CN110334586A - A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing - Google Patents
A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN110334586A CN110334586A CN201910429460.9A CN201910429460A CN110334586A CN 110334586 A CN110334586 A CN 110334586A CN 201910429460 A CN201910429460 A CN 201910429460A CN 110334586 A CN110334586 A CN 110334586A
- Authority
- CN
- China
- Prior art keywords
- automobile
- car
- difference
- british standard
- information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012790 confirmation Methods 0.000 claims abstract description 33
- 239000000284 extract Substances 0.000 claims abstract description 5
- 238000013527 convolutional neural network Methods 0.000 claims description 21
- 238000013075 data extraction Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 7
- 239000000203 mixture Substances 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 229910000838 Al alloy Inorganic materials 0.000 description 15
- 238000012015 optical character recognition Methods 0.000 description 11
- 239000000463 material Substances 0.000 description 9
- 238000012549 training Methods 0.000 description 9
- 238000011478 gradient descent method Methods 0.000 description 8
- 230000000052 comparative effect Effects 0.000 description 6
- 229910001018 Cast iron Inorganic materials 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 239000000446 fuel Substances 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000002347 injection Methods 0.000 description 4
- 239000007924 injection Substances 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000035800 maturation Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 240000005319 Sedum acre Species 0.000 description 1
- 235000014327 Sedum acre Nutrition 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 239000007858 starting material Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- 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
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Multimedia (AREA)
- Finance (AREA)
- Economics (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing, are based on artificial intelligence, comprising the following steps: shooting Vehicle Identify Number simultaneously obtains vehicle frame number；British Standard, Automobile Series information is extracted according to vehicle frame number；Result or differences are confirmed according to British Standard, Automobile Series information acquisition；Automobile is shot according to differences, to obtain difference collection；Difference collection is compared with British Standard, Automobile Series, and obtains confirmation result.The present invention is by shooting Vehicle Identify Number and obtains vehicle frame number, then extracts British Standard, Automobile Series information；Realize the identification to the British Standard, Automobile Series of automobile；Differences are obtained using contrast difference's module, and difference collection is obtained using shooting identification module shooting automobile, difference collection and British Standard, Automobile Series are finally compared into acquisition confirmation result, realize the identification to the car's style information of automobile, and then be conducive to the definite price that user determines automobile, the risk for reducing credit enterprise avoids credit enterprise because of the case where causing damages to automobile valuation mistake.
Description
Technical field
The present invention relates to field of artificial intelligence more particularly to a kind of automobile recognition methods, device, computer system and
Readable storage medium storing program for executing.
Background technique
The mode of current identification vehicle style has artificial rule of thumb identification, manually according to atlas matching identification, shooting vapour
The identification of vehicle distant view photograph, input Vehicle Identify Number method etc..
Wherein, the poor process control of manual identified, and the mode for shooting or inputting, knowing vehicle system at present can identify
Vehicle, and cannot recognize that the specific style of automobile.In credit scene, the automobile credit amount of different styles be it is differentiated,
If which style therefore cannot be recognized accurately is, it will increase non-controllable risk to automobile credit, in turn result in automobile credit
Traffic lost.
Summary of the invention
The object of the present invention is to provide a kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing, are used for
It solves the problems, such as of the existing technology.
To achieve the above object, the present invention provides a kind of automobile recognition methods, comprising the following steps:
S1: vehicle frame number is obtained according to the Vehicle Identify Number of the automobile of photographic device shooting；
S2: the corresponding British Standard, Automobile Series information of the Vehicle Identify Number is extracted according to vehicle frame number；
S3: judging whether the British Standard, Automobile Series information includes multiple car's styles, if so, according to the multiple automobile is obtained
The differences of style, and execute step S4；If it is not, then obtaining the confirmation result of the corresponding car's style of the Vehicle Identify Number；
S4: the corresponding automobile component image of the differences of photographic device shooting is obtained, to obtain difference collection；
S5: by the difference collection and British Standard, Automobile Series information comparison, and confirmation result is obtained.
It further include step S0 in above scheme, the step S0 includes founding database；
Database in the step S0 has British Standard, Automobile Series information, and the British Standard, Automobile Series information association has and the vehicle frame
The corresponding serial number of number.
In above scheme, the step S1 the following steps are included:
S11: shooting instruction is exported to shoot the Vehicle Identify Number of automobile；Wherein, according to the reflective to vehicle of Vehicle Identify Number position
Frame number carries out auto-focusing, to shoot the Vehicle Identify Number picture with predetermined size；
S12: Vehicle Identify Number picture is received；
S13: Vehicle Identify Number picture is identified using OCR, the Vehicle Identify Number in Vehicle Identify Number picture, which is converted to, to be identified by computer
Vehicle frame number.
