CN108052866A - Car license recognition learning method and system based on artificial neural network - Google Patents
Car license recognition learning method and system based on artificial neural network Download PDFInfo
- Publication number
- CN108052866A CN108052866A CN201711146253.XA CN201711146253A CN108052866A CN 108052866 A CN108052866 A CN 108052866A CN 201711146253 A CN201711146253 A CN 201711146253A CN 108052866 A CN108052866 A CN 108052866A
- Authority
- CN
- China
- Prior art keywords
- car
- license recognition
- car license
- neural network
- artificial neural
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention provides a kind of Car license recognition learning method and system based on artificial neural network, including：Using each Car license recognition instrument as artificial neuron, the artificial neural network of Vehicle License Plate Recognition System is made up of network, sample car plate is provided to Car license recognition instrument and is learnt to obtain experience table, according to the experience table Forecasting recognition result；By regression analysis, understand unified car plate multiple characters or multiple car plates multiple characters whether related, related direction and intensity, founding mathematical models；By verifying target license plate, the response variables of mathematical model is adjusted with explaining parameter；By classification analysis, according to the feature of known sample car plate, judge which kind of known sample class new sample car plate belongs to, select characteristic parameter by calculating, establish discriminant function, classify to sample car plate.The present invention can accurately identify effective identity of vehicle, reduce the requirement to identification distance and environment, save installation maintenance cost.
Description
Technical field
The present invention relates to vehicle recongnition technique fields, and in particular, to a kind of Car license recognition based on artificial neural network
Learning method and system.
Background technology
In existing Vehicle License Plate Recognition System, car plate in fine day and the overcast and rainy, weather that snows be visually it is different, together
When, car plate species itself is more, and shape, color, size etc. are all had nothing in common with each other, and distance difference, the shooting angle of shooting point are different
Final car plate result will be influenced.
Machine learning is a branch of artificial intelligence.Substantial amounts of statistics involved in learning algorithm, Approximation Theory, convex point
The multi-door subject such as analysis, computational complexity theory.Neutral net is coupled by substantial amounts of artificial neuron to be calculated.In neutral net
Neuron connection weight and neuron excitation value be variation；How excitation function defines neuron according to other nerves
The activity of member changes the excitation value of oneself；How the weight that learning rules are specified in network adjusts with time stepping method.
How machine learning is carried out to row Car license recognition, realized under all weather conditions, different installing environmental conditionses to difference
It is the technical issues of this field is badly in need of solving that car plate, which accurately identify,.
The content of the invention
For in the prior art the defects of, the object of the present invention is to provide a kind of Car license recognitions based on artificial neural network
Learning method and system.
A kind of Car license recognition learning method based on artificial neural network provided according to the present invention, including step：
Artificial neural network construction step：Using each Car license recognition instrument as artificial neuron, car plate is formed by network
The artificial neural network of identifying system provides Car license recognition instrument sample car plate and is learnt to obtain experience table, according to the warp
Test table Forecasting recognition result；
Mathematical model establishment step：By regression analysis, understand unified car plate multiple characters or multiple car plates it is more
Whether a character is related, related direction and intensity, founding mathematical models；
Mathematical model set-up procedure：By verifying target license plate, the response variables of mathematical model is adjusted with explaining parameter；
Classification analysis step：By classification analysis, according to the feature of known sample car plate, judge that new sample car plate belongs to
Which kind of known sample class, selects characteristic parameter by calculating, establishes discriminant function, classify to sample car plate.
Preferably, it further includes：
Weight set-up procedure：Specify the weight of Car license recognition instrument connection, the Car license recognition instrument of Same Site is mutual
Connection weight is higher than the connection weight between the Car license recognition instrument in different places.
Preferably, it further includes：
Excitation value set-up procedure：Adjust the excitation value of Car license recognition instrument according to vehicle flowrate, the excitation value of Car license recognition instrument with
The increase of vehicle flowrate and increase, reduced with the reduction of vehicle flowrate.
A kind of Car license recognition learning system based on artificial neural network provided according to the present invention, including：
Artificial neural network sets up module：Using each Car license recognition instrument as artificial neuron, car plate is formed by network
The artificial neural network of identifying system provides Car license recognition instrument sample car plate and is learnt to obtain experience table, according to the warp
Test table Forecasting recognition result；
Mathematical model establishes module：By regression analysis, understand unified car plate multiple characters or multiple car plates it is more
Whether a character is related, related direction and intensity, founding mathematical models；
Mathematical model adjusts module：By verifying target license plate, the response variables of mathematical model is adjusted with explaining parameter；
Classification analysis module：By classification analysis, according to the feature of known sample car plate, judge that new sample car plate belongs to
Which kind of known sample class, selects characteristic parameter by calculating, establishes discriminant function, classify to sample car plate.
Preferably, it further includes：
Weight adjusts module：Specify the weight of Car license recognition instrument connection, the Car license recognition instrument of Same Site is mutual
Connection weight is higher than the connection weight between the Car license recognition instrument in different places.
Preferably, it further includes：
Excitation value adjusts module：Adjust the excitation value of Car license recognition instrument according to vehicle flowrate, the excitation value of Car license recognition instrument with
The increase of vehicle flowrate and increase, reduced with the reduction of vehicle flowrate.
Compared with prior art, the present invention has following advantageous effect：
1st, the Car license recognition study based on artificial neural network, reduces the requirement to identification distance and environment, saves peace
Fill maintenance cost.
2nd, improve license plate recognition result, efficiently identify vehicle identification, accurately control the disengaging of vehicle, reduce management into
This.
3rd, effective identity of vehicle is accurately identified, is automatically performed and opens a sluice gate, improves the actual use experience of user.
Description of the drawings
Upon reading the detailed description of non-limiting embodiments with reference to the following drawings, other feature of the invention,
Objects and advantages will become more apparent upon：
Fig. 1 is the flow chart of the present invention；
Fig. 2 is the principle of the present invention figure.
Specific embodiment
With reference to specific embodiment, the present invention is described in detail.Following embodiment will be helpful to the technology of this field
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill to this field
For personnel, without departing from the inventive concept of the premise, several changes and improvements can also be made.These belong to the present invention
Protection domain.
As shown in Figure 1, 2, a kind of Car license recognition learning method based on artificial neural network provided by the invention, including step
Suddenly：
Artificial neural network construction step：Using each Car license recognition instrument as artificial neuron, car plate is formed by network
The artificial neural network of identifying system provides Car license recognition instrument sample car plate and is learnt to obtain experience table, according to the warp
Test table Forecasting recognition result.The parameter of typing Car license recognition instrument, property, above and below ground and environmental parameter.
Mathematical model establishment step：By regression analysis, understand unified car plate multiple characters or multiple car plates it is more
Whether a character is related, related direction and intensity, founding mathematical models.Given input identification character exports vehicle as independent variable
Board character establishes multiple linear regression mathematical model as dependent variable.
Mathematical model set-up procedure：By verifying target license plate, the response variables of mathematical model is adjusted with explaining parameter, i.e.,
Regression coefficient is adjusted, represents influence degree of the input character to output character.
Classification analysis step：By classification analysis, according to the feature of known sample car plate, judge that new sample car plate belongs to
Which kind of known sample class, selects characteristic parameter by calculating, establishes discriminant function, classify to sample car plate.Analysis machine
Motor-car number plate coding rule establishes the regular expression of various car plates.
Weight set-up procedure：Specify the weight of Car license recognition instrument connection, the Car license recognition instrument of Same Site is mutual
Connection weight is higher than the connection weight between the Car license recognition instrument of different places (such as on the ground with low).
Excitation value set-up procedure：Adjust the excitation value of Car license recognition instrument according to vehicle flowrate, the excitation value of Car license recognition instrument with
The increase of vehicle flowrate and increase, reduced with the reduction of vehicle flowrate.By defining excitation function, how to determine an identifier
The excitation value of oneself is changed according to the activity of other identifiers.When the vehicle flowrate of the identifier of entrance becomes larger, excitation function is just
The excitation value of increase outlet identifier.
How learning rules define weight in network as time stepping method adjusts.Enhancing study passes through environment of observation light
Line inductor, the feedback of the ambient enviroment arrived according to the observation are analyzed in the experience table of slave phase near-ambient.
By there is the study of supervision, the confidence level of license plate recognition result constantly learns, reductive analysis empirical value.Credible
In the case that degree is high, result is adopted；Car plate is made in the case of with a low credibility and reasonably being speculated.
Based on a kind of above-mentioned Car license recognition learning method based on artificial neural network, the present invention also provides one kind to be based on people
The Car license recognition learning system of artificial neural networks, including：
Artificial neural network sets up module：Using each Car license recognition instrument as artificial neuron, car plate is formed by network
The artificial neural network of identifying system provides Car license recognition instrument sample car plate and is learnt to obtain an experience table, according to institute
State experience table Forecasting recognition result；
Mathematical model establishes module：By regression analysis, understand unified car plate multiple characters or multiple car plates it is more
Whether a character is related, related direction and intensity, founding mathematical models；
Mathematical model adjusts module：By verifying target license plate, the response variables of mathematical model is adjusted with explaining parameter；
Classification analysis module：By classification analysis, according to the feature of known sample car plate, judge that new sample car plate belongs to
Which kind of known sample class, selects characteristic parameter by calculating, establishes discriminant function, classify to sample car plate.
Weight adjusts module：Specify the weight of Car license recognition instrument connection, the Car license recognition instrument of Same Site is mutual
Connection weight is higher than the connection weight between the Car license recognition instrument in different places.
Excitation value adjusts module：Adjust the excitation value of Car license recognition instrument according to vehicle flowrate, the excitation value of Car license recognition instrument with
The increase of vehicle flowrate and increase, reduced with the reduction of vehicle flowrate.
One skilled in the art will appreciate that except realizing system provided by the invention in a manner of pure computer readable program code
It, completely can be by the way that method and step progress programming in logic be provided come the present invention and its beyond each device, module, unit
System and its each device, module, unit with logic gate, switch, application-specific integrated circuit, programmable logic controller (PLC) and embedding
Enter the form of the controller that declines etc. to realize identical function.So system provided by the invention and its every device, module, list
Member is considered a kind of hardware component, and the device for being used to implement various functions, module, unit to including in it also may be used
To be considered as the structure in hardware component；The device for being used to implement various functions, module, unit can also be considered as either real
The software module of existing method can be the structure in hardware component again.
Specific embodiments of the present invention are described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make a variety of changes or change within the scope of the claims, this not shadow
Ring the substantive content of the present invention.In the case where there is no conflict, the feature in embodiments herein and embodiment can arbitrary phase
Mutually combination.
Claims (6)
1. a kind of Car license recognition learning method based on artificial neural network, which is characterized in that including step：
Artificial neural network construction step：Using each Car license recognition instrument as artificial neuron, Car license recognition is formed by network
The artificial neural network of system provides Car license recognition instrument sample car plate and is learnt to obtain experience table, according to the experience table
Forecasting recognition result；
Mathematical model establishment step：By regression analysis, multiple characters of unified car plate or multiple words of multiple car plates are understood
Whether symbol is related, related direction and intensity, founding mathematical models；
Mathematical model set-up procedure：By verifying target license plate, the response variables of mathematical model is adjusted with explaining parameter；
Classification analysis step：By classification analysis, according to the feature of known sample car plate, judge which kind of new sample car plate belongs to
Known sample class selects characteristic parameter by calculating, establishes discriminant function, classify to sample car plate.
2. the Car license recognition learning method according to claim 1 based on artificial neural network, which is characterized in that also wrap
It includes：
Weight set-up procedure：Specify the weight of Car license recognition instrument connection, the mutual connection of the Car license recognition instrument of Same Site
Weight is higher than the connection weight between the Car license recognition instrument in different places.
3. the Car license recognition learning method according to claim 1 based on artificial neural network, which is characterized in that also wrap
It includes：
Excitation value set-up procedure：The excitation value of Car license recognition instrument is adjusted according to vehicle flowrate, the excitation value of Car license recognition instrument is with wagon flow
The increase of amount and increase, reduced with the reduction of vehicle flowrate.
4. a kind of Car license recognition learning system based on artificial neural network, which is characterized in that including：
Artificial neural network sets up module：Using each Car license recognition instrument as artificial neuron, Car license recognition is formed by network
The artificial neural network of system provides Car license recognition instrument sample car plate and is learnt to obtain experience table, according to the experience table
Forecasting recognition result；
Mathematical model establishes module：By regression analysis, multiple characters of unified car plate or multiple words of multiple car plates are understood
Whether symbol is related, related direction and intensity, founding mathematical models；
Mathematical model adjusts module：By verifying target license plate, the response variables of mathematical model is adjusted with explaining parameter；
Classification analysis module：By classification analysis, according to the feature of known sample car plate, judge which kind of new sample car plate belongs to
Known sample class selects characteristic parameter by calculating, establishes discriminant function, classify to sample car plate.
5. the Car license recognition learning system according to claim 4 based on artificial neural network, which is characterized in that also wrap
It includes：
Weight adjusts module：Specify the weight of Car license recognition instrument connection, the mutual connection of the Car license recognition instrument of Same Site
Weight is higher than the connection weight between the Car license recognition instrument in different places.
6. the Car license recognition learning system according to claim 4 based on artificial neural network, which is characterized in that also wrap
It includes：
Excitation value adjusts module：The excitation value of Car license recognition instrument is adjusted according to vehicle flowrate, the excitation value of Car license recognition instrument is with wagon flow
The increase of amount and increase, reduced with the reduction of vehicle flowrate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711146253.XA CN108052866A (en) | 2017-11-17 | 2017-11-17 | Car license recognition learning method and system based on artificial neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711146253.XA CN108052866A (en) | 2017-11-17 | 2017-11-17 | Car license recognition learning method and system based on artificial neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108052866A true CN108052866A (en) | 2018-05-18 |
Family
ID=62120210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711146253.XA Pending CN108052866A (en) | 2017-11-17 | 2017-11-17 | Car license recognition learning method and system based on artificial neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108052866A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110596685A (en) * | 2019-09-13 | 2019-12-20 | 福建工程学院 | Monocular distance measurement method and device for random scene under rapid movement |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1694130A (en) * | 2005-03-24 | 2005-11-09 | 上海大学 | Identification method of mobile number plate based on three-channel parallel artificial nerve network |
| CN101105893A (en) * | 2006-07-14 | 2008-01-16 | 沈阳江龙软件开发科技有限公司 | Automobile video frequency discrimination speed-testing method |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| CN104657748A (en) * | 2015-02-06 | 2015-05-27 | 中国石油大学(华东) | Vehicle type recognition method based on convolutional neural network |
| CN105184271A (en) * | 2015-09-18 | 2015-12-23 | 苏州派瑞雷尔智能科技有限公司 | Automatic vehicle detection method based on deep learning |
| CN106096602A (en) * | 2016-06-21 | 2016-11-09 | 苏州大学 | Chinese license plate recognition method based on convolutional neural network |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| CN107220638A (en) * | 2017-07-03 | 2017-09-29 | 深圳市唯特视科技有限公司 | A kind of car plate detection recognition methods based on deep learning convolutional neural networks |
-
2017
- 2017-11-17 CN CN201711146253.XA patent/CN108052866A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1694130A (en) * | 2005-03-24 | 2005-11-09 | 上海大学 | Identification method of mobile number plate based on three-channel parallel artificial nerve network |
| CN101105893A (en) * | 2006-07-14 | 2008-01-16 | 沈阳江龙软件开发科技有限公司 | Automobile video frequency discrimination speed-testing method |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| CN104657748A (en) * | 2015-02-06 | 2015-05-27 | 中国石油大学(华东) | Vehicle type recognition method based on convolutional neural network |
| CN105184271A (en) * | 2015-09-18 | 2015-12-23 | 苏州派瑞雷尔智能科技有限公司 | Automatic vehicle detection method based on deep learning |
| CN106096602A (en) * | 2016-06-21 | 2016-11-09 | 苏州大学 | Chinese license plate recognition method based on convolutional neural network |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| CN107220638A (en) * | 2017-07-03 | 2017-09-29 | 深圳市唯特视科技有限公司 | A kind of car plate detection recognition methods based on deep learning convolutional neural networks |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110596685A (en) * | 2019-09-13 | 2019-12-20 | 福建工程学院 | Monocular distance measurement method and device for random scene under rapid movement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Machine vision for natural gas methane emissions detection using an infrared camera | |
| CN108734283B (en) | Neural network system | |
| Fourati | Multiple neural control of a greenhouse | |
| Lv et al. | A long Short-Term memory cyclic model with mutual information for hydrology forecasting: A Case study in the xixian basin | |
| Wang et al. | A new ARMAX model based on evolutionary algorithm and particle swarm optimization for short-term load forecasting | |
| CN106951919A (en) | A kind of flow monitoring implementation method based on confrontation generation network | |
| Hosseini et al. | Artificial intelligence for resilience enhancement of power distribution systems | |
| CN106254330A (en) | A kind of software defined network intrusion detection method based on BP neutral net | |
| CN108091111A (en) | A kind of oil truck oil and gas leakage intelligent early-warning system | |
| CN116629465B (en) | Smart power grids video monitoring and risk prediction response system | |
| Jozi et al. | Energy consumption forecasting based on hybrid neural fuzzy inference system | |
| CN112381142B (en) | Method and system for generating explainability confrontation sample based on important features | |
| CN112272074B (en) | Information transmission rate control method and system based on neural network | |
| Narang et al. | Weight pattern evaluation for multiobjective hydrothermal generation scheduling using hybrid search technique | |
| Swaroop et al. | Load forecasting for power system planning using fuzzy-neural networks | |
| Kabir et al. | Partial adversarial training for prediction interval | |
| CN108052866A (en) | Car license recognition learning method and system based on artificial neural network | |
| CN104463130B (en) | A kind of license plate image photo-irradiation treatment method based on assessment reponse system | |
| CN107220168A (en) | It is a kind of based on integrated information system system maturity assessment method | |
| CN111967003A (en) | Automatic wind control rule generation system and method based on black box model and decision tree | |
| Manikonda et al. | Power quality event classification using transfer learning on images | |
| Yap et al. | Compressing and improving fuzzy rules using genetic algorithm and its application to fault detection | |
| CN116488325A (en) | Smart power grid anomaly detection and classification method, smart power grid anomaly detection and classification equipment and readable storage medium | |
| Amogha | Load Forecasting Algorithms with Simulation & Coding | |
| El-Sharkh et al. | Forecasting electricity demand using dynamic artificial neural network model |
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: 20180518 |
|
| RJ01 | Rejection of invention patent application after publication |