CN106650724A - Method and device for building traffic sign database - Google Patents
Method and device for building traffic sign database Download PDFInfo
- Publication number
- CN106650724A CN106650724A CN201610932080.3A CN201610932080A CN106650724A CN 106650724 A CN106650724 A CN 106650724A CN 201610932080 A CN201610932080 A CN 201610932080A CN 106650724 A CN106650724 A CN 106650724A
- Authority
- CN
- China
- Prior art keywords
- traffic
- field picture
- signss
- traffic signss
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
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
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a method and a device for building a traffic sign database, and relates to the technical field of traffic measurement. The method comprises the steps of acquiring location information and traffic sign size and distance information of a traffic sign in each frame image acquired by camera shooting equipment, wherein the distance information is used for representing the distance between the traffic sign in each frame image and the camera shooting equipment; determining the location of the traffic sign in each frame image in the frame image according to the location information and the traffic sign size and distance information of the traffic sign in each frame image, and acquiring a traffic sign image in each frame image according to the location; and generating a traffic sign database according to the traffic sign image in each frame image and the distance information of the traffic sign in each frame image. According to the invention, the speed of acquiring a traffic sign from an image can be improved, and the error making probability can be reduced.
Description
Technical field
It relates to technical field of transportation measurement, more particularly to a kind of method and dress for setting up traffic signss data base
Put.
Background technology
Vehicle will realize Function for Automatic Pilot, it is desirable to be able to accurately identify the various traffic signss on road, it is therefore desirable to pre-
A traffic signss data base is set up first.In prior art, the method for setting up traffic signss data base is to clap first with video camera
Image is taken the photograph, then the image to shooting carries out the artificial image scratched figure, obtain the traffic signss in the image for shooting, and be stored in data
Storehouse.As video camera is the image by frame shooting, therefore data volume is very huge, due to workload by the way of artificial stingy figure
It is huge, so speed is slow and easily malfunctions.
The content of the invention
The disclosure provides a kind of method and apparatus for setting up traffic signss data base, to the speed for solving prior art
Slowly, error-prone problem.
To achieve these goals, the first aspect according to the embodiment of the present disclosure, there is provided one kind sets up traffic signss data
The method in storehouse, methods described include:
Obtain the positional information of traffic signss in each two field picture of picture pick-up device collection, traffic signss size and away from
From information, the distance of traffic signss and the picture pick-up device that the range information is used to representing in each two field picture;
Positional information, traffic signss size and range information according to the traffic signss in each two field picture determines
Position of the traffic signss in each two field picture in each two field picture, and it is each according to the position acquisition
Traffic Sign Images in two field picture;
According to the distance of the traffic signss in the Traffic Sign Images in each two field picture and each two field picture
Information generates traffic signss data base.
Optionally, the positional information of the traffic signss according in each two field picture, traffic signss size and
Range information determines the position of traffic signss in each two field picture in each two field picture, and according to the position
The Traffic Sign Images in each two field picture are obtained, including:
Positional information, the traffic signss size according to the traffic signss in each two field picture and it is described away from
The position of traffic signss in each two field picture in each two field picture is determined from information, and obtained according to the position
The traffic signss taken in each two field picture slightly match image;
Image is slightly matched to the traffic signss in each two field picture carries out default image procossing, obtains described each
Traffic signss essence matching image in two field picture;
Using the traffic signss essence matching image in each two field picture as the traffic signss in each two field picture
Image.
Optionally, methods described also includes:
The positional information of the positional information and traffic signss to be captured of the picture pick-up device is obtained before collection image;
Take the photograph according to the positional information of the positional information and the traffic signss to be captured of the picture pick-up device determines
As the distance between equipment and described traffic signss to be captured;
When the distance is less than the first setting value, the image in the picture pick-up device collection coverage is begun through;
When the distance is less than the second setting value, stop shooting.
Optionally, the positional information of the traffic signss in each two field picture includes the traffic in each two field picture
The geographical location information and elevation information of mark, the positional information of the traffic signss according in each two field picture, friendship
Logical mark size and range information determine the position of traffic signss in each two field picture in each two field picture,
And the traffic signss in each two field picture according to the position acquisition slightly match image, including:
Geographical location information, elevation information, traffic signss size and range information according to each two field picture enters
Row coordinate system transformation, determines thick matched position of the traffic signss in each two field picture in each two field picture;
FIG pull handle is carried out to the thick matched position in each two field picture, the traffic in each two field picture is obtained
The thick matching image of mark;
The traffic signss in each two field picture slightly match image carries out default image procossing, obtains described
Traffic signss essence matching image in each two field picture, including:
Image is slightly matched to the traffic signss in each two field picture carries out RGB normalizeds;To each frame
The traffic signss carried out after RGB normalizeds of image slightly match image, and to carry out color threshold segmentation described every to determine
Traffic signss in one two field picture slightly match the smart matched position that the traffic signss in image are located;
FIG pull handle is carried out to the described smart matched position in each two field picture, is obtained in each two field picture
Traffic signss essence matching image.
Optionally, the traffic in the Traffic Sign Images and each two field picture according in each two field picture
The range information of mark generates traffic signss data base, including:
According to the range information of the traffic signss in each two field picture to the traffic signss in each two field picture
Image is grouped, and obtains the packet of at least one Traffic Sign Images;
Each Traffic Sign Images packet is labeled, what described each Traffic Sign Images were grouped is noted for indicating
The classification belonging to Traffic Sign Images in described each Traffic Sign Images packet.
According to the second aspect of the embodiment of the present disclosure, there is provided a kind of device for setting up traffic signss data base, described device
Including:Navigation module, image processing module, image storage module;
The navigation module, for obtain picture pick-up device collection each two field picture in traffic signss positional information,
Traffic signss size and range information, the traffic signss that the range information is used to representing in each two field picture with it is described
The distance of picture pick-up device;
Described image processing module, for the positional information according to the traffic signss in each two field picture, traffic mark
Will size and range information determine the position of traffic signss in each two field picture in each two field picture, and root
According to the Traffic Sign Images in each two field picture described in the position acquisition;
Described image memory module, for according to the Traffic Sign Images in each two field picture and each frame figure
The range information of the traffic signss as in generates traffic signss data base.
Optionally, described image processing module is used for:
Positional information, the traffic signss size according to the traffic signss in each two field picture and it is described away from
The position of traffic signss in each two field picture in each two field picture is determined from information, and obtained according to the position
The traffic signss taken in each two field picture slightly match image;
Image is slightly matched to the traffic signss in each two field picture carries out default image procossing, obtains described each
Traffic signss essence matching image in two field picture;
Using the traffic signss essence matching image in each two field picture as the traffic signss in each two field picture
Image.
Optionally, described device also includes:Apart from acquisition module and image capture module;
The navigation module, is additionally operable to obtain the positional information of the picture pick-up device and to be captured before collection image
The positional information of traffic signss;
It is described apart from acquisition module, for the positional information according to the picture pick-up device and the traffic signss to be captured
Positional information determine the distance between the picture pick-up device and described traffic signss to be captured;
Described image acquisition module, for when the distance is less than the first setting value, beginning through the picture pick-up device
Image in collection coverage;When the distance is less than the second setting value, stop shooting.
Optionally, the positional information of the traffic signss in each two field picture includes the traffic in each two field picture
The geographical location information and elevation information of mark, described image processing module include:Thick matching module and smart matching module;
The thick matching module is used for, geographical location information, elevation information according to each two field picture, traffic signss
Size and range information carry out coordinate system transformation, determine the traffic signss in each two field picture in each two field picture
In thick matched position;FIG pull handle is carried out to the thick matched position in each two field picture, each two field picture is obtained
In traffic signss slightly match image;
The smart matching module is used for, and slightly matching image to the traffic signss in each two field picture carries out RGB normalizings
Change is processed;Image is slightly matched to the traffic signss carried out after RGB normalizeds of each two field picture carries out color
Threshold segmentation slightly matches the smart match bit at the place of the traffic signss in image to determine the traffic signss in each two field picture
Put;FIG pull handle is carried out to the described smart matched position in each two field picture, the traffic in each two field picture is obtained
Mark essence matching image.
Optionally, described image memory module is used for:
According to the range information of the traffic signss in each two field picture to the traffic signss in each two field picture
Image is grouped, and obtains the packet of at least one Traffic Sign Images;
Each Traffic Sign Images packet is labeled, what described each Traffic Sign Images were grouped is noted for indicating
The classification belonging to Traffic Sign Images in described each Traffic Sign Images packet.
By above-mentioned technical proposal, positional information, traffic signss size and the range information of traffic signss are combined, it is real
Figure is scratched by the automatization for having showed traffic signss, by the way of comparing in traditional method using artificial stingy figure, it is possible to increase obtain from image
The speed of traffic signss is taken, and the probability for malfunctioning can be reduced.
Other feature and advantage of the disclosure will be described in detail in subsequent specific embodiment part, it should be understood that
, the general description of the above and detailed description hereinafter are only exemplary and explanatory, can not limit the disclosure.
Description of the drawings
Accompanying drawing is, for providing further understanding of the disclosure, and to constitute the part of description, with following tool
Body embodiment is used for explaining the disclosure together, but does not constitute restriction of this disclosure.In the accompanying drawings:
Fig. 1 is a kind of flow chart of method for setting up traffic signss data base that one embodiment of the disclosure is provided;
Fig. 2 is the flow chart of the method that the another kind that one embodiment of the disclosure is provided sets up traffic signss data base;
Fig. 3 is the flow chart of the method that another of one embodiment of disclosure offer sets up traffic signss data base;
Fig. 4 is the flow chart of the method that another of one embodiment of disclosure offer sets up traffic signss data base;
Fig. 5 is a kind of block diagram of device for setting up traffic signss data base that one embodiment of the disclosure is provided;
Fig. 6 is the block diagram of the device that the another kind that one embodiment of the disclosure is provided sets up traffic signss data base;
Fig. 7 is a kind of block diagram of image processing module that one embodiment of the disclosure is provided.
Specific embodiment
Here in detail exemplary embodiment will be illustrated, its example is illustrated in the accompanying drawings.Explained below 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 embodiment do not represent all embodiments consistent with the disclosure.Conversely, they be only with as appended by
The example of consistent apparatus and method in terms of some described in detail in claims, the disclosure.
The disclosure is provided set up traffic signss data base method and apparatus explanation before, first to the disclosure each
Involved by embodiment, application scenarios are introduced.The application scenarios can be the vehicle for being mounted with picture pick-up device, for shooting
The traffic signss run into during traveling.Wherein vehicle can be the vehicle that any one is furnished with navigation module, such as automobile,
Other kinds of motor vehicles or non-motor vehicle, picture pick-up device can be the equipment such as video camera, high speed camera, drive recorder,
In the disclosure each embodiment, with vehicle as automobile, picture pick-up device as video camera as a example by illustrate.
Fig. 1 is a kind of flow chart of the method for setting up traffic signss data base according to an exemplary embodiment, such as
Shown in Fig. 1, the method includes:
Step 101, the positional information of the traffic signss in each two field picture of acquisition picture pick-up device collection, traffic signss are big
Little and range information, the range information are used for the distance for representing the traffic signss in each two field picture and picture pick-up device.
Wherein, the positional information of traffic signss can include the geographical location information of traffic signss(Such as latitude and longitude coordinates)
And elevation information, the geographical location information, elevation information and traffic signss size can from vehicle-mounted navigation module institute
Obtain in the high accuracy map of offer, the range information of traffic signss can be according to the vehicle position information of navigation module acquisition
Determine with the information of the traffic signss, can also be obtained to measure using laser equipment, infrared equipment or radar, or the shooting
Equipment can be the picture pick-up device with distance measurement function, can directly measure the distance of reference object.Vehicle-mounted navigation module institute
The navigation system of support can be global positioning system(English:Global Position System, abbreviation:GPS), the whole world defends
Star navigation system(English:Global Navigation Satellite System, abbreviation:GLONASS), Galilean satellite leads
Boat system(English:Galileo Satellite Navigation System, referred to as:GSNS)Or Beidou satellite navigation system
(English:BeiDou Navigation Satellite System, abbreviation BDS)In one or several.
Step 102, the positional information, traffic signss size and range information according to the traffic signss in each two field picture
Determine position of the traffic signss in each two field picture in each two field picture, and according in each two field picture of the position acquisition
Traffic Sign Images.
Example, for arbitrary two field picture, the geographical location information and height of the traffic signss provided according to navigation module
Information, it becomes possible to determine the three-dimensional coordinate geographically of the traffic signss, with reference between the traffic signss and picture pick-up device
Range information and traffic signss size, through coordinate transform, it is possible to determine the traffic signss in the two field picture in the frame figure
Position as in, it is determined that position of the traffic signss in the two field picture can be obtained by Traffic Sign Images.
Optionally, in order that the Traffic Sign Images for arriving are more accurate, will can regard as from the image of the position acquisition
It is the thick matching image of the traffic signss, and further image procossing is carried out to the thick matching image, traffic mark can be obtained
Will essence matching image, and as the Traffic Sign Images of the two field picture.In the same manner, can obtain each by above-mentioned method
Traffic Sign Images in two field picture.
Step 103, according to the distance of the traffic signss in the Traffic Sign Images in each two field picture and each two field picture
Information generates traffic signss data base.
Example, memory module can be stored in by the Traffic Sign Images in each two field picture that said method is obtained
In, wherein during Traffic Sign Images in each two field picture is stored, can be by the distance of the traffic signss in each two field picture
Information is stored together, and the traffic indication map of so storage seems the Traffic Sign Images with range information.Wherein, store
Module can be included in the vehicle-mounted industrial computer of vehicle, the friendship in Traffic Sign Images and each two field picture in each two field picture
The range information of logical mark can be stored during memory module is transferred to wire transmission or by way of being wirelessly transferred, example
Such as, it is possible to use Ethernet switch is transmitted by Ethernet, it is also possible to through vehicle bus, such as CAN transmission
Stored on vehicle-mounted industrial computer.
Fig. 2 is the flow chart of the method that the another kind according to an exemplary embodiment sets up traffic signss data base,
As shown in Fig. 2 the positional information according to the traffic signss in each two field picture, traffic signss size described in step 102 with
It is and range information determines the position of traffic signss in each two field picture in each two field picture and each according to the position acquisition
The step of Traffic Sign Images in two field picture, includes:
Step 1021, the positional information, traffic signss size and distance letter according to the traffic signss in each two field picture
Position of the traffic signss in each two field picture of breath determination in each two field picture, and according to each frame of the position acquisition
Traffic signss in image slightly match image.
It is possible, firstly, to geographical location information, elevation information, traffic signss size according to each two field picture and away from
Coordinate system transformation is carried out from information, thick matching of the traffic signss in each two field picture in each two field picture is determined
Position;
Secondly, FIG pull handle can be carried out to the thick matched position in each two field picture, obtains each frame figure
Traffic signss as in slightly match image.
Step 1022, slightly matching image to the traffic signss in each two field picture carries out default image procossing, obtains institute
State the traffic signss essence matching image in each two field picture.
It is possible, firstly, to slightly matching image to the traffic signss in each two field picture carries out RGB normalizeds;Carry out
After RGB normalizeds, color threshold segmentation is carried out by matching image thick to traffic signss and determines traffic signss slightly matching figure
The smart matched position that traffic signss as in are located.
Secondly, FIG pull handle is carried out to smart matched position, obtains the traffic signss essence matching image in each two field picture.
Step 1023, using the traffic signss essence matching image in each two field picture as the traffic signss in each two field picture
Image.
Example, for arbitrary two field picture, the geographical location information and height of the traffic signss provided according to navigation module
Information, can determine the three-dimensional coordinate geographically of the traffic signss, and the navigation module can also provide the ground of picture pick-up device
Reason positional information and elevation information, so as to also can determine picture pick-up device three-dimensional coordinate geographically, in conjunction with traffic signss
With the distance between picture pick-up device information and traffic signss size, through coordinate transform, it may be determined that the friendship in the two field picture
Position of the logical mark in the two field picture(The position can be understood as the region covered by traffic signss in image), the position can
To be interpreted as thick matched position of the traffic signss in the two field picture, then the image of the position is amplified, figure is scratched, you can
Obtain the thick matching image of the traffic signss.Afterwards, RGB normalizeds can be carried out to the thick matching image, removes illumination
Interference image caused with shade.Again to the image after process through color threshold segmentation(Can be according to the traffic signss
Type chooses corresponding flag colors), the corresponding minimum enclosed rectangle of the traffic signss is found, according to the minimum enclosed rectangle just
The smart matched position that traffic signss slightly match traffic signss place in image can be obtained, then smart matched position is carried out
FIG pull handle, it is possible to obtain the traffic signss essence matching image of the two field picture, and as the traffic signss of the two field picture
Image.In the same manner, the Traffic Sign Images in each two field picture can be obtained by above-mentioned method.
In addition, the geographical three-dimensional coordinate of traffic signss can be understood as seat of the traffic signss in geographical three-dimensional system of coordinate
Mark, position of the traffic signss in its affiliated image can be understood as seat of the traffic signss in the image coordinate system of the image
Mark, accordingly, the traffic signss just can be regarded as the distance between origin of two coordinate systems with the distance of picture pick-up device, because
This relation according to geographical three-dimensional system of coordinate and image coordinate system, using the geographical three-dimensional coordinate of the traffic signss, the traffic mark
The relation of the range information of will and the two coordinate systems carries out coordinate transform and can just draw traffic signss in its affiliated image
In position.
Fig. 3 is the flow chart of the method that another according to an exemplary embodiment sets up traffic signss data base,
As shown in figure 3, the method also includes:
Step 104, obtained the position of the positional information and traffic signss to be captured of picture pick-up device before collection image
Information.
Wherein, the positional information of video camera can be obtained according to onboard navigation system positioning, traffic signss to be captured
Positional information can be obtained by the high accuracy map of onboard navigation system.
According to the positional information of the positional information and traffic signss to be captured of picture pick-up device, step 105, determines that shooting sets
It is standby the distance between with traffic signss to be captured.
Example, the distance between video camera and traffic signss to be captured, the shooting that can be provided by navigation module
The geographical location information determination of the geographical location information and traffic signss of machine, can also utilize laser equipment, infrared equipment or thunder
Up to obtaining to measure.
Step 106, when distance is less than the first setting value, begins through the image in picture pick-up device collection coverage;
When distance is less than the second setting value, stop shooting.
Example, in order that the image clearly that photographs of picture pick-up device effectively, should avoid apart from traffic signss too far
Or it is excessively near in the case of shot, therefore, when distance is less than the first setting value, picture pick-up device is entered and clearly can be shot
The scope of traffic signss, picture pick-up device can proceed by shooting;When distance is less than the second setting value, picture pick-up device is entered not
In the range of effectively shooting traffic signss, therefore the picture pick-up device stops shooting.Wherein, first setting value can be 100
Rice, the second setting value can be 2 meters.In addition, it is noted that picture pick-up device starts to clap during shooting value closes shooting every time
The all images taken the photograph can be stored under same catalogue.
Fig. 4 be one embodiment of the disclosure provide another set up traffic signss data base method flow chart, such as Fig. 4
It is shown, described in step 103 according to the traffic signss in the Traffic Sign Images in each two field picture and each two field picture
The step of range information generates traffic signss data base includes:
Step 1031, according to the range information of the traffic signss in each two field picture to the traffic signss in each two field picture
Image is grouped, and obtains the packet of at least one Traffic Sign Images.
Example, what picture pick-up device was collected be a period of time in image sequence, the image comprising multiple frames, wherein often
Traffic Sign Images in one two field picture all correspond to a range information(The acquisition of the range information may refer to step 101, away from
Storage from information may refer to step 103), within this time, multiple traffic signss can be photographed, and as shooting sets
For when shooting, within the unit interval shoot multi-frame images, such as 24 frame per second, therefore each traffic signs may shoot
Multiple image, it is possible to which the every two field picture in the image sequence for collecting picture pick-up device is carried out point according to different traffic signss
Group, the packet can be with the partial gradients of range information as foundation.Example, in this time, photograph two traffic marks
Will, first traffic signss correspondence N frame(1st arrives N frames)Image, second traffic signss correspondence M frame(N+1 is to N+M frames)
N+M two field pictures are contained in image, i.e. image sequence, wherein N, M is positive integer.Assume when nth frame image and N+1 two field pictures
Between partial gradient it is maximum when, image sequence can be grouped from nth frame and N+1 frames, front N two field pictures are one
Group, N+1 to N+M two field pictures are one group.If the picture frame only comprising a traffic signss correspondence collection in image sequence, that
Packet only one of which.
Step 1032, is labeled to each Traffic Sign Images packet, and the mark of each Traffic Sign Images packet is used
The classification belonging to Traffic Sign Images in each Traffic Sign Images packet is indicated.
Example, can contrast in each Traffic Sign Images packet with the traffic signss pattern of official's issue as foundation
Traffic Sign Images in any one two field picture, you can complete mark, can be obtained by according to the Traffic Sign Images after mark
Traffic signss data base with range information.As each Traffic Sign Images packet only needs to mark once, greatly reduce
The workload of mark, decreases the probability of error.
In sum, the disclosure combines the positional information of traffic signss, traffic signss size and range information, realizes
Figure is scratched by the automatization of traffic signss, by the way of comparing in traditional method using artificial stingy figure, it is possible to increase the acquisition from image
The speed of traffic signss, and the probability for malfunctioning can be reduced.
Fig. 5 is a kind of block diagram of device for setting up traffic signss data base that one embodiment of the disclosure is provided, such as Fig. 5 institutes
Show, the device includes navigation module 501, image processing module 502, image storage module 503.
Navigation module 501, for obtaining the positional information of the traffic signss in each two field picture that picture pick-up device is gathered, handing over
Logical mark size and range information, the traffic signss that range information is used to representing in each two field picture and picture pick-up device away from
From.
Image processing module 502, for the positional information according to the traffic signss in each two field picture, traffic signss size
And range information determines the position of traffic signss in each two field picture in each two field picture, and according to the position acquisition
Traffic Sign Images in each two field picture.
Image storage module 503, for according to the friendship in the Traffic Sign Images in each two field picture and each two field picture
The range information of logical mark generates traffic signss data base.
Optionally, image processing module 502 is used for:
Positional information, traffic signss size and range information according to the traffic signss in each two field picture determines every
Position of the traffic signss in one two field picture in each two field picture, and according to the traffic in each two field picture of the position acquisition
The thick matching image of mark.
Image is slightly matched to the traffic signss in each two field picture carries out default image procossing, obtains in each two field picture
Traffic signss essence matching image.
Using the traffic signss essence matching image in each two field picture as the Traffic Sign Images in each two field picture.
Fig. 6 is the block diagram of the device that the another kind that one embodiment of the disclosure is provided sets up traffic signss data base, such as Fig. 6 institutes
Show, the device is also included apart from acquisition module 504 and image capture module 505.
The navigation module 501, is additionally operable to obtain the positional information of picture pick-up device and friendship to be captured before collection image
The positional information of logical mark.
Apart from acquisition module 504, the position for the positional information according to picture pick-up device and traffic signss to be captured is believed
Breath determines the distance between picture pick-up device and traffic signss to be captured.
Image capture module 505, shoots model for when distance is less than the first setting value, beginning through picture pick-up device collection
Image in enclosing;When distance is less than the second setting value, stop shooting.
Optionally, the positional information of the traffic signss in each two field picture includes the ground of the traffic signss in each two field picture
Reason positional information and elevation information, Fig. 7 are a kind of block diagrams of image processing module that one embodiment of the disclosure is provided, such as Fig. 7 institutes
Show, image processing module 502 includes:Thick matching module 5021 and smart matching module 5022.
Thick matching module 5021 is used for, the geographical location information, elevation information, traffic signss size according to each two field picture
And range information carries out coordinate system transformation, thick match bit of the traffic signss in each two field picture in each two field picture is determined
Put;Thick matched position in each two field picture carries out FIG pull handle, obtains traffic signss in each two field picture slightly matching figure
Picture.
Smart matching module 5022 is used for, and slightly matching image to the traffic signss in each two field picture is carried out at RGB normalization
Reason;Image is slightly matched to the traffic signss carried out after RGB normalizeds of each two field picture carries out color threshold segmentation with true
Traffic signss in fixed each two field picture slightly match the smart matched position that the traffic signss in image are located;To in each two field picture
Smart matched position carry out FIG pull handle, obtain in each two field picture traffic signss essence matching image.
Optionally, image storage module 503 is used for:
The Traffic Sign Images in each two field picture are carried out according to the range information of the traffic signss in each two field picture
Packet, obtains the packet of at least one Traffic Sign Images.
Each Traffic Sign Images packet is labeled, what each Traffic Sign Images was grouped is noted for indicating each
The classification belonging to Traffic Sign Images in Traffic Sign Images packet.
Wherein, illustrating for above-mentioned modules realization of functions has been carried out in detail in said method embodiment
Description, here is omitted.
In sum, the disclosure combines the positional information of traffic signss, traffic signss size and range information, realizes
Figure is scratched by the automatization of traffic signss, by the way of comparing in traditional method using artificial stingy figure, it is possible to increase the acquisition from image
The speed of traffic signss, and the probability for malfunctioning can be reduced.
The preferred implementation of the disclosure is described 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 description and practice
After the disclosure, other embodiments of the disclosure are readily apparent that, the protection domain of the disclosure is belonged to.
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 embodiments of the disclosure
Combination in any can also be carried out, as long as which is without prejudice to the thought of the disclosure, which 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 limited only by appended claim
System.
Claims (10)
1. a kind of method for setting up traffic signss data base, it is characterised in that methods described includes:
Positional information, traffic signss size and the distance for obtaining the traffic signss in each two field picture of picture pick-up device collection is believed
Breath, the range information are used for the distance for representing traffic signss and the picture pick-up device in each two field picture;
Described in positional information, traffic signss size and range information according to the traffic signss in each two field picture determines
Position of the traffic signss in each two field picture in each two field picture, and each frame figure according to the position acquisition
Traffic Sign Images as in;
According to the range information of the traffic signss in the Traffic Sign Images in each two field picture and each two field picture
Generate traffic signss data base.
2. method according to claim 1, it is characterised in that the traffic signss according in each two field picture
Positional information, traffic signss size and range information determine traffic signss in each two field picture in each frame figure
Position as in, and the Traffic Sign Images in each two field picture according to the position acquisition, including:
Positional information, traffic signss size and range information according to the traffic signss in each two field picture determines institute
State position of the traffic signss in each two field picture in each two field picture, and each frame according to the position acquisition
Traffic signss in image slightly match image;
Image is slightly matched to the traffic signss in each two field picture carries out default image procossing, obtains each frame figure
Traffic signss essence matching image as in;
Using the traffic signss essence matching image in each two field picture as the Traffic Sign Images in each two field picture.
3. method according to claim 1, it is characterised in that methods described also includes:
The positional information of the positional information and traffic signss to be captured of the picture pick-up device is obtained before collection image;
Determine that the shooting sets according to the positional information of the positional information and the traffic signss to be captured of the picture pick-up device
It is standby the distance between with the traffic signss to be captured;
When the distance is less than the first setting value, the image in the picture pick-up device collection coverage is begun through;Work as institute
When stating distance less than the second setting value, stop shooting.
4. method according to claim 2, it is characterised in that the positional information of the traffic signss in each two field picture
It is including the geographical location information and elevation information of the traffic signss in each two field picture, described according to each two field picture
In the positional information of traffic signss, traffic signss size and range information determine the traffic signss in each two field picture
Position in each two field picture, and the traffic signss in each two field picture according to the position acquisition slightly match figure
Picture, including:
Geographical location information, elevation information, traffic signss size and range information according to each two field picture is sat
The conversion of mark system, determines thick matched position of the traffic signss in each two field picture in each two field picture;
FIG pull handle is carried out to the thick matched position in each two field picture, the traffic signss in each two field picture are obtained
Thick matching image;
The traffic signss in each two field picture slightly match image carries out default image procossing, obtains described each
Traffic signss essence matching image in two field picture, including:
Image is slightly matched to the traffic signss in each two field picture carries out RGB normalizeds;To each two field picture
The traffic signss carried out after RGB normalizeds slightly match image and carry out color threshold segmentation to determine each frame
Traffic signss in image slightly match the smart matched position that the traffic signss in image are located;
FIG pull handle is carried out to the described smart matched position in each two field picture, the traffic in each two field picture is obtained
Mark essence matching image.
5. the method according to any one of Claims 1-4, it is characterised in that described according in each two field picture
The range information of the traffic signss in Traffic Sign Images and each two field picture generates traffic signss data base, including:
According to the range information of the traffic signss in each two field picture to the Traffic Sign Images in each two field picture
It is grouped, is obtained the packet of at least one Traffic Sign Images;
Each Traffic Sign Images packet is labeled, it is described that what described each Traffic Sign Images were grouped is noted for instruction
The classification belonging to Traffic Sign Images in each Traffic Sign Images packet.
6. a kind of device for setting up traffic signss data base, it is characterised in that described device includes:Navigation module, image procossing
Module, image storage module;
The navigation module, for obtaining positional information, the traffic of the traffic signss in each two field picture that picture pick-up device is gathered
Mark size and range information, the range information are used to representing the traffic signss in each two field picture and the shooting
The distance of equipment;
Described image processing module is big for the positional information according to the traffic signss in each two field picture, traffic signss
Little and range information determines the position of traffic signss in each two field picture in each two field picture, and according to institute
State the Traffic Sign Images in each two field picture described in position acquisition;
Described image memory module, for according in the Traffic Sign Images in each two field picture and each two field picture
Traffic signss range information generate traffic signss data base.
7. device according to claim 6, it is characterised in that described image processing module is used for:
Positional information, the traffic signss size and the distance letter according to the traffic signss in each two field picture
Position of the traffic signss in breath determination each two field picture in each two field picture, and according to the position acquisition institute
State the traffic signss in each two field picture and slightly match image;
Image is slightly matched to the traffic signss in each two field picture carries out default image procossing, obtains each frame figure
Traffic signss essence matching image as in;
Using the traffic signss essence matching image in each two field picture as the Traffic Sign Images in each two field picture.
8. device according to claim 6, it is characterised in that described device also includes:Adopt apart from acquisition module and image
Collection module;
The navigation module, is additionally operable to obtain the positional information of the picture pick-up device and traffic to be captured before collection image
The positional information of mark;
It is described apart from acquisition module, for the positional information according to the picture pick-up device and the position of the traffic signss to be captured
Confidence breath determines the distance between the picture pick-up device and described traffic signss to be captured;
Described image acquisition module, for when the distance is less than the first setting value, beginning through the picture pick-up device collection
Image in coverage;When the distance is less than the second setting value, stop shooting.
9. device according to claim 7, it is characterised in that the positional information of the traffic signss in each two field picture
Including the geographical location information and elevation information of the traffic signss in each two field picture, described image processing module includes:
Thick matching module and smart matching module;
The thick matching module is used for, geographical location information, elevation information according to each two field picture, traffic signss size
And range information carries out coordinate system transformation, determine the traffic signss in each two field picture in each two field picture
Thick matched position;FIG pull handle is carried out to the thick matched position in each two field picture, is obtained in each two field picture
Traffic signss slightly match image;
The smart matching module is used for, and slightly matching image to the traffic signss in each two field picture is carried out at RGB normalization
Reason;Image is slightly matched to the traffic signss carried out after RGB normalizeds of each two field picture carries out color threshold
Split to determine that the traffic signss in each two field picture slightly match the smart matched position at the place of the traffic signss in image;It is right
Described smart matched position in each two field picture carries out FIG pull handle, obtains the traffic signss essence in each two field picture
Matching image.
10. the device according to any one of claim 6 to 9, it is characterised in that described image memory module is used for:
According to the range information of the traffic signss in each two field picture to the Traffic Sign Images in each two field picture
It is grouped, is obtained the packet of at least one Traffic Sign Images;
Each Traffic Sign Images packet is labeled, it is described that what described each Traffic Sign Images were grouped is noted for instruction
The classification belonging to Traffic Sign Images in each Traffic Sign Images packet.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610932080.3A CN106650724A (en) | 2016-10-31 | 2016-10-31 | Method and device for building traffic sign database |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610932080.3A CN106650724A (en) | 2016-10-31 | 2016-10-31 | Method and device for building traffic sign database |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106650724A true CN106650724A (en) | 2017-05-10 |
Family
ID=58820970
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610932080.3A Pending CN106650724A (en) | 2016-10-31 | 2016-10-31 | Method and device for building traffic sign database |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106650724A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021042792A1 (en) * | 2019-09-05 | 2021-03-11 | 华为技术有限公司 | Method and apparatus for determining position of camera |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345630A (en) * | 2013-06-14 | 2013-10-09 | 合肥工业大学 | Traffic sign positioning method based on spherical panoramic video |
CN103925927A (en) * | 2014-04-18 | 2014-07-16 | 中国科学院软件研究所 | Traffic sign positioning method based on vehicle-mounted video |
CN104504904A (en) * | 2015-01-08 | 2015-04-08 | 杭州智诚惠通科技有限公司 | Mobile acquisition method of transportation facility |
CN105320725A (en) * | 2015-05-29 | 2016-02-10 | 杨振贤 | Method and apparatus for acquiring geographic object in collection point image |
-
2016
- 2016-10-31 CN CN201610932080.3A patent/CN106650724A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345630A (en) * | 2013-06-14 | 2013-10-09 | 合肥工业大学 | Traffic sign positioning method based on spherical panoramic video |
CN103925927A (en) * | 2014-04-18 | 2014-07-16 | 中国科学院软件研究所 | Traffic sign positioning method based on vehicle-mounted video |
CN104504904A (en) * | 2015-01-08 | 2015-04-08 | 杭州智诚惠通科技有限公司 | Mobile acquisition method of transportation facility |
CN105320725A (en) * | 2015-05-29 | 2016-02-10 | 杨振贤 | Method and apparatus for acquiring geographic object in collection point image |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021042792A1 (en) * | 2019-09-05 | 2021-03-11 | 华为技术有限公司 | Method and apparatus for determining position of camera |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11474247B2 (en) | Methods and systems for color point cloud generation | |
CN108694882B (en) | Method, device and equipment for labeling map | |
JP7326720B2 (en) | Mobile position estimation system and mobile position estimation method | |
CN107084727B (en) | Visual positioning system and method based on high-precision three-dimensional map | |
WO2020000137A1 (en) | Integrated sensor calibration in natural scenes | |
CN111448591A (en) | System and method for locating a vehicle in poor lighting conditions | |
CN111065043B (en) | System and method for fusion positioning of vehicles in tunnel based on vehicle-road communication | |
EP3650814A1 (en) | Vision augmented navigation | |
JP6950832B2 (en) | Position coordinate estimation device, position coordinate estimation method and program | |
CN112184890B (en) | Accurate positioning method of camera applied to electronic map and processing terminal | |
CN106650705A (en) | Region labeling method and device, as well as electronic equipment | |
CN110858414A (en) | Image processing method and device, readable storage medium and augmented reality system | |
CN113850126A (en) | Target detection and three-dimensional positioning method and system based on unmanned aerial vehicle | |
KR101493418B1 (en) | Method for eliminating shadow of aerial photograph and apparatus thereof | |
CN110515110B (en) | Method, device, equipment and computer readable storage medium for data evaluation | |
CN106227732A (en) | A kind of method of real-time acquisition mobile video photographed scene position | |
CN113240813A (en) | Three-dimensional point cloud information determination method and device | |
CN113137973A (en) | Image semantic feature point truth value determining method and device | |
WO2018098635A1 (en) | Method and system for generating environment model and for positioning using cross-sensor feature point referencing | |
WO2020113425A1 (en) | Systems and methods for constructing high-definition map | |
CN106650724A (en) | Method and device for building traffic sign database | |
CN110554420B (en) | Equipment track obtaining method and device, computer equipment and storage medium | |
CN113009533A (en) | Vehicle positioning method and device based on visual SLAM and cloud server | |
CN115235493A (en) | Method and device for automatic driving positioning based on vector map | |
CN115665553A (en) | Automatic tracking method and device for unmanned aerial vehicle, electronic equipment and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170510 |
|
RJ01 | Rejection of invention patent application after publication |