In above scheme, the step S2 the following steps are included:
S21: by vehicle frame number successively with series number comparing；
S22: it obtains and the consistent serial number of vehicle frame number；
S23: the British Standard, Automobile Series information with the series number data correlation is extracted.
In above scheme, the step S3 the following steps are included:
If only having a car's style information in British Standard, Automobile Series information, tied the car's style information as confirmation
Fruit；
If in British Standard, Automobile Series information tool there are two and more than two car's style information, extract each car's style information
Between differences.
In above scheme, the step S4 the following steps are included:
S41: according to differences to photographic device output difference instruction, make the differences institute of the photographic device shooting automobile
Corresponding position, and generate difference pictures；
S42: the difference pictures are received, and using convolutional neural networks Recognition Different picture and generate variance data；
S43: the variance data is associated with the differences, generates difference results；
S44: step S41- step S43 successively is repeated to the differences between several car's style information, and is finally obtained
Obtain the difference collection being made of at least one difference results.
In above scheme, the step S5 the following steps are included:
S51: the configuration item by the difference results in difference collection successively with car's style information each in British Standard, Automobile Series compares；
S52: it obtains and the matched car's style information of the difference collection；
S53: using the car's style information as confirmation result.
To achieve the above object, the present invention also provides a kind of automobile identification devices, comprising:
The Vehicle Identify Number of shooting module, the automobile for being shot according to photographic device obtains vehicle frame number；
Data extraction module, for extracting the corresponding British Standard, Automobile Series information of the Vehicle Identify Number according to vehicle frame number；
Contrast difference's module, for judging whether the British Standard, Automobile Series information includes multiple car's styles, if so, according to obtaining
Take the differences of the multiple car's style；If it is not, then obtaining the confirmation result of the corresponding car's style of the Vehicle Identify Number；
Identification module is shot, for obtaining the corresponding automobile component image of the differences of photographic device shooting, with
To difference collection；
Style confirmation module is used for the difference collection and British Standard, Automobile Series information comparison, and obtains confirmation result.
To achieve the above object, the present invention also provides a kind of computer systems comprising multiple computer equipments, it is each to calculate
Machine equipment includes memory processor and stores the computer program that can be run on a memory and on a processor, described
The step of processor of multiple computer equipments realizes above-mentioned automobile recognition methods when executing the computer program jointly.
To achieve the above object, the present invention also provides a kind of computer readable storage mediums comprising multiple storage mediums,
Computer program is stored on each storage medium, the computer program of the multiple storage medium storage is executed by processor
Shi Gongtong realizes the step of above-mentioned automobile recognition methods.
A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing provided by the invention are based on artificial intelligence
Can, after shooting Vehicle Identify Number by shooting module, image procossing is carried out to obtain vehicle frame to Vehicle Identify Number picture using OCR identification technology
Vehicle Identify Number in number picture, and the Vehicle Identify Number is switched to the vehicle frame number that can be identified by computer；Pass through data extraction module
Each automobile money in the British Standard, Automobile Series information is obtained according to Vehicle Identify Number data acquisition British Standard, Automobile Series information, then by contrast difference's module
The differences of formula information, if only one car's style information in British Standard, Automobile Series information, which is directly made
To confirm result；Automobile is shot according to differences using shooting identification module, and obtains difference collection, then true by style
Recognize module judges which car's style information in the British Standard, Automobile Series information according to difference collection, is the vapour of automobile of being currently taken
Vehicle style realizes the quickly accurate purpose for determining car's style information；And then be conducive to the definite price that user determines automobile,
Under credit scene, the risk of credit enterprise is reduced, avoids credit enterprise because of the feelings that cause damages to automobile valuation mistake
Condition occurs.
Detailed description of the invention
Fig. 1 is the flow chart of automobile recognition methods embodiment one of the present invention；
Fig. 2 is the work flow diagram in automobile recognition methods embodiment one of the present invention between automobile identification device and service system；
Fig. 3 is the program module schematic diagram of automobile identification device embodiment two of the present invention；
Fig. 4 is the hardware structural diagram of computer equipment in computer system embodiment three of the present invention.
Appended drawing reference:
1, automobile identification device 2, photographic device 3, OCR network service port
4, computer equipment
11, shooting module 12, data extraction module 13, contrast difference's module
14, it shoots identification module 15, style confirmation module 16, found module
111, photographing control unit 112, picture receiving unit 113, Vehicle Identify Number recognition unit
121, Vehicle Identify Number comparing unit 122, comparison receiving unit 123, information extraction unit
131, difference shooting unit 132, variance data generation unit 133, association results unit
134, difference collection unit 151, configuration comparing unit 152, style receiving unit
153, confirm result output unit 41, memory 42, processor
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.Based on the embodiments of the present invention, those of ordinary skill in the art are not before making creative work
Every other embodiment obtained is put, shall fall within the protection scope of the present invention.
A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing provided by the invention are suitable for artificial
Smart field, it is a kind of true based on shooting module, data extraction module, contrast difference's module, shooting identification module, style to provide
Recognize module and founds the automobile recognition methods of module.After the present invention shoots Vehicle Identify Number by shooting module, OCR identification technology is utilized
Image procossing is carried out to obtain the Vehicle Identify Number in Vehicle Identify Number picture to Vehicle Identify Number picture, and the Vehicle Identify Number is switched to can be by computer
The vehicle frame number of identification；By data extraction module according to Vehicle Identify Number data acquisition British Standard, Automobile Series information, then pass through difference pair
Than the differences that module obtains each car's style information in the British Standard, Automobile Series information, if only one automobile in British Standard, Automobile Series information
Style information, then by the car's style information directly as confirmation result；Using shooting identification module according to differences to automobile
It is shot, and obtains difference collection, then which in the British Standard, Automobile Series information is judged according to difference collection by style confirmation module
One car's style information realizes the quickly accurate purpose for determining car's style information for the car's style for the automobile that is currently taken；
Be conducive to the definite price that user determines automobile in turn reduces the risk of credit enterprise, avoids letter under credit scene
Enterprise is borrowed because of the case where causing damages to automobile valuation mistake.
Embodiment one
Please refer to Fig. 1 and Fig. 2, a kind of automobile recognition methods of the present embodiment, using automobile identification device 1, including it is following
Step:
S1: vehicle frame number is obtained according to the Vehicle Identify Number of the automobile of photographic device shooting；
S2: the corresponding British Standard, Automobile Series information of the Vehicle Identify Number is extracted according to vehicle frame number；
S3: judging whether the British Standard, Automobile Series information includes multiple car's styles, if so, according to the multiple automobile is obtained
The differences of style, and execute step S4；If it is not, then obtaining the confirmation result of the corresponding car's style of the Vehicle Identify Number；
S4: the corresponding automobile component image of the differences of photographic device shooting is obtained, to obtain difference collection；
S5: by the difference collection and British Standard, Automobile Series information comparison, and confirmation result is obtained.
It preferably, further include step S0, the step S0 includes founding database.
Specifically, the database in the step S0 has British Standard, Automobile Series information, the British Standard, Automobile Series information association, which has, is
Row number data, the series number data include manufacturer's code, engine code and chassis serial number.
Further, a car's style information is included at least in the British Standard, Automobile Series information；The car's style information
With at least one configuration item, configuration item such as: it is tire size, suspension, cylinder block material, steering wheel type, rim material, complete
Red-spotted stonecrop window etc..
Preferably, the database at least has a British Standard, Automobile Series information.
Further, British Standard, Automobile Series information is obtained in internet in real time；The British Standard, Automobile Series letter having in setting database
Breath is prestored message, and the British Standard, Automobile Series information obtained from internet is comparative information；
The comparative information is compared with prestored message, if the comparative information is consistent with prestored message, is abandoned
Comparative information；If the comparative information and prestored message are inconsistent, comparative information is stored in database, lead to database can
The British Standard, Automobile Series information that internet real-time update has been deposited is crossed, to guarantee the integrality of database information, and then guarantees automobile identification
Reliability.
Specifically, the step S1 the following steps are included:
S11: shooting instruction is exported to shoot the Vehicle Identify Number of automobile；Wherein, according to the reflective to vehicle of Vehicle Identify Number position
Frame number carries out auto-focusing, to shoot the Vehicle Identify Number picture with predetermined size；
Specifically, the step S11 includes exporting shooting instruction to photographic device 2 to shoot the Vehicle Identify Number of automobile.
Further, the predetermined size can be set as needed.
S12: Vehicle Identify Number picture is received；
S13: calling OCR network service port 3, using OCR (OCR that is, Optical Character Recognition,
Optical character identification refers to that electronic equipment, such as scanner or digital camera check the character printed on paper, dark by detection,
Bright mode determines its shape, then shape is converted to the method for the process of computword with character identifying method) identification
Vehicle Identify Number picture, and the Vehicle Identify Number in Vehicle Identify Number picture is converted to the vehicle frame number that can be identified by computer.
Wherein, Vehicle Identify Number data include manufacturer's code, engine code and chassis serial number.
Specifically, the step S2 the following steps are included:
S21: by vehicle frame number successively with series number comparing；
S22: it obtains and the consistent serial number of the vehicle frame number；
S23: the British Standard, Automobile Series information with the series number data correlation is extracted.
Specifically, the step S3 the following steps are included:
If only having a car's style information in British Standard, Automobile Series information, tied the car's style information as confirmation
Fruit；Such as: British Standard, Automobile Series information is 2013 sections of automobiles of R board M system, only has car's style information A in the British Standard, Automobile Series information
One car's style information；
If in British Standard, Automobile Series information tool there are two and more than two car's style information, extract each car's style information
Between differences；Such as: British Standard, Automobile Series information is 2013 sections of automobiles of R board M system, has car's style in the British Standard, Automobile Series information
Two car's style information of information A and car's style information B.
Wherein, different configuration items of the differences between several car's style information.
Further, the configuration item includes steering wheel type, wheel rim material, without key starter, keyless entry
Device etc..
Such as: the configuration item of car's style information A be ... fuel system-direct-injection, cylinder block material-aluminium alloy, steering wheel
Type-multi-functional steering wheel, wheel rim material-aluminium alloy ... }；
The configuration item of car's style information B are as follows: ... fuel system-direct-injection, cylinder block material-aluminium alloy, steering wheel type
Type-common handle, wheel rim material-cast iron ... }；
Then the differences of car's style information A and car's style information B are { steering wheel type, wheel rim material }.
Specifically, the step S4 the following steps are included:
S41: difference instruction is exported to photographic device 2 according to differences, the photographic device 2 is made to shoot the differences of automobile
Corresponding position, and generate difference pictures；Such as: exporting the difference that content is " steering wheel type " to photographic device 2 and instruct, make
Photographic device 2 shoots the steering wheel of automobile according to difference instruction, and generates the difference pictures of entitled XXX.png；
The difference that content is " wheel rim material " is exported to photographic device 2 again to instruct, and instructs photographic device 2 to automobile according to the difference
Wheel rim is shot, and generates the difference pictures of entitled YYY.png；
Optionally, operator obtains " steering wheel " information by photographic device 2, and photographic device 2 is moved to automobile side
In front of to disk, the difference pictures of entitled XXX.png are shot and generated to steering wheel；Operator passes through photographic device
2 obtain " wheel rim material " information, photographic device 2 are moved in front of the wheel rim of automobile, to be shot to wheel rim and generate name
The referred to as difference pictures of YYY.png.
Optionally, photographic device 2 is mountable on intelligent robot, and photographic device 2 exports " direction to intelligent robot
Disk " instruction after photographic device 2 is moved in front of the steering wheel of automobile by intelligent robot, is completed to refer to the output of photographic device 2
It enables, photographic device 2 is shot to the steering wheel and generated the difference pictures of entitled XXX.png according to completion instruction；
For photographic device 2 again to intelligent robot output " wheel rim material " instruction, photographic device 2 is moved to the wheel of automobile by intelligent robot
It behind circle front, is exported to photographic device 2 and completes instruction, photographic device 2 shoots simultaneously the wheel rim according to completion instruction
Generate the difference pictures of entitled YYY.png.
S42: the difference pictures are received, and using mature convolutional neural networks Recognition Different picture and generate difference number
According to；
Specifically, including: to load mature convolutional neural networks model；Receive the difference pictures；Run the mature volume
Product neural network model identifies the difference pictures and generates variance data；
Such as: receiving the difference pictures of entitled XXX.png, the maturation convolutional neural networks carry out the difference pictures
Identification, and judge the steering wheel in the difference pictures for multi-functional steering wheel or common handle；Receive entitled YYY.png
Difference pictures, the maturation convolutional neural networks identify the difference pictures, and judge the direction in the difference pictures
Disk is aluminium alloy or cast iron；
If in difference pictures being multi-functional steering wheel, the variance data that content is " multi-functional steering wheel " is exported；If poor
Wheel rim material in different picture is aluminium alloy, then exports the variance data that content is " aluminium alloy ".
S43: the variance data is associated with the differences, generates difference results；Such as: being " multi-functional side by content
To disk " variance data and content be that the differences of " steering wheel type " be associated withs, generation content is " steering wheel type-multi-functional
The difference results of steering wheel "；The variance data that content is " aluminium alloy " is associated with the differences that content is " wheel rim material ", it is raw
It is the difference results of " wheel rim material-aluminium alloy " at content；
S44: step S41- step S43 successively is repeated to the differences between several car's style information, and is finally obtained
Obtain the difference collection being made of at least one difference results；
Such as: repeating step S41- step S43 and obtain difference collection；
Wherein, difference, which is concentrated, at least has a difference results, therefore the manifestation mode of difference collection can are as follows:
{ steering wheel type-multi-functional steering wheel, wheel rim material-aluminium alloy }.
Further, in the step S42, mature convolutional neural networks can be trained by following steps:
S42-1: prepare the picture at each position of enough automobiles, and mark the picture.Such as: picture:
Xxx.png, label: multi-functional steering wheel；Picture: xx.png, label: common handle；And picture: yyy.png, label:
Aluminium alloy；Picture: yy.png, label: cast iron；
S42-2: prepare primary convolution neural network model；
Wherein, the convolutional neural networks model can be and be not limited to classical mature convolutional neural networks VGG model
(" VGG " represents Oxonian Oxford Visual Geometry Group, which is under the jurisdiction of establishment in 1985
Robotics Research Group, the Group research range include machine learning to mobile robot), Resnet model
(ResNet (residual error neural network) was proposed by 4 Chinese such as He Kaiming of Microsoft Research in 2015, successfully had trained 152
The super deep convolutional neural networks of layer, effect is very prominent, and is readily incorporated into other network structures), Inception mould
Type etc.；
Preferably, the convolutional neural networks model in the present embodiment preferably use Densenet model (by Highway,
The inspiration of ResNet scheduling algorithm thinking, it is thus proposed that DenseNet model proposes a kind of connection network of cross-layer.
DenseNet design comparison is bold, each layer of input include before all layers of information, by by the more a layers of front N
Feature combines, and forms the description and differentiation richer to feature)；
S42-3: preparing multiple batches of training sample, and the training sample is marked, the label and the trained sample
This type matching；
In this step, the training sample is picture, and the batch of training sample is classified according to attribute, per batch of instruction
Practicing sample is similar similar sample, i.e., attribute is same type, different types of picture；
Wherein, attribute is auto parts and components, such as steering wheel, wheel hub etc.；Type is the type of auto parts and components, such as multi-functional
Steering wheel, common handle etc..
S42-3: a batch of training sample will be appointed to carry out in batches to convolutional neural networks model according to gradient descent method
More wheel training, until the convolutional neural networks model can accuracy rate reach preset value；
Wherein, gradient descent method is one kind of iterative method, can be used for solving least square problem.Solving machine learning
When the model parameter, i.e. unconstrained optimization problem of algorithm, it is commonly used method that gradient, which declines (Gradient Descent),
One of, when solving the minimum value of loss function, can be minimized by gradient descent method come iterative solution step by step
Loss function and model parameter value.In turn, it if we need to solve the maximum value of loss function, at this moment just needs with ladder
Degree rise method carrys out iteration.In machine learning, two kinds of gradient descent methods are developed based on basic gradient descent method, respectively
For stochastic gradient descent method and batch gradient descent method, the present embodiment is then using batch gradient descent method.
S42-4: the structure and weight and the mature convolution for obtaining the batch of the primary convolution neural network model are saved
Neural network, the maturation convolutional neural networks have lot label, the lot label and the training sample of the batch
Attributes match；Via the primary convolution neural network model that the training sample that attribute is " steering wheel " is trained, will finally obtain
There must be the mature convolutional neural networks of " steering wheel " lot label.
S42-5: successively the training sample of each batch is trained according to the method for S42-3 and S42-4, and obtains each batch
Secondary mature convolutional neural networks summarize each mature convolutional neural networks and obtain neural network collection.
Further, the step S42 includes: the mature convolutional neural networks for successively concentrating differences with neural network
Lot label compare, obtain and the matched mature convolutional neural networks of the differences, and the mature convolutional Neural
Network Recognition difference pictures simultaneously generate variance data.
Specifically, step S5 the following steps are included:
S51: the configuration item by the difference results in difference collection successively with car's style information each in British Standard, Automobile Series compares；
Such as: difference collection is { steering wheel type-multi-functional steering wheel, wheel rim material-aluminium alloy }, the difference that difference is concentrated
As a result " steering wheel type-multi-functional steering wheel ", " wheel rim material-aluminium alloy " successively with the configuration item of car's style information ratio
It is right；
For another example: car's style information A is 2013 sections of automobile comfort types of R board M system；Car's style information B is R board M system 2013
Money automobile is economical；
The configuration item of car's style information A be ... fuel system-direct-injection, cylinder block material-aluminium alloy, steering wheel type-
Multi-functional steering wheel, wheel rim material-aluminium alloy ... }；
The configuration item of car's style information B are as follows: ... fuel system-direct-injection, cylinder block material-aluminium alloy, steering wheel type
Type-common handle, wheel rim material-cast iron ... }；
The difference results successively configuration with the configuration item of car's style information A and car's style information B that difference is concentrated
Item is compared.
S52: it obtains and the matched car's style information of the difference collection；
Such as: it is based on above-mentioned example, it is known that car's style information A is matched with all differences result that difference is concentrated.
S53: using the car's style information as confirmation result.
Such as: above-mentioned example is based on, using car's style information A as confirmation result.
It preferably, further include step S6；
S6: AUTO QUOTED PRICE database is founded, and is joined from the AUTO QUOTED PRICE database according to the confirmation result
It examines quotation and stores.
Further, step S6 the following steps are included:
S61: AUTO QUOTED PRICE database is founded, there are each all automobiles of British Standard, Automobile Series information in the AUTO QUOTED PRICE database
The quotation information of style information；Wherein, the quotation information includes quotation for new cars and used car quotation；
S62: it will confirm that result is brought AUTO QUOTED PRICE database, acquisition and the matched quotation information of the confirmation result into and incited somebody to action
The quotation information is set as target offers；Summarize the target offers and forms quotation collection；
S63: being set as quotation lower limit for the smallest target offers of the quotation lumped values, and the quotation lumped values are maximum
Target offers are set as price cap；The quotation lower limit and price cap as info quote and are stored.
Embodiment two
Referring to Fig. 3, a kind of automobile identification device 1 of the present embodiment, comprising:
The Vehicle Identify Number of shooting module 11, the automobile for being shot according to photographic device obtains vehicle frame number；
Data extraction module 12, for extracting the corresponding British Standard, Automobile Series information of the Vehicle Identify Number according to vehicle frame number；
Contrast difference's module 13, for judging whether the British Standard, Automobile Series information includes multiple car's styles, if so, according to
Obtain the differences of the multiple car's style；If it is not, then obtaining the confirmation result of the corresponding car's style of the Vehicle Identify Number；
Identification module 14 is shot, for obtaining the corresponding automobile component image of the differences of photographic device shooting, with
Obtain difference collection；
Style confirmation module 15 is used for the difference collection and British Standard, Automobile Series information comparison, and obtains confirmation result.
Preferably, automobile identification device 1 further include:
Module 16 is founded, for founding the database for having British Standard, Automobile Series information.
Specifically, the shooting module 11 includes:
Photographing control unit 111 shoots the Vehicle Identify Number of automobile for exporting shooting instruction；
Picture receiving unit 112, for receiving Vehicle Identify Number picture；
Vehicle Identify Number recognition unit 113 identifies Vehicle Identify Number picture using OCR, and will for calling OCR network service port 3
Vehicle Identify Number in Vehicle Identify Number picture is converted to the vehicle frame number that can be identified by computer.
Specifically, the data extraction module 12 includes:
Vehicle Identify Number comparing unit 121, by vehicle frame number successively with series number comparing；
Receiving unit 122 is compared, is obtained and the consistent serial number of the vehicle frame number；
Information extraction unit 123 extracts the British Standard, Automobile Series information with the series number data correlation.
Specifically, contrast difference's module 13 includes:
Difference shooting unit 131 makes the photographic device 2 for exporting difference instruction to photographic device 2 according to differences
Position corresponding to the differences of automobile is shot, and generates difference pictures；
Variance data generation unit 132, for using mature convolutional neural networks Recognition Different picture and generating difference number
According to；
Association results unit 133 generates difference results for the variance data to be associated with the differences；
Difference collection unit 134 is clapped for successively reusing difference to the differences between several car's style information
Unit, variance data generation unit and association results unit are taken the photograph, finally at least obtains a difference results, and obtain by least one
The difference collection of a difference results composition.
Specifically, the style confirmation module 15 includes:
Configure comparing unit 151, by the difference results in difference collection successively with car's style information each in British Standard, Automobile Series
Configuration item compares；
Style receiving unit 152 obtains and the matched car's style information of the difference collection；
Result output unit 153 is confirmed, using the car's style information as confirmation result.
The technical program is based on artificial intelligence, after shooting Vehicle Identify Number by shooting module, using OCR identification technology to vehicle frame
Number picture carries out image procossing to obtain the Vehicle Identify Number in Vehicle Identify Number picture, and the Vehicle Identify Number is switched to can be identified by computer
Vehicle frame number；By data extraction module according to Vehicle Identify Number data acquisition British Standard, Automobile Series information, then pass through contrast difference's module
The differences of each car's style information in the British Standard, Automobile Series information are obtained, if only one car's style is believed in British Standard, Automobile Series information
Breath, then by the car's style information directly as confirmation result；Automobile is clapped according to differences using shooting identification module
It takes the photograph, and obtains difference collection, then which automobile in the British Standard, Automobile Series information judged according to difference collection by style confirmation module
Style information realizes the quickly accurate purpose for determining car's style information for the car's style for the automobile that is currently taken.
Embodiment three:
To achieve the above object, the present invention also provides a kind of computer system, which includes multiple computers
The component part of equipment 4, the automobile identification device 1 of embodiment two is dispersed in different computer equipments, computer equipment
It can be the smart phone for executing program, tablet computer, laptop, desktop computer, rack-mount server, blade type clothes
It is engaged in device, tower server or Cabinet-type server (including server set composed by independent server or multiple servers
Group) etc..The computer equipment of the present embodiment includes, but is not limited to: the memory that connection can be in communication with each other by system bus
41, processor 42, as shown in Figure 4.It should be pointed out that Fig. 4 illustrates only the computer equipment with component-, but should manage
Solution is, it is not required that implements all components shown, the implementation that can be substituted is more or less component.
In the present embodiment, memory 41 (i.e. readable storage medium storing program for executing) includes flash memory, hard disk, multimedia card, card-type memory
(for example, SD or DX memory etc.), random access storage device (RAM), static random-access memory (SRAM), read-only memory
(ROM), electrically erasable programmable read-only memory (EEPROM), programmable read only memory (PROM), magnetic storage, magnetic
Disk, CD etc..In some embodiments, memory 41 can be the internal storage unit of computer equipment, such as the computer
The hard disk or memory of equipment.In further embodiments, memory 41 is also possible to the External memory equipment of computer equipment, example
The plug-in type hard disk being equipped in such as computer equipment, intelligent memory card (Smart Media Card, SMC), secure digital
(Secure Digital, SD) card, flash card (Flash Card) etc..Certainly, memory 41 can also both be set including computer
Standby internal storage unit also includes its External memory equipment.In the present embodiment, memory 41 is installed on meter commonly used in storage
Calculate machine equipment operating system and types of applications software, such as embodiment one automobile identification device program code etc..In addition,
Memory 41 can be also used for temporarily storing the Various types of data that has exported or will export.
Processor 42 can be in some embodiments central processing unit (Central Processing Unit, CPU),
Controller, microcontroller, microprocessor or other data processing chips.The processor 42 is commonly used in control computer equipment
Overall operation.In the present embodiment, program code or processing data of the processor 42 for being stored in run memory 41, example
Automobile identification device is run, such as to realize the automobile recognition methods of embodiment one.
Example IV:
To achieve the above object, the present invention also provides a kind of computer-readable storage systems comprising multiple storage mediums,
Such as flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory), random access storage device (RAM), static state
Random access storage device (SRAM), electrically erasable programmable read-only memory (EEPROM), may be programmed read-only memory (ROM)
Read-only memory (PROM), magnetic storage, disk, CD, server, App are stored thereon with computer using store etc.
Program, program realize corresponding function when being executed by processor 42.The computer readable storage medium of the present embodiment is for storing vapour
Vehicle identification device realizes the automobile recognition methods of embodiment one when being executed by processor 42.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of automobile recognition methods, which comprises the following steps:
S1: vehicle frame number is obtained according to the Vehicle Identify Number of the automobile of photographic device shooting；
S2: the corresponding British Standard, Automobile Series information of the Vehicle Identify Number is extracted according to vehicle frame number；
S3: judging whether the British Standard, Automobile Series information includes multiple car's styles, if so, according to the multiple car's style is obtained
Differences, and execute step S4；If it is not, then obtaining the confirmation result of the corresponding car's style of the Vehicle Identify Number；
S4: the corresponding automobile component image of the differences of photographic device shooting is obtained, to obtain difference collection；
S5: by the difference collection and British Standard, Automobile Series information comparison, and confirmation result is obtained.
2. a kind of automobile recognition methods according to claim 1, which is characterized in that it further include step S0, the step S0
Including founding database；
Database in the step S0 has British Standard, Automobile Series information, and the British Standard, Automobile Series information association has and the vehicle frame number
According to corresponding serial number.
3. a kind of automobile recognition methods according to claim 2, which is characterized in that the step S1 the following steps are included:
S11: shooting instruction is exported to shoot the Vehicle Identify Number of automobile；Wherein, according to the reflective to Vehicle Identify Number of Vehicle Identify Number position
Auto-focusing is carried out, to shoot the Vehicle Identify Number picture with predetermined size；
S12: Vehicle Identify Number picture is received；
S13: identifying Vehicle Identify Number picture using OCR, and the Vehicle Identify Number in Vehicle Identify Number picture is converted to the vehicle that can be identified by computer
Frame number.
4. a kind of automobile recognition methods according to claim 2, which is characterized in that the step S2 the following steps are included:
S21: by vehicle frame number successively with series number comparing；
S22: it obtains and the consistent serial number of vehicle frame number；
S23: the British Standard, Automobile Series information with the series number data correlation is extracted.
5. a kind of automobile recognition methods according to claim 2, which is characterized in that the step S3 the following steps are included:
If only having a car's style information in British Standard, Automobile Series information, using the car's style information as confirmation result；
If in British Standard, Automobile Series information tool there are two and more than two car's style information, extract between each car's style information
Differences.
6. a kind of automobile recognition methods according to claim 2, which is characterized in that the step S4 the following steps are included:
S41: according to differences to photographic device output difference instruction, corresponding to the differences for making the photographic device shooting automobile
Position, and generate difference pictures；
S42: the difference pictures are received, and using convolutional neural networks Recognition Different picture and generate variance data；
S43: the variance data is associated with the differences, generates difference results；
S44: successively between several car's style information differences repeat step S41- step S43, and finally obtain by
The difference collection of at least one difference results composition.
7. a kind of automobile recognition methods according to claim 2, which is characterized in that step S5 the following steps are included:
S51: the configuration item by the difference results in difference collection successively with car's style information each in British Standard, Automobile Series compares；
S52: it obtains and the matched car's style information of the difference collection；
S53: using the car's style information as confirmation result.
8. a kind of automobile identification device characterized by comprising
The Vehicle Identify Number of shooting module, the automobile for being shot according to photographic device obtains vehicle frame number；
Data extraction module, for extracting the corresponding British Standard, Automobile Series information of the Vehicle Identify Number according to vehicle frame number；
Contrast difference's module, for judging whether the British Standard, Automobile Series information includes multiple car's styles, if so, according to institute is obtained
State the differences of multiple car's styles；If it is not, then obtaining the confirmation result of the corresponding car's style of the Vehicle Identify Number；
Identification module is shot, for obtaining the corresponding automobile component image of the differences of photographic device shooting, to obtain difference
Different collection；
Style confirmation module is used for the difference collection and British Standard, Automobile Series information comparison, and obtains confirmation result.
9. a kind of computer system comprising multiple computer equipments, each computer equipment include memory, processor and deposit
Store up the computer program that can be run on a memory and on a processor, which is characterized in that the place of the multiple computer equipment
Reason device realizes the step of any one of claim 1 to 7 automobile recognition methods jointly when executing the computer program.
10. a kind of computer readable storage medium comprising multiple storage mediums are stored with computer journey on each storage medium
Sequence, which is characterized in that the computer program of the multiple storage medium storage realizes right when being executed by processor jointly
It is required that the step of any one of 1 to 7 automobile recognition methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910429460.9A CN110334586A (en) | 2019-05-22 | 2019-05-22 | A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910429460.9A CN110334586A (en) | 2019-05-22 | 2019-05-22 | A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110334586A true CN110334586A (en) | 2019-10-15 |
Family
ID=68139070
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910429460.9A Pending CN110334586A (en) | 2019-05-22 | 2019-05-22 | A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110334586A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110796709A (en) * | 2019-10-29 | 2020-02-14 | 上海眼控科技股份有限公司 | Method and device for acquiring size of frame number, computer equipment and storage medium |
| CN112634066A (en) * | 2020-12-25 | 2021-04-09 | 明觉科技（北京）有限公司 | Method and device for analyzing sales vehicle type through vehicle identification number |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130054832A (en) * | 2011-11-17 | 2013-05-27 | 현대자동차주식회사 | Wireless operating apparatus for vehicle |
| CN106600421A (en) * | 2016-11-21 | 2017-04-26 | 中国平安财产保险股份有限公司 | Intelligent car insurance loss assessment method and system based on image recognition |
| CN107423732A (en) * | 2017-07-26 | 2017-12-01 | 大连交通大学 | Vehicle VIN recognition methods based on Android platform |
| CN108038423A (en) * | 2017-11-22 | 2018-05-15 | 广东数相智能科技有限公司 | The recognition methods of automotive type based on image recognition and device |
| CN108197626A (en) * | 2017-12-28 | 2018-06-22 | 深圳开思时代科技有限公司 | information identifying method, system and computer readable storage medium |
| CN108319907A (en) * | 2018-01-26 | 2018-07-24 | 腾讯科技（深圳）有限公司 | A kind of vehicle identification method, device and storage medium |
| CN108573198A (en) * | 2017-03-14 | 2018-09-25 | 优信互联（北京）信息技术有限公司 | A kind of method and device identifying vehicle information according to Vehicle Identify Number |
| CN108733719A (en) * | 2017-04-24 | 2018-11-02 | 优信拍（北京）信息科技有限公司 | A kind of recