CN107578483A - A kind of electric non-stop toll method of commerce, server and system - Google Patents
A kind of electric non-stop toll method of commerce, server and system Download PDFInfo
- Publication number
- CN107578483A CN107578483A CN201710719980.4A CN201710719980A CN107578483A CN 107578483 A CN107578483 A CN 107578483A CN 201710719980 A CN201710719980 A CN 201710719980A CN 107578483 A CN107578483 A CN 107578483A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- transacted
- board units
- overlay area
- transaction
- 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
Landscapes
- Traffic Control Systems (AREA)
- Devices For Checking Fares Or Tickets At Control Points (AREA)
Abstract
The present invention relates to a kind of electric non-stop toll method of commerce, including：Image in the camera overlay area is obtained by camera, and the position of vehicle to be transacted in the overlay area is identified based on deep learning；Board units position in the roadside unit transaction area is obtained by roadside unit；Determine whether the vehicle to be transacted is provided with board units according to the position of the position of the vehicle to be transacted and the board units；Judge whether the current transaction vehicle in each vehicle to be transacted is provided with board units, if so, then completing transaction with the board units installed on the current transaction vehicle.The present invention by video depth study with the fusion application of wireless radiofrequency location technology in electronic charging system without parking, avoid with phenomenon of deducting fees caused by car interference and monkey chatter by mistake.
Description
Technical field
The present invention relates to intelligent transportation (Intelligent Transportation System, abbreviation ITS) field, especially
It is related to a kind of electric non-stop toll method of commerce, server and system.
Background technology
Electric non-stop toll technology is a kind of emerging toll collection technique, be by the board units installed on vehicle with
Roadside unit (Road Side in ETC (Electronic Toll Collection, electric non-stop toll) dedicated Lanes
Unit, abbreviation RSU) communicated, realize non-parking charge, have exempt from parking quickly through, exempt cash transaction, simplify charge
Management, the advantages that environmental pollution is reduced, be one of important applied field of intelligent transportation (ITS).ETC system at present, pass through RSU
Communicated with board units (On Board Unit, abbreviation OBU), realization has label vehicle location, but can not realize to without mark
Sign the positioning of vehicle, it is necessary to ordinatedly feel Coil Detector statistics without label vehicle number.Using ground induction coil there is, such as：
Coil lays costly, coil and is not easy adjustment and compatible poor, pass through car speed faster situation inspection big to vehicle flowrate
Survey the not high many deficiencies of positional precision.So causing ETC system antijamming capability poor, can not effectively solve lie interference
With mistake and the fee evasion problem of being deducted fees when being merchandised with ETC caused by car interference phenomenon.
The content of the invention
To solve the above problems, the present invention provides a kind of electric non-stop toll method of commerce, including：
Video identification step：Image in the camera overlay area is obtained by camera, and is based on deep learning
Identify the position of vehicle to be transacted in the overlay area；
Radio frequency identification step：Board units position in the roadside unit transaction area, institute are obtained by roadside unit
Transaction area and the overlay area is stated at least partly to overlap；
Judgment step：The car to be transacted is determined according to the position of the position of the vehicle to be transacted and the board units
Whether board units are installed, and determine the sequencing of each vehicle to be transacted；
Transaction step：Judge whether the current transaction vehicle in each vehicle to be transacted is provided with board units, if so,
Then transaction is completed with the board units installed on the current transaction vehicle.
Further, the video identification step is specially：
The image of the overlay area is demarcated, the position established in the overlay area image and real position
Corresponding relation；
Training pattern is preset into the image input of the overlay area, the position range of vehicle to be transacted is obtained, according to institute
The position that position range tries to achieve the vehicle to be transacted is stated, wherein the position of the vehicle to be transacted is real position.
Further, the image to the overlay area is demarcated, the position established in the overlay area image
Put and be specially with the corresponding relation step of real position：
The point described at least four in overlay area is chosen as calibration point；
Obtain position and real position of at least four calibration point in the overlay area image；
The Transformation Parameters of the position and real position in the overlay area image are obtained by camera calibration method.
Further, before the video identification step, in addition to default training pattern step is established, it is described to establish in advance
If training pattern step is specially to be based on Region CNN algorithms, faster R-CNN methods, YOLO networks or SSD deep learnings
Framework establishes the default training pattern.
Further, it is described the default training pattern is established based on SSD deep learning frameworks to be specially：
The image in collection overlay area generates a plurality of training image data in advance, and wherein described image data are included just
Sample and negative sample；
The training image data are pre-processed, wherein pretreatment is that the vehicle in the training image data is entered
Row is confined and/or marked；
Deep learning training is carried out to a plurality of pretreated training image data, i.e., point-to-point loss function
The renewal that calculating and backpropagation calculate, finally gives and obtains the default training pattern.
Further, the image by the overlay area inputs default training pattern, obtains the position of vehicle to be transacted
Scope is put, the position step that the vehicle to be transacted is tried to achieve according to the position range is specially：
The image of the overlay area is inputted into the default training pattern, obtains the vehicle to be transacted in the covering
The fractional value prob of position range and prediction result in area image, when fractional value prob is not less than preset value, then basis
The position range calculates the position of the vehicle to be transacted.
Further, the identification step includes：
Verification step：If the position of any vehicle to be transacted includes the position of the board units, waiting to hand over described in judgement
Easy vehicle is provided with board units；
The verification step is performed respectively according to each board units in the transaction area successively, determines the trading post
Whether each vehicle to be transacted in the range of domain and camera overlay area coincidence is provided with board units；
The elder generation of the vehicle to be transacted is determined according to the position of the position of the board units and/or the vehicle to be transacted
Order afterwards.
Another aspect, the invention also discloses a kind of highway electric non-stop toll trading server, including：
Video identification module, for obtaining the image in the camera overlay area by camera, and it is based on depth
Study identifies the position of vehicle to be transacted in the overlay area；
Radio frequency identification module, for obtaining the board units position in the roadside unit transaction area by roadside unit
Put, the transaction area and the overlay area at least partly overlap；
Judge module, for waiting to hand over described in the position according to the vehicle to be transacted and the determination of the position of the board units
Whether easy vehicle is provided with board units, and determines the sequencing of each vehicle to be transacted；
Transaction modules, for judging whether the current transaction vehicle in each vehicle to be transacted is provided with board units,
If so, then complete transaction with the board units installed on the current transaction vehicle.
On the other hand, present invention also offers a kind of highway electric non-stop toll trading server, including place
Device and memory are managed, the memory instructs for storage program, so that processor loading described program instruction, is completed such as
Upper described method and step.
Another aspect, additionally provide a kind of highway electric non-stop toll transaction system, including camera, trackside
The transaction area of the coverage of unit, wherein camera and the roadside unit at least partly overlaps, the camera and institute
Roadside unit is stated respectively with server as described above to be connected.
Implement technical scheme, video depth can be learnt with the fusion application of wireless radiofrequency location technology in
In electronic charging system without parking, avoid with deducting fees phenomenon caused by car interference and monkey chatter by mistake, it is ensured that system is more increased
Effect, economic, accurately work.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.In accompanying drawing：
Fig. 1 is application scenarios schematic diagram；
Fig. 2 is a kind of flow chart of highway electric non-stop toll method of commerce of the present invention.
Embodiment
The present invention is applied to ETC system as shown in Figure 1, and the system includes camera 2, roadside unit 3 and server 1,
Wherein server 1 can be the industrial computer connected respectively with camera 2, roadside unit 3.Camera 2 and roadside unit 3 are using dragon
Door frame is arranged on above ETC tracks, the image (fan section that dotted line represents in figure that camera 2 is used to shoot in its overlay area
Domain) and it is transferred to server 1；Roadside unit 3 under the control of server 1 with transaction area (in figure dotted line represent ellipse
Shape region) board units (not shown in figure, wherein OBU1 and OBU2 are the vehicle for being provided with board units) communication, and to car
Carrier unit is positioned, and obtains the information of vehicles in board units.When OBU1 vehicles turn into direct of travel on most it is forward not yet
When completing the vehicle to be transacted of transaction, roadside unit 3 is completed to merchandise with the board units that OBU1 vehicles carry, and completion of letting pass is handed over
Easy OBU1 vehicles.Wherein transaction area refers to that roadside unit can complete the region of board units communication, still incomplete transaction
Both included not yet completing ETC transaction, and also do not completed and manually wait other transaction.The system, which is removed, can be used for highway ETC tracks
Outside, also there is application prospect in fields such as city free flows.
Embodiment 1, based on said system, the present invention provides a kind of highway electric non-stop toll method of commerce, bag
Include：
Video identification step：Image in the overlay area of camera 2 is obtained by camera 2, and is based on depth
Practise the position for identifying vehicle to be transacted in the overlay area；It is readily appreciated that, server 1 and camera 2 connect, and receive shooting
It is first 2 shooting overlay areas image, when need to be merchandised on ETC tracks vehicle when, camera photographs vehicle to be transacted, then
The vehicle to be transacted in image is identified based on deep learning by server 1, and calculates the real position of vehicle to be transacted.
Radio frequency identification step：Board units position in the roadside unit transaction area, institute are obtained by roadside unit 3
Transaction area and the overlay area is stated at least partly to overlap；It should be understood that transaction area and the overlay area of camera by
It is different in caused equipment, therefore will not typically be completely superposed.In the present embodiment, transaction area and overlay area intersection
Including an ETC tracks part, the part in the ETC tracks comprises at least roadside unit 3 and board units are completed corresponding to transaction
Position.It is readily appreciated that, intersection can also include longer ETC tracks, to be transacted for shooting and positioning some
Vehicle.In order to ensure the accuracy of board units position acquisition, roadside unit 3 can be phased array roadside unit.
Judgment step：The car to be transacted is determined according to the position of the position of the vehicle to be transacted and the board units
Whether board units are installed, and determine the sequencing of each vehicle to be transacted；When vehicle to be transacted is more,
On direction of traffic headed by vehicle of currently merchandising, the precedence of each vehicle to be transacted is recorded successively.
Transaction step：Judge whether the current transaction vehicle in each vehicle to be transacted is provided with board units, if so,
Then transaction is completed with the board units installed on the current transaction vehicle.Identify the precedence of each vehicle to be transacted and be
It is no board units are installed after, if it is determined that currently transaction vehicle board units are installed, then server 1 control roadside unit 3
It is traded with the board units on vehicle, current transaction vehicle of being let pass after deducting fees, it is readily appreciated that, in each current trade car
Complete transaction after, other vehicles to be transacted are advanced successively, and most forward still incomplete transaction waits to hand over wherein on direct of travel
Easy vehicle is current transaction vehicle.If it is determined that it is fitted without board units, then uses and manually deduct fees, or warn it to roll away from
ETC tracks.Further, outside above-mentioned function, due to roadside unit 3 with can obtain car in board units process of exchange
The information of vehicles recorded in the ID and board units of carrier unit in itself, the image of further camera collection can also parse
Some vehicles information, therefore may also be used for verifying charge, prevent from charging by mistake.
The present invention treats vehicle of merchandising using video depth study and detected, in combination with the wireless location of roadside unit 3,
Wherein deep learning algorithm positions to the vehicle Real time identification to be transacted on ETC tracks, has both included the to be transacted of board units
Vehicle, also include the vehicle to be transacted without board units.Positioning result is identified by deep learning, can accurately determine ETC
The position relationship of each car in vehicle platoon to be transacted on track, and the result of the wireless location of roadside unit 3 is combined, can
To judge each car in vehicle platoon to be transacted whether there is board units situation.Can efficiently solve what vehicle flowrate increase triggered
There is the problem of monkey chatter with car interference problem and in multilane in the same direction application.
Embodiment 2, it is deep in the positioning result that the two methods of video depth study and wireless location of roadside unit 3 obtain
The position that degree learning outcome belongs in image, and the result of roadside unit 3 belongs to real position, in order to merge two kinds of localization methods
Result, on the basis of embodiment 1, the video identification step is specially：
The image of the overlay area is demarcated, the position established in the overlay area image and real position
Corresponding relation；
Training pattern is preset into the image input of the overlay area, the position range of vehicle to be transacted is obtained, according to institute
The position that position range tries to achieve the vehicle to be transacted is stated, wherein the position of the vehicle to be transacted is real position.
It is readily appreciated that, real position is physical location information of the vehicle to be transacted on the horizontal plane of ETC tracks, and unit is
Rice；The vehicle location scope to be transacted that positioning acquisition is identified by image in deep learning is to be transacted under image coordinate system
Pixel coverage shared by vehicle, unit are pixels.Therefore need to demarcate the image of camera 2, establish itself and reality
The corresponding relation of position：
Specifically, the point described in selection at least four in overlay area is as calibration point；
Obtain position and real position of at least four calibration point in the overlay area image；Namely obtain
Coordinate of all calibration points in image coordinate system in point coordinates and reality, as long as the selection of coordinate system can complete corresponding function i.e.
Can, it is not specifically limited herein.
Position and reality in the overlay area image is calculated in the calibration matrix derived by camera calibration method
The Transformation Parameters of position.
Pass through Transformation Parameters, you can to facilitate position in the image of overlay area and real position to be changed.
Embodiment 3, on the basis of embodiment 2, before the video identification step, in addition to establish default training mould
Type step, the default training pattern step of foundation is specially to be based on Region CNN (Region-based
Convolutional Neural Network, region convolutional neural networks) algorithm, faster R-CNN (FasterRegion-
Based Convolutional Neural Network, fast area convolutional neural networks algorithm) method, YOLO (You
Only Look Once) network or SSD deep learning frameworks establish the default training pattern.
The core of SSD (Single Deep Nerual Network) deep learning framework is prediction object, and is calculated
The score of its belonging kinds.SSD is to propagate CNN networks forward based on one, by inputting single-frame images data, according to depth
Practise training mode to calculate, export a series of bounding box of fixed sizes, and each frame includes the possibility of object example
Property, i.e. fraction.By carrying out a non-maxima suppression, final prediction result is obtained.SSD methods are in detection time, detection
There is relatively good performance in precision.The selection SSD deep learning frameworks establish the default training pattern, are specially：
The image in collection overlay area generates a plurality of training image data in advance, and wherein described image data are included just
Sample and negative sample；It should be understood that the number of training image data directly affects the accuracy of calculating, therefore should not be very few,
Selection positive sample 2000 is opened herein opens with negative sample 1000, and positive sample is the sample for including vehicle to be transacted, and negative sample is not wrap
Sample containing vehicle to be transacted.
The training image data are pre-processed, wherein pretreatment is that the vehicle in the training image data is entered
Row is confined and/or marked；
Deep learning training is carried out to a plurality of pretreated training image data, i.e., point-to-point loss function
The renewal that calculating and backpropagation calculate, finally gives and obtains the default training pattern.
Further, the image by the overlay area inputs default training pattern, obtains the position of vehicle to be transacted
Scope is put, the position step that the vehicle to be transacted is tried to achieve according to the position range is specially：
The image of the overlay area is inputted into the default training pattern, obtains the vehicle to be transacted in the covering
The fractional value prob of position range and prediction result in area image, when fractional value prob is not less than preset value, then basis
The position range calculates the position of the vehicle to be transacted.
It should be understood that using each two field picture gathered in real time as input, calculated by default training pattern, you can
Into image vehicle to be transacted image in position range, be expressed as position range bottom right angular coordinate and upper left in the picture
Angular coordinate (unit pixel), and the bigger explanation of the fractional value prob of the prediction result, fractional value prob are the possibilities of vehicle
It is bigger, during general fraction prob >=0.68 (full marks are 1.0 points), it is taken as vehicle to be transacted.SSD methods can simultaneously handle
All vehicles to be transacted in the image are all identified, detection and localization, then according to the position of all vehicles to be transacted of acquisition
Put scope and to be transacted vehicle location of the be described vehicle to be transacted on ETC tracks is finally tried to achieve by Transformation Parameters.
Further, the judgment step includes：
Verification step：If the position of any vehicle to be transacted includes the position of the board units, waiting to hand over described in judgement
Easy vehicle is provided with board units；It should be understood that the position of board units, which is roadside unit 3, utilizes DSRC (Dedicated
Short Range Communication, abbreviation DSRC) technology communicated with the board units in communication zone, acquisition
The position of board units position, wherein board units is point value.And the vehicle location in image is through default training pattern identification
It is a value range, i.e. position range, and board units are mounted on vehicle to be transacted, if so board units position
If point value fall into the position range of vehicle to be transacted, illustrate to be mounted with board units on vehicle to be transacted.
The verification step is performed respectively according to each board units in the transaction area successively, determines the trading post
Whether each vehicle to be transacted in the range of domain and camera overlay area coincidence is provided with board units；
The elder generation of the vehicle to be transacted is determined according to the position of the position of the board units and/or the vehicle to be transacted
Order afterwards.
It should be understood that except it is above-mentioned picture position is converted into real position in addition to, can also unify vehicle to be transacted
The position being converted into the position of board units in image, essence is identical, specifically, after Transformation Parameters can also being obtained, by road
The real position (unit, rice) of side unit 3RSU wireless locations acquisition switchs to the position (unit, pixel) in image, eventually through two
Kind distinct methods obtain the position of vehicle and board units to be transacted in the picture respectively, synchronous in data fusion module, lead to
Cross and calculate the distance that each truck position completes position corresponding to exchange relative to roadside unit 3 and board units, just
It may determine that each vehicle to be transacted goes out the position relationship on ETC tracks.
Embodiment 4, the invention also discloses a kind of highway electric non-stop toll trading server 1, including：
Video identification module, for obtaining the image in the overlay area of camera 2 by camera 2, and based on deep
Degree study identifies the position of vehicle to be transacted in the overlay area；
Radio frequency identification module, for obtaining the board units position in the transaction area of roadside unit 3 by roadside unit 3
Put, the transaction area and the overlay area at least partly overlap；
Judge module, for waiting to hand over described in the position according to the vehicle to be transacted and the determination of the position of the board units
Whether easy vehicle is provided with board units, and determines the sequencing of each vehicle to be transacted；
Transaction modules, for judging whether the current transaction vehicle in each vehicle to be transacted is provided with board units,
If so, then complete transaction with the board units installed on the current transaction vehicle.
On the other hand, present invention also offers a kind of highway electric non-stop toll trading server 1, including place
Device and memory are managed, the memory instructs for storage program, so that processor loading described program instruction, is completed such as
Method and step described in embodiment 1-3.
Another aspect, as shown in figure 1, present invention also offers a kind of highway electric non-stop toll transaction system
System, including camera 2, roadside unit 3, the wherein transaction area of the coverage of camera 2 and the roadside unit 3 at least portion
Divide and overlap, the camera 2 is connected with server 1 as described above respectively with the roadside unit 3.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, any bun made
Change, equivalent substitution, improvement etc., should be included within scope of the presently claimed invention.
Claims (10)
- A kind of 1. electric non-stop toll method of commerce, it is characterised in that including：Video identification step：Image in the camera overlay area is obtained by camera, and identified based on deep learning The position of vehicle to be transacted in the overlay area；Radio frequency identification step：Board units position in the roadside unit transaction area, the friendship are obtained by roadside unit Easy region and the overlay area at least partly overlap；Judgment step：Determine that the vehicle to be transacted is according to the position of the position of the vehicle to be transacted and the board units It is no that board units are installed；Transaction step：Judge whether the current transaction vehicle in each vehicle to be transacted is provided with board units, if so, then and The board units installed on the current transaction vehicle complete transaction.
- 2. according to the method for claim 1, it is characterised in that the video identification step is specially：The image of the overlay area is demarcated, the position established in the overlay area image and the correspondence of real position Relation；Training pattern is preset into the image input of the overlay area, the position range of vehicle to be transacted is obtained, according to institute's rheme The position that scope tries to achieve the vehicle to be transacted is put, wherein the position of the vehicle to be transacted is real position.
- 3. according to the method for claim 2, it is characterised in that the image to the overlay area is demarcated, and is built Stand the position in the overlay area image and the corresponding relation step of real position is specially：The point described at least four in overlay area is chosen as calibration point；Obtain position and real position of at least four calibration point in the overlay area image；The Transformation Parameters of the position and real position in the overlay area image are obtained by camera calibration method.
- 4. according to the method for claim 2, it is characterised in that before the video identification step, in addition to establish pre- It is described to establish default training pattern step specially based on Region CNN algorithms, faster R-CNN if training pattern step Method, YOLO networks or SSD deep learning frameworks establish the default training pattern.
- 5. according to the method for claim 4, it is characterised in that described that described preset is established based on SSD deep learning frameworks Training pattern is specially：The image in collection overlay area generates a plurality of training image data in advance, and wherein described image data include positive sample And negative sample；The training image data are pre-processed, wherein pretreatment is to carry out frame to the vehicle in the training image data Fixed and/or mark；Deep learning training, i.e., the calculating of point-to-point loss function are carried out to a plurality of pretreated training image data And the renewal that backpropagation calculates, finally give and obtain the default training pattern.
- 6. according to the method for claim 5, it is characterised in that the default training of image input by the overlay area Model, the position range of vehicle to be transacted is obtained, the position step tool of the vehicle to be transacted is tried to achieve according to the position range Body is：The image of the overlay area is inputted into the default training pattern, obtains the vehicle to be transacted in the overlay area The fractional value prob of position range and prediction result in image, when fractional value prob is not less than preset value, then according to Position range calculates the position of the vehicle to be transacted.
- 7. according to the method for claim 6, it is characterised in that the judgment step includes：Verification step：If the position of any vehicle to be transacted includes the position of the board units, the car to be transacted is judged Board units are installed；Perform the verification step respectively according to each board units in the transaction area successively, determine the transaction area and Whether each vehicle to be transacted in the range of the camera overlay area coincidence is provided with board units；Determine that the priority of the vehicle to be transacted is suitable according to the position of the position of the board units and/or the vehicle to be transacted Sequence.
- A kind of 8. highway electric non-stop toll trading server, it is characterised in that including：Video identification module, for obtaining the image in the camera overlay area by camera, and it is based on deep learning Identify the position of vehicle to be transacted in the overlay area；Radio frequency identification module, for obtaining the board units position in the roadside unit transaction area, institute by roadside unit Transaction area and the overlay area is stated at least partly to overlap；Judge module, the car to be transacted is determined for the position according to the vehicle to be transacted and the position of the board units Whether board units are installed；Transaction modules, for judging whether the current transaction vehicle in each vehicle to be transacted is provided with board units, if so, Then transaction is completed with the board units installed on the current transaction vehicle.
- 9. a kind of highway electric non-stop toll trading server, it is characterised in that described including processor and memory Memory instructs for storage program, so that processor loading described program instruction, is completed such as any one of claim 1-7 Described method and step.
- A kind of 10. highway electric non-stop toll transaction system, it is characterised in that including camera, roadside unit, wherein The transaction area of the coverage of camera and the roadside unit at least partly overlaps, the camera and the roadside unit It is connected respectively with server as claimed in claim 8 or 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710719980.4A CN107578483A (en) | 2017-08-21 | 2017-08-21 | A kind of electric non-stop toll method of commerce, server and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710719980.4A CN107578483A (en) | 2017-08-21 | 2017-08-21 | A kind of electric non-stop toll method of commerce, server and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107578483A true CN107578483A (en) | 2018-01-12 |
Family
ID=61034588
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710719980.4A Pending CN107578483A (en) | 2017-08-21 | 2017-08-21 | A kind of electric non-stop toll method of commerce, server and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107578483A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108717789A (en) * | 2018-06-28 | 2018-10-30 | 深圳市金溢科技股份有限公司 | A kind of the acquisition mask method and device of vehicle sample |
| CN108765603A (en) * | 2018-04-02 | 2018-11-06 | 江苏数慧信息科技有限公司 | Toll collection system |
| CN108986465A (en) * | 2018-07-27 | 2018-12-11 | 深圳大学 | A kind of method of vehicle Flow Detection, system and terminal device |
| CN110956709A (en) * | 2019-11-26 | 2020-04-03 | 广州铭创通讯科技有限公司 | ETC phased array antenna system for highway toll collection |
| CN113129460A (en) * | 2021-03-17 | 2021-07-16 | 深圳成谷软件有限公司 | Method for determining driving direction of vehicle in intelligent traffic system and vehicle-mounted unit |
| CN113487300A (en) * | 2021-04-29 | 2021-10-08 | 杭州恒生芸擎网络科技有限公司 | Safety verification method and device for expense data earmark process and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101350109A (en) * | 2008-09-05 | 2009-01-21 | 交通部公路科学研究所 | Method for locating and controlling multilane free flow video vehicle |
| CN102622783A (en) * | 2012-03-29 | 2012-08-01 | 北京速通科技有限公司 | Multilane free-flow electronic toll collection method based on accurate position match |
| CN102622784A (en) * | 2012-03-29 | 2012-08-01 | 北京速通科技有限公司 | Method for realizing multilane free stream electronic charging through orthogonal phase control array antenna |
| CN103456047A (en) * | 2013-09-04 | 2013-12-18 | 广州华工信息软件有限公司 | Electronic toll collection lane system and method based on dynamic vehicle detection technology |
| CN103729892A (en) * | 2013-06-20 | 2014-04-16 | 深圳市金溢科技有限公司 | Vehicle positioning method and device and processor |
| CN104574540A (en) * | 2014-12-24 | 2015-04-29 | 北京速通科技有限公司 | Electronic toll collection system and method |
-
2017
- 2017-08-21 CN CN201710719980.4A patent/CN107578483A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101350109A (en) * | 2008-09-05 | 2009-01-21 | 交通部公路科学研究所 | Method for locating and controlling multilane free flow video vehicle |
| CN102622783A (en) * | 2012-03-29 | 2012-08-01 | 北京速通科技有限公司 | Multilane free-flow electronic toll collection method based on accurate position match |
| CN102622784A (en) * | 2012-03-29 | 2012-08-01 | 北京速通科技有限公司 | Method for realizing multilane free stream electronic charging through orthogonal phase control array antenna |
| CN103729892A (en) * | 2013-06-20 | 2014-04-16 | 深圳市金溢科技有限公司 | Vehicle positioning method and device and processor |
| CN103456047A (en) * | 2013-09-04 | 2013-12-18 | 广州华工信息软件有限公司 | Electronic toll collection lane system and method based on dynamic vehicle detection technology |
| CN104574540A (en) * | 2014-12-24 | 2015-04-29 | 北京速通科技有限公司 | Electronic toll collection system and method |
Non-Patent Citations (1)
| Title |
|---|
| 马慧彬: "《基于机器学习的乳腺图像辅助诊断算法研究》", 31 August 2016 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108765603A (en) * | 2018-04-02 | 2018-11-06 | 江苏数慧信息科技有限公司 | Toll collection system |
| CN108717789A (en) * | 2018-06-28 | 2018-10-30 | 深圳市金溢科技股份有限公司 | A kind of the acquisition mask method and device of vehicle sample |
| CN108986465A (en) * | 2018-07-27 | 2018-12-11 | 深圳大学 | A kind of method of vehicle Flow Detection, system and terminal device |
| CN110956709A (en) * | 2019-11-26 | 2020-04-03 | 广州铭创通讯科技有限公司 | ETC phased array antenna system for highway toll collection |
| CN113129460A (en) * | 2021-03-17 | 2021-07-16 | 深圳成谷软件有限公司 | Method for determining driving direction of vehicle in intelligent traffic system and vehicle-mounted unit |
| CN113129460B (en) * | 2021-03-17 | 2024-01-02 | 深圳成谷科技有限公司 | Method for determining driving direction of vehicle in intelligent traffic system and vehicle-mounted unit |
| CN113487300A (en) * | 2021-04-29 | 2021-10-08 | 杭州恒生芸擎网络科技有限公司 | Safety verification method and device for expense data earmark process and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107578483A (en) | A kind of electric non-stop toll method of commerce, server and system | |
| CN103886647B (en) | The Fare Collection System of compatible ETC and MTC based on the Big Dipper, device and method | |
| CN104064031B (en) | The vehicle peccancy monitoring of a kind of applying electronic car plate and tracking location system | |
| Zhang et al. | A traffic surveillance system for obtaining comprehensive information of the passing vehicles based on instance segmentation | |
| CN108229307A (en) | For the method, apparatus and equipment of object detection | |
| CN102622783B (en) | Multilane free-flow electronic toll collection method based on accurate position match | |
| CN104021594B (en) | Vehicle on highway based on the Big Dipper freely flows E-payment system, device and method | |
| CN109064495A (en) | A kind of bridge floor vehicle space time information acquisition methods based on Faster R-CNN and video technique | |
| CN110189424A (en) | Multilane free-flow vehicle detection method and system based on multiple target radar | |
| CN105489053B (en) | A kind of parallel shutdown system construction method based on ACP methods | |
| CN106846870A (en) | The intelligent parking system and method for the parking lot vehicle collaboration based on centralized vision | |
| CN104574540A (en) | Electronic toll collection system and method | |
| CN103500472B (en) | Based on the expressway non-stop charging method of dipper system | |
| CN109410623A (en) | Parking navigation method and device | |
| CN103761877B (en) | Based on the red-lamp running automatic recording system and method that radar triggers | |
| CN103150772A (en) | Multi-lane free flow electronic toll collection system arranged on side of road | |
| CN104021593B (en) | Vehicle electric charging method based on global position system, system and device | |
| CN104881897B (en) | ETC tracks vehicle detecting system and detection method | |
| CN109782364B (en) | Traffic sign board missing detection method based on machine vision | |
| CN109637137A (en) | Traffic control system based on bus or train route collaboration | |
| CN111260808B (en) | Free flow vehicle charging device, system and method based on multi-data fusion | |
| CN110379168A (en) | A kind of vehicular traffic information acquisition method based on Mask R-CNN | |
| CN108154146A (en) | A kind of car tracing method based on image identification | |
| CN109544710A (en) | A kind of intelligent road side cellular system | |
| CN206489633U (en) | A kind of chargeable service device |
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: 20180112 |
|
| RJ01 | Rejection of invention patent application after publication |