angles
The passage situation on road and the image of traffic lights.Traffic lights include base plate and indicator lamp two parts, and signal lamp can be with
There is multiple, in the disclosure each embodiment, illustrated as a example by there are three indicator lamps on traffic lights.
Fig. 1 is a kind of flow chart of the recognition methods of the traffic lights according to an exemplary embodiment, such as Fig. 1 institutes
Show, the method includes：
Step 101, obtains the monitored area image of picture pick-up device collection.
Example, there is the crossing of traffic lights at one, it is desirable to which monitoring is by the vehicle at the crossing with the presence or absence of in violation of rules and regulations
Behavior, except the current state of vehicle to be obtained, in addition it is also necessary to coordinate the state for obtaining traffic lights in the corresponding time period.Traffic
The state of signal lamp can gather the image containing traffic lights by the picture pick-up device in traffic monitoring system, and the image is made
It is monitored area image.
Step 102, obtains the gray level image of monitored area image.
It should be noted that the acquisition of gray level image, can be any one existing gray level image treatment technology, will supervise
Red in area image, green and blue three color ranges of passage are surveyed, is processed according to default ratio, for example, can used
Floating point, displacement method or mean value method to monitored area image process obtaining gray level image.
Step 103, using the graphics template for pre-setting, obtains the target area most matched with graphics template in gray level image
Domain；Graphics template is to be set according to the shape of the indicator lamp on traffic lights base plate configuration and traffic lights base plate.
, wherein it is desired to explanation, graphics template can be the shape in the image of monitored area with traffic lights base plate
Shape is identical with size, and N number of circle corresponding with the position of the N number of indicator lamp on traffic lights base plate is provided with graphics template
Shape subregion, wherein for the signal lamp for generally, on traffic lights having green three colors of reddish yellow, now N is equal to 3,
3 circular sub-areas corresponding with the position of 3 indicator lamps on traffic lights base plate are provided with graphics template.Figure mould
Plate can be set traffic monitoring system when, what the image that can be collected according to picture pick-up device set in advance.For example, can
With the image collected in advance according to picture pick-up device, position, the shapes and sizes of the traffic lights in the image are found, according to
The shapes and sizes set graphics template, while the quantity of indicator lamp and in base plate on traffic lights in the image
On position, be correspondingly also provided with that quantity is identical and position identical circular sub-area on graphics template.Graphics template can be with
Replacing according to variety classes traffic lights carries out accommodation.
For example, it is standard with the graphics template for pre-setting, according to default distance interval to institute in a step 102
Traveled through on the gray level image of acquisition, obtained multiple size and shape identical regions with graphics template and graphics template
Matching degree, and matching degree highest region is therefrom found, as target area.
Step 104, the traffic lights figure using the image of target area as traffic lights in the image of monitored area
Picture.
According to the target area that step 103 is obtained, it is possible to identify in the image of monitored area where traffic lights
Position such that it is able to coordinate the transport condition of vehicle on road to judge whether unlawful practice.Identification traffic lights exist
Position directly by the image interception in the region out, then can be processed by the way of figure is scratched in the image of monitored area,
Can be processed by adding mark to the pixel in the region.
Fig. 2 is the flow chart of the recognition methods of another traffic lights according to an exemplary embodiment, such as Fig. 2
Shown, step 103 includes：
Step 1031, n-th comparison area is selected using graphics template on gray level image.
Step 1032, is located in the graphics template and region outside N number of circular sub-area in n-th comparison area of acquisition
The gray value sum of middle pixel, as the first gray value.
Step 1033, obtains the N number of area grayscale value for being located at N number of circular sub-area respectively in n-th comparison area, and
Minimum value is selected in N number of area grayscale value as the second gray value, wherein i-th area grayscale value of circular sub-area is the
The gray value sum of all pixels being located in n comparison area in i-th circular sub-area, wherein 1≤i≤N.
Step 1034, according to the first gray value and the second gray value, obtains the first of n-th comparison area and graphics template
Matching degree, wherein, n is positive integer, and 1≤n≤M, M are for needed for using the graphics template by monitored area image traversal one time
Comparison number of times.
Optionally, step 1034 includes
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is the first gray value, a is the
Two gray values.
Step 1031 to step 1034 is performed again after taking n=n+1, until M the first matching degree is obtained, by M first
The maximum corresponding comparison area of the first matching degree is used as the target area most matched with graphics template in gray level image in matching degree
Domain.
For example, it is standard with the graphics template for pre-setting, according to default distance interval to institute in a step 102
Traveled through on the gray level image of acquisition, obtained M comparison area.As a example by when n is 5, the 5th comparison area is selected first.
Fig. 3 is the schematic diagram of the comparison area of method identification according to Fig. 2, and wherein region A, region B and region C are respectively the 5th
Three circular sub-areas in comparison area, the indicator lamp in difference corresponding traffic light, region D represents the 5th comparison area
In be located at graphics template in and the region outside 3 circular sub-areas, the base plate of corresponding traffic light.
Secondly, to the gray value summation of all pixels in the D of region, it is designated as the first gray value d.Respectively to region A, region B
With the gray value summation of all pixels in the C of region, trizonal gray value is obtained, by the minimum value note in these three gray values
It is the second gray value a.
Then, first of the 5th comparison area and graphics template is calculated according to the first gray value d and the second gray value a
With degree X5=(d-3*a)/a.
Again by n+1, i.e., the 6th comparison area is repeated the above steps, until obtaining M the first matching degree：X1, X2,....
XM-1, XM, the first matching degree of maximum is therefrom selected, then during the maximum corresponding comparison area of the first matching degree is exactly step 102
The target area most matched with graphics template in acquired gray level image.
Fig. 4 is the flow chart of the recognition methods of another traffic lights according to an exemplary embodiment, such as Fig. 4
Shown, step 103 includes：
Step 1035, n-th comparison area is selected using graphics template on gray level image.
Step 1036, is located at picture in each circular sub-area in N number of circular sub-area in n-th comparison area of acquisition
The gray scale difference of element, obtains N number of gray scale difference；Wherein i-th gray scale difference is to be located at i-th circular sub-area in n-th comparison area
In image outer shroud pixel average gray value and inner ring pixel average gray value difference, wherein 1≤i≤N.
Step 1037, according to N number of gray scale difference, obtains the second matching degree of n-th comparison area and graphics template, wherein, n
Be positive integer, and 1≤n≤M, M for using graphics template by the comparison number of times needed for monitored area image traversal one time.
Optionally, step 1037 includes：
Wherein, YnIt is n-th comparison area and the second matching degree of graphics template, eiIt is i-th gray scale difference.
Step 1035 to step 1037 is performed again after taking n=n+1, until M the second matching degree is obtained, by M second
The maximum corresponding comparison area of the second matching degree is used as the target area most matched with graphics template in gray level image in matching degree
Domain.
For example, it is standard with the graphics template for pre-setting, according to default distance interval to institute in a step 102
Traveled through on the gray level image of acquisition, obtained M comparison area.As a example by when n is 9, the 9th comparison area is selected first.
Fig. 5 is the schematic diagram of the comparison area of method identification according to Fig. 4, and wherein region A1 regions A2 represents the 9th comparison respectively
The outer region of image and endocyclic area in first circular sub-area in region, first instruction in corresponding traffic light
Lamp.Region B1 regions B2 represents in the 9th comparison area the outer region of image and inner ring in second circular sub-area respectively
Region, second indicator lamp in corresponding traffic light.Region C1 regions C2 is represented the 3rd in the 9th comparison area respectively
The outer region of image and endocyclic area in circular sub-area, the 3rd indicator lamp in corresponding traffic light.
Secondly, three gray scale difference e are calculated respectively to three circular sub-areas in the 9th comparison area1、e2And e3.Wherein
e1The average gray value of all pixels in the A2 of region, wherein e are subtracted for the average gray value of all pixels in the A1 of region2It is region
The average gray value of all pixels subtracts the average gray value of all pixels in the B2 of region, wherein e in B13To own in the C1 of region
The average gray value of pixel subtracts the average gray value of all pixels in the C2 of region.
Then, according to three gray scale difference e1、e2And e3Can be matched with the second of graphics template in the hope of the 9th comparison area
Degree Y9=e1+e2+e3。
Again by n+1, i.e., the 10th comparison area is repeated the above steps, until obtaining M the second matching degree：Y1,
Y2,....YM-1, YM, therefrom select the second matching degree of maximum, then the maximum corresponding comparison area of the second matching degree is exactly to walk
The target area most matched with graphics template in acquired gray level image in rapid 102.
Fig. 6 is the flow chart of the recognition methods of another traffic lights according to an exemplary embodiment, such as Fig. 6
Shown, step 103 includes：
Step 103a, n-th comparison area is selected using graphics template on gray level image.
Step 103b, is located in the graphics template and region outside N number of circular sub-area in n-th comparison area of acquisition
The gray value sum of middle pixel, as the first gray value.
Step 103c, obtains the N number of area grayscale value for being located at N number of circular sub-area respectively in n-th comparison area, and
Minimum value is selected in N number of area grayscale value as the second gray value, wherein i-th area grayscale value of circular sub-area is the
The gray value sum of all pixels being located in n comparison area in i-th circular sub-area, wherein 1≤i≤N.
Step 103d, according to the first gray value and the second gray value, obtains the first of n-th comparison area and graphics template
Matching degree.
Optionally, step 103d includes：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of graphics template, d is the first gray value, and a is the second ash
Angle value.
Step 103e, is located at picture in each circular sub-area in N number of circular sub-area in n-th comparison area of acquisition
The gray scale difference of element, obtains N number of gray scale difference；Wherein i-th gray scale difference is to be located at i-th circular sub-area in n-th comparison area
In image outer shroud pixel average gray value and inner ring pixel average gray value difference, wherein 1≤i≤N.
Step 103f, according to N number of gray scale difference, obtains the second matching degree of n-th comparison area and graphics template, wherein, n
Be positive integer, and 1≤n≤M, M for using graphics template by the comparison number of times needed for monitored area image traversal one time.
Optionally, step 103f includes：
Wherein, YnIt is n-th comparison area and the second matching degree of graphics template, eiIt is i-th gray scale difference.
Step 103g, obtains n-th comprehensive matching degree of comparison area, and the comprehensive matching degree is first matching degree
With the product of second matching degree；
Step 103a to step 103g is performed again after taking n=n+1, until M comprehensive matching degree is obtained, it is comprehensive by M
The maximum corresponding comparison area of comprehensive matching degree is used as the target area most matched with graphics template in gray level image in matching degree
Domain.
Example, above-described embodiment combines the two methods of Fig. 2 and embodiment illustrated in fig. 4, chooses n-th comparison area
Calculate the first matching degree X in the region respectively afterwardsnWith the second matching degree Yn, n-th comparison area is calculated according still further to step 103g
Comprehensive matching degree：Xn*Yn.Again by n+1, repeat the above steps, until obtaining M comprehensive matching degree：X1*Y1, X2*Y2,....
XM-1*YM-1, XM*YM, therefrom select the comprehensive matching degree of maximum, then the maximum corresponding comparison area of comprehensive matching degree is exactly to walk
The target area most matched with graphics template in acquired gray level image in rapid 102.
Fig. 7 is the flow chart of the recognition methods of another traffic lights according to an exemplary embodiment, such as Fig. 7
Shown, the method also includes：
Step 105, obtains the indicator lamp information in traffic lights, and indicator lamp information includes each face in traffic lights
The position of the indicator lamp of color.
Example, when traffic monitoring system is set, the image that be able to can be collected according to picture pick-up device obtains traffic
Indicator lamp information in signal lamp, the information can be believed according to the number of indicator lamp, including the indicator lamp of each color in traffic
Position on signal lamp, the information can carry out accommodation according to the replacing of variety classes traffic lights.
Step 106, the brightness of each indicator lamp in traffic lights is obtained according to monitored area image.
It should be noted that traffic lights of the traffic lights according to step 104 acquisition in the image of monitored area
Image, can the quantity of indicator lamp and the position on base plate according to the size of image template, shape and on traffic lights, obtain
Obtain the brightness of each indicator lamp.
The position of step 107, the position of the indicator lamp according to each color, and brightness highest indicator lamp determines current
The color of the indicator lamp of unlatching.
Step 108, the color according to the current indicator lamp opened judges whether violation vehicle.
Example, when step 107 determines that it is the indicator lamp of current unlatching to be second indicator lamp, with reference in step 105
Indicator lamp information in the traffic lights of acquisition, it is possible to it is determined that the color of the current indicator lamp opened, it is possible thereby to judge
Vehicle on present road whether there is unlawful practice, play a part of Traffic monitoring.
In sum, the disclosure utilizes the gray level image of monitored area image, in the graphics template analyzed and pre-set
The gray scale of the image pixel of traffic lights base plate and indicator lamp opposite position come obtain in the image of monitored area with figure mould
The region that plate is most matched, so as to identify the position of traffic lights, can solve the problem that traffic lights use amber lamp bottom
Plate and shell, cause the problem of the accuracy for recognizing traffic lights, with the degree of accuracy for improving identification traffic lights
Effect.
Fig. 8 is a kind of block diagram of the identifying device of the traffic lights according to an exemplary embodiment, such as Fig. 8 institutes
Show, the device 200 includes：Image capture module 201, gray level image acquisition module 202, matching module 203 and identification module
204。
Image capture module 201, the monitored area image for obtaining picture pick-up device collection.
Gray level image acquisition module 202, the gray level image for obtaining monitored area image.
Matching module 203, for using the graphics template for pre-setting, most being matched with graphics template in acquisition gray level image
Target area；Graphics template is the shape according to the indicator lamp on traffic lights base plate configuration and traffic lights base plate
Set.
Identification module 204, for the traffic using the image of target area as traffic lights in the image of monitored area
Signal lamp image.
Fig. 9 is the block diagram of the identifying device of another traffic lights according to an exemplary embodiment, such as Fig. 9 institutes
Show, matching module 203 includes：Selection submodule 2031, the first gray value acquisition submodule 2032, the second gray value obtain submodule
The matching degree acquisition submodule 2034 of block 2033 and first.
Selection submodule 2031, for selecting n-th comparison area on gray level image using graphics template.
First gray value acquisition submodule 2032, for being located in graphics template and positioned at N in n-th comparison area of acquisition
The gray value sum of pixel in region outside individual circular sub-area, as the first gray value.
Second gray value acquisition submodule 2033, for obtain n-th comparison area in respectively be located at N number of circular sub-district
N number of area grayscale value in domain, and minimum value is selected in N number of area grayscale value as the second gray value, wherein i-th circular son
The area grayscale value in region is the gray value sum of all pixels being located in n-th comparison area in i-th circular sub-area,
Wherein 1≤i≤N.
First matching degree acquisition submodule 2034, for according to the first gray value and the second gray value, obtaining n-th comparison
First matching degree of region and graphics template, wherein, n is positive integer, and 1≤n≤M, M are by monitored area using graphics template
Comparison number of times needed for image traversal one time.
Selection submodule 2031, the first gray value acquisition submodule 2032, the second gray value are performed again after taking n=n+1
Step performed by the matching degree acquisition submodule 2034 of acquisition submodule 2033 and first, until M the first matching degree is obtained, will
The maximum corresponding comparison area of the first matching degree with graphics template in gray level image used as most matching in M the first matching degrees
Target area.
Optionally, the first matching degree acquisition submodule 2034 includes：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is first gray scale
Value, a is second gray value.
Figure 10 is the block diagram of the identifying device of another traffic lights according to an exemplary embodiment, such as Figure 10
Shown, matching module 203 includes：Selection submodule 2035, the matching degree of gray scale difference acquisition submodule 2036 and second obtain submodule
Block 2037.
Selection submodule 2035, for selecting n-th comparison area on gray level image using graphics template.
Gray scale difference acquisition submodule 2036, for every in N number of circular sub-area in n-th comparison area of acquisition
The gray scale difference of pixel in individual circular sub-area, obtains N number of gray scale difference；Wherein i-th gray scale difference is to be located in n-th comparison area
The difference of the average gray value of the outer shroud pixel of the image in i-th circular sub-area and the average gray value of inner ring pixel, its
In 1≤i≤N.
Second matching degree acquisition submodule 2037, for according to N number of gray scale difference, obtaining n-th comparison area and figure mould
Second matching degree of plate, wherein, n is positive integer, and 1≤n≤M, M are by monitored area image traversal one time using graphics template
Required comparison number of times.
Selection submodule 2035, the acquisition of the matching degree of gray scale difference acquisition submodule 2036 and second are performed again after taking n=n+1
Step performed by submodule 2037, until M the second matching degree is obtained, by the second matching maximum in M the second matching degree
Corresponding comparison area is spent as the target area most matched with graphics template in gray level image.
Optionally, the second matching degree acquisition submodule 2037 includes：
Wherein, YnIt is n-th comparison area and the second matching degree of graphics template, eiIt is i-th gray scale difference.
Figure 11 is the block diagram of the identifying device of another traffic lights according to an exemplary embodiment, such as Figure 11
Shown, matching module 203 includes：Selection submodule 203a, the first gray value acquisition submodule 203b, the second gray value obtain son
Module 203c, the first matching degree acquisition submodule 203d, gray scale difference acquisition submodule 203e, the second matching degree acquisition submodule
203f and comprehensive matching degree acquisition submodule 203g.
Selection submodule 203a, for selecting n-th comparison area on gray level image using graphics template.
First gray value acquisition submodule 203b, for being located in graphics template and positioned at N in n-th comparison area of acquisition
The gray value sum of pixel in region outside individual circular sub-area, as the first gray value.
Second gray value acquisition submodule 203c, for obtain n-th comparison area in respectively be located at N number of circular sub-district
N number of area grayscale value in domain, and minimum value is selected in N number of area grayscale value as the second gray value, wherein i-th circular son
The area grayscale value in region is the gray value sum of all pixels being located in n-th comparison area in i-th circular sub-area,
Wherein 1≤i≤N.
First matching degree acquisition submodule 203d, for according to the first gray value and the second gray value, obtaining n-th comparison
Region and the first matching degree of graphics template.
Gray scale difference acquisition submodule 203e, for every in N number of circular sub-area in n-th comparison area of acquisition
The gray scale difference of pixel in individual circular sub-area, obtains N number of gray scale difference；Wherein i-th gray scale difference is to be located in n-th comparison area
The difference of the average gray value of the outer shroud pixel of the image in i-th circular sub-area and the average gray value of inner ring pixel, its
In 1≤i≤N.
Second matching degree acquisition submodule 203f, for according to N number of gray scale difference, obtaining n-th comparison area and figure mould
Second matching degree of plate, wherein, n is positive integer, and 1≤n≤M, M are by monitored area image traversal one time using graphics template
Required comparison number of times.
Comprehensive matching degree acquisition submodule 203g, for obtaining n-th comprehensive matching degree of comparison area, comprehensive matching degree
It is the first matching degree and the product of the second matching degree.
Selection submodule 203a, the first gray value acquisition submodule 203b, the second gray value are performed again after taking n=n+1
Acquisition submodule 203c, the first matching degree acquisition submodule 203d, gray scale difference acquisition submodule 203e, the second matching degree obtain son
Step performed by module 203f and comprehensive matching degree acquisition submodule 203g, until the M comprehensive matching degree is obtained, by M
The maximum corresponding comparison area of comprehensive matching degree with graphics template in gray level image used as most matching in individual comprehensive matching degree
Target area.
Optionally, the first matching degree acquisition submodule 203f includes：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of graphics template, d is the first gray value, and a is the second ash
Angle value；
Second matching degree acquisition submodule 203f includes：
Wherein, YnIt is n-th comparison area and the second matching degree of graphics template, eiIt is i-th gray scale difference.
Figure 12 is the block diagram of the identifying device of another traffic lights according to an exemplary embodiment, such as Figure 12
Shown, the device includes：Data obtaining module 205, luminance acquisition module 206, indicator lamp identification module 207 and judge module
208。
Data obtaining module 205, for obtaining the indicator lamp information in traffic lights, indicator lamp information is believed including traffic
The position of the indicator lamp of each color in signal lamp.
Luminance acquisition module 206, the brightness for obtaining each indicator lamp in traffic lights according to monitored area image.
Indicator lamp identification module 207, for the position of the indicator lamp according to each color, and brightness highest indicator lamp
Position determine the color of the current indicator lamp opened.
Judge module 208, for judging whether violation vehicle according to the color of the current indicator lamp opened.
Wherein, illustrating for above-mentioned modules realization of functions has been carried out in detail in above method embodiment
Description, here is omitted.
In sum, the disclosure utilizes the gray level image of monitored area image, in the graphics template analyzed and pre-set
The gray scale of the image pixel of traffic lights base plate and indicator lamp opposite position come obtain in the image of monitored area with figure mould
The region that plate is most matched, so as to identify the position of traffic lights, can solve the problem that traffic lights use amber lamp bottom
Plate and shell, cause the problem of the accuracy for recognizing traffic lights, with the degree of accuracy for improving identification traffic lights
Effect.
Describe the preferred embodiment of the disclosure in detail above in association with accompanying drawing, but, the disclosure is not limited to above-mentioned reality
The detail in mode is applied, in the range of the technology design of the disclosure, those skilled in the art are considering specification and practice
After the disclosure, other embodiments of the disclosure are readily apparent that, belong to the protection domain of the disclosure.
It is further to note that each particular technique feature described in above-mentioned specific embodiment, in not lance
In the case of shield, can be combined by any suitable means.Simultaneously between a variety of implementation methods of the disclosure
Can also be combined, as long as it is without prejudice to the thought of the disclosure, it should equally be considered as disclosure disclosure of that.
The disclosure is not limited to the precision architecture being described above out, and the scope of the present disclosure is only limited by appended claim
System.
Claims (10)
1. a kind of recognition methods of traffic lights, it is characterised in that methods described includes：
Obtain the monitored area image of picture pick-up device collection；
Obtain the gray level image of the monitored area image；
Using the graphics template for pre-setting, the target area most matched with the graphics template in the gray level image is obtained；
The graphics template is to be set according to the shape of the indicator lamp on traffic lights base plate configuration and the traffic lights base plate
Put；
Traffic lights image using the image of the target area as traffic lights in the monitored area image.
2. method according to claim 1, it is characterised in that the graphics template is with the traffic lights base plate in institute
The shapes and sizes stated in the image of monitored area are identical, and are provided with the graphics template and the N on traffic lights base plate
The corresponding N number of circular sub-area in position of individual indicator lamp, it is described using the graphics template for pre-setting, obtain the gray level image
In the target area that is most matched with the graphics template, including：
A. n-th comparison area is selected on the gray level image using the graphics template；
B. obtain and be located in n-th comparison area in the graphics template and the area outside N number of circular sub-area
The gray value sum of pixel described in domain, as the first gray value；
C. obtain in n-th comparison area respectively positioned at N number of area grayscale value of N number of circular sub-area, and
Minimum value is selected in N number of area grayscale value as the second gray value, wherein the area grayscale value of i-th circular sub-area is
The gray value sum of all pixels being located in n-th comparison area in i-th circular sub-area, wherein 1≤i≤
N；
D. according to first gray value and second gray value, n-th comparison area and the graphics template are obtained
The first matching degree, wherein, n is positive integer, and 1≤n≤M, M are by the monitored area image time using the graphics template
Comparison number of times needed for going through one time；
Step a to step d is performed again after taking n=n+1, until obtaining M the first matching degree；
Using the corresponding comparison area of the first matching degree maximum in M the first matching degrees as in the gray level image with the figure
The target area that shape template is most matched.
3. method according to claim 2, it is characterised in that described according to first gray value and second gray scale
Value, obtain n-th comparison area includes with the first matching degree of the graphics template：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is first gray value, and a is
Second gray value.
4. method according to claim 1, it is characterised in that the graphics template is with the traffic lights base plate in institute
The shapes and sizes stated in the image of monitored area are identical, and are provided with the graphics template and the N on traffic lights base plate
The corresponding N number of circular sub-area in position of individual indicator lamp, it is described using the graphics template for pre-setting, obtain the gray level image
In the target area that is most matched with the graphics template, including：
E. n-th comparison area is selected on the gray level image using the graphics template；
F. obtain and be located in n-th comparison area pixel in each circular sub-area in N number of circular sub-area
Gray scale difference, obtain N number of gray scale difference；Wherein described i-th gray scale difference is circular to be located at i-th in n-th comparison area
The difference of the average gray value of the outer shroud pixel of the image in subregion and the average gray value of inner ring pixel, wherein 1≤i≤N；
G. according to N number of gray scale difference, the second matching degree of n-th comparison area and the graphics template is obtained, wherein,
N is positive integer, and 1≤n≤M, M for using the graphics template by the comparison time needed for the monitored area image traversal one time
Number；
Step e to step g is performed again after taking n=n+1, until obtaining M the second matching degree；
Using the corresponding comparison area of the second matching degree maximum in M the second matching degrees as in the gray level image with the figure
The target area that shape template is most matched.
5. method according to claim 4, it is characterised in that described according to N number of gray scale difference, obtains described n-th
Comparison area includes with the second matching degree of the graphics template：
Wherein, YnIt is n-th comparison area and the second matching degree of the graphics template, eiIt is i-th gray scale difference.
6. method according to claim 1, it is characterised in that the graphics template is with the traffic lights base plate in institute
The shapes and sizes stated in the image of monitored area are identical, and are provided with the graphics template and the N on traffic lights base plate
The corresponding N number of circular sub-area in position of individual indicator lamp, it is described using the graphics template for pre-setting, obtain the gray level image
In the target area that is most matched with the graphics template, including：
A. n-th comparison area is selected on the gray level image using the graphics template；
B. obtain and be located in n-th comparison area in the graphics template and the area outside N number of circular sub-area
The gray value sum of pixel described in domain, as the first gray value；
C. obtain in n-th comparison area respectively positioned at N number of area grayscale value of N number of circular sub-area, and
Minimum value is selected in N number of area grayscale value as the second gray value, wherein the area grayscale value of i-th circular sub-area is
The gray value sum of all pixels being located in n-th comparison area in i-th circular sub-area, wherein 1≤i≤
N；
D. according to first gray value and second gray value, n-th comparison area and the graphics template are obtained
The first matching degree；
E. obtain and be located in n-th comparison area pixel in each circular sub-area in N number of circular sub-area
Gray scale difference, obtain N number of gray scale difference；Wherein described i-th gray scale difference is circular to be located at i-th in n-th comparison area
The difference of the average gray value of the outer shroud pixel of the image in subregion and the average gray value of inner ring pixel, wherein 1≤i≤N；
F. according to N number of gray scale difference, the second matching degree of n-th comparison area and the graphics template is obtained, wherein,
N is positive integer, and 1≤n≤M, M for using the graphics template by the comparison time needed for the monitored area image traversal one time
Number；
G. n-th comprehensive matching degree of comparison area is obtained, the comprehensive matching degree is first matching degree and described second
The product of matching degree；
Step a to step g is performed again after taking n=n+1, until obtaining the M comprehensive matching degree；
Using the corresponding comparison area of comprehensive matching degree maximum in M comprehensive matching degree as in the gray level image with the figure
The target area that shape template is most matched.
7. method according to claim 6, it is characterised in that described according to first gray value and second gray scale
Value, obtain n-th comparison area includes with the first matching degree of the graphics template：
Xn=(d-N*a)/a
Wherein, XnIt is n-th comparison area and the first matching degree of the graphics template, d is first gray value, and a is
Second gray value；
It is described according to N number of gray scale difference, obtain the second matching degree bag of n-th comparison area and the graphics template
Include：
Wherein, YnIt is n-th comparison area and the second matching degree of the graphics template, eiIt is i-th gray scale difference.
8. method according to claim 1, it is characterised in that methods described also includes：
The indicator lamp information in the traffic lights is obtained, the indicator lamp information includes each face in the traffic lights
The position of the indicator lamp of color；
The brightness of each indicator lamp in the traffic lights is obtained according to the monitored area image；
The position of the indicator lamp according to each color, and the position of brightness highest indicator lamp determines the current finger opened
Show the color of lamp；
The color of the indicator lamp according to the current unlatching judges whether violation vehicle.
9. a kind of identifying device of traffic lights, it is characterised in that described device includes：Image capture module, gray level image
Acquisition module, matching module and identification module；
Described image acquisition module, the monitored area image for obtaining picture pick-up device collection；
The gray level image acquisition module, the gray level image for obtaining the monitored area image；
The matching module, for using the graphics template for pre-setting, with the graphics template in the acquisition gray level image
The target area for most matching；The graphics template is according on traffic lights base plate configuration and the traffic lights base plate
Indicator lamp shape set；
The identification module, for using the image of the target area as traffic lights in the monitored area image
Traffic lights image.
10. device according to claim 9, it is characterised in that the graphics template exists with the traffic lights base plate
Shapes and sizes in the monitored area image are identical, and be provided with the graphics template with traffic lights base plate
The corresponding N number of circular sub-area in position of N number of indicator lamp, the matching module includes：Selection submodule, the first gray value are obtained
Submodule, the second gray value acquisition submodule and the first matching degree acquisition submodule；
The selection submodule, for selecting n-th comparison area on the gray level image using the graphics template；
The first gray value acquisition submodule, be located in n-th comparison area in the graphics template for obtaining and
The gray value sum of pixel described in region outside N number of circular sub-area, as the first gray value；
The second gray value acquisition submodule, N number of circle is located at for obtaining in n-th comparison area respectively
N number of area grayscale value of subregion, and minimum value is selected in N number of area grayscale value as the second gray value, wherein i-th
The area grayscale value of individual circular sub-area is owning in i-th circular sub-area in n-th comparison area
The gray value sum of pixel, wherein 1≤i≤N；
The first matching degree acquisition submodule, for according to first gray value and second gray value, obtaining described
First matching degree of n-th comparison area and the graphics template, wherein, n is positive integer, and 1≤n≤M, M are using described
Graphics template is by the comparison number of times needed for the monitored area image traversal one time；
The selection submodule, the first gray value acquisition submodule, second gray value are performed again after taking n=n+1
Step performed by acquisition submodule and the first matching degree acquisition submodule, until obtaining M the first matching degree；
Using the corresponding comparison area of the first matching degree maximum in M the first matching degrees as in the gray level image with the figure
The target area that shape template is most matched.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611265156.8A CN106781521B (en) | 2016-12-30 | 2016-12-30 | Traffic signal lamp identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611265156.8A CN106781521B (en) | 2016-12-30 | 2016-12-30 | Traffic signal lamp identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106781521A true CN106781521A (en) | 2017-05-31 |
| CN106781521B CN106781521B (en) | 2020-12-25 |
Family
ID=58952518
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611265156.8A Active CN106781521B (en) | 2016-12-30 | 2016-12-30 | Traffic signal lamp identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106781521B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107316488A (en) * | 2017-08-23 | 2017-11-03 | 苏州豪米波技术有限公司 | The recognition methods of signal lamp, device and system |
| CN109389079A (en) * | 2018-09-30 | 2019-02-26 | 无锡职业技术学院 | A kind of traffic lights recognition methods |
| CN109544955A (en) * | 2018-12-26 | 2019-03-29 | 广州小鹏汽车科技有限公司 | A kind of state acquiring method and system of traffic lights |
| CN109784317A (en) * | 2019-02-28 | 2019-05-21 | 东软睿驰汽车技术（沈阳）有限公司 | A kind of recognition methods of traffic lights and device |
| CN110717438A (en) * | 2019-10-08 | 2020-01-21 | 东软睿驰汽车技术（沈阳）有限公司 | Traffic signal lamp identification method and device |
| CN112131922A (en) * | 2019-06-24 | 2020-12-25 | 丰田自动车株式会社 | Signal recognition system |
| CN112507956A (en) * | 2020-12-21 | 2021-03-16 | 北京百度网讯科技有限公司 | Signal lamp identification method and device, electronic equipment, road side equipment and cloud control platform |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101882217A (en) * | 2010-02-26 | 2010-11-10 | 杭州海康威视软件有限公司 | Target classification method of video image and device |
| US20110235896A1 (en) * | 2010-03-26 | 2011-09-29 | Hon Hai Precision Industry Co., Ltd. | Burr detecting apparatus and burr detection method thereof |
| CN103577809A (en) * | 2013-11-12 | 2014-02-12 | 北京联合大学 | Ground traffic sign real-time detection method based on intelligent driving |
| CN103607558A (en) * | 2013-11-04 | 2014-02-26 | 深圳市中瀛鑫科技股份有限公司 | Video monitoring system, target matching method and apparatus thereof |
| CN103680176A (en) * | 2012-09-18 | 2014-03-26 | 中国移动通信集团公司 | Signal identification system based on position information |
| CN104637304A (en) * | 2015-03-08 | 2015-05-20 | 无锡桑尼安科技有限公司 | Vehicle violation detection method for overtaking left lane |
| CN105809138A (en) * | 2016-03-15 | 2016-07-27 | 武汉大学 | Road warning mark detection and recognition method based on block recognition |
| CN105913041A (en) * | 2016-04-27 | 2016-08-31 | 浙江工业大学 | Pre-marked signal lights based identification method |
| CN106023623A (en) * | 2016-07-28 | 2016-10-12 | 南京理工大学 | Recognition and early warning method of vehicle-borne traffic signal and symbol based on machine vision |
| CN106128096A (en) * | 2016-06-29 | 2016-11-16 | 安徽金赛弗信息技术有限公司 | A kind of motor vehicles is not opened steering indicating light by regulation and is broken rules and regulations lane change intelligent identification Method and device thereof |
-
2016
- 2016-12-30 CN CN201611265156.8A patent/CN106781521B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101882217A (en) * | 2010-02-26 | 2010-11-10 | 杭州海康威视软件有限公司 | Target classification method of video image and device |
| US20110235896A1 (en) * | 2010-03-26 | 2011-09-29 | Hon Hai Precision Industry Co., Ltd. | Burr detecting apparatus and burr detection method thereof |
| CN103680176A (en) * | 2012-09-18 | 2014-03-26 | 中国移动通信集团公司 | Signal identification system based on position information |
| CN103607558A (en) * | 2013-11-04 | 2014-02-26 | 深圳市中瀛鑫科技股份有限公司 | Video monitoring system, target matching method and apparatus thereof |
| CN103577809A (en) * | 2013-11-12 | 2014-02-12 | 北京联合大学 | Ground traffic sign real-time detection method based on intelligent driving |
| CN104637304A (en) * | 2015-03-08 | 2015-05-20 | 无锡桑尼安科技有限公司 | Vehicle violation detection method for overtaking left lane |
| CN105809138A (en) * | 2016-03-15 | 2016-07-27 | 武汉大学 | Road warning mark detection and recognition method based on block recognition |
| CN105913041A (en) * | 2016-04-27 | 2016-08-31 | 浙江工业大学 | Pre-marked signal lights based identification method |
| CN106128096A (en) * | 2016-06-29 | 2016-11-16 | 安徽金赛弗信息技术有限公司 | A kind of motor vehicles is not opened steering indicating light by regulation and is broken rules and regulations lane change intelligent identification Method and device thereof |
| CN106023623A (en) * | 2016-07-28 | 2016-10-12 | 南京理工大学 | Recognition and early warning method of vehicle-borne traffic signal and symbol based on machine vision |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107316488A (en) * | 2017-08-23 | 2017-11-03 | 苏州豪米波技术有限公司 | The recognition methods of signal lamp, device and system |
| CN109389079A (en) * | 2018-09-30 | 2019-02-26 | 无锡职业技术学院 | A kind of traffic lights recognition methods |
| CN109389079B (en) * | 2018-09-30 | 2022-02-15 | 无锡职业技术学院 | Traffic signal lamp identification method |
| CN109544955A (en) * | 2018-12-26 | 2019-03-29 | 广州小鹏汽车科技有限公司 | A kind of state acquiring method and system of traffic lights |
| CN109784317A (en) * | 2019-02-28 | 2019-05-21 | 东软睿驰汽车技术（沈阳）有限公司 | A kind of recognition methods of traffic lights and device |
| CN109784317B (en) * | 2019-02-28 | 2021-02-23 | 东软睿驰汽车技术（沈阳）有限公司 | Traffic signal lamp identification method and device |
| CN112131922A (en) * | 2019-06-24 | 2020-12-25 | 丰田自动车株式会社 | Signal recognition system |
| CN112131922B (en) * | 2019-06-24 | 2024-05-28 | 丰田自动车株式会社 | Signal identification system |
| CN110717438A (en) * | 2019-10-08 | 2020-01-21 | 东软睿驰汽车技术（沈阳）有限公司 | Traffic signal lamp identification method and device |
| CN112507956A (en) * | 2020-12-21 | 2021-03-16 | 北京百度网讯科技有限公司 | Signal lamp identification method and device, electronic equipment, road side equipment and cloud control platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106781521B (en) | 2020-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106781521A (en) | The recognition methods of traffic lights and device | |
| CN103345766B (en) | A kind of signal lamp recognition methods and device | |
| CN110197589B (en) | Deep learning-based red light violation detection method | |
| CN102834309B (en) | Automotive vehicle monitoring of tools, warning and control system | |
| CN103559507B (en) | The method for traffic sign detection being combined based on color with shape facility | |
| CN111212772A (en) | Method and device for determining a driving strategy of a vehicle | |
| JP4863951B2 (en) | Traffic light recognition device | |
| CN104408424B (en) | A kind of multi signal lamp recognition methods based on image procossing | |
| CN104769652B (en) | Method and system for detecting traffic lights | |
| CN103488987A (en) | Video-based method and device for detecting traffic lights | |
| CN106570494A (en) | Traffic signal lamp recognition method and device based on convolution neural network | |
| CN104778833B (en) | The method for identifying traffic lights | |
| CN107273838A (en) | Traffic lights capture the processing method and processing device of picture | |
| CN109564462A (en) | Electronic tracking equipment, Third-party Logistics and electronic tracking method | |
| CN105023452B (en) | A kind of method and device of multichannel traffic lights signal acquisition | |
| CN104637031A (en) | Eye image processing method and device | |
| CN103544480A (en) | Vehicle color recognition method | |
| CN103324957A (en) | Identification method and identification device of state of signal lamps | |
| JP3006471B2 (en) | Sign recognition method | |
| CN105430352A (en) | Processing method for video monitor image | |
| CN102306276A (en) | Method for identifying color of vehicle body in video vehicle image based on block clustering | |
| CN110675650A (en) | Parking space state monitoring method | |
| Chiang et al. | Road speed sign recognition using edge-voting principle and learning vector quantization network | |
| CN110570347B (en) | Color image graying method for lane line detection | |
| CN107316002A (en) | A kind of night front vehicles recognition methods based on Active Learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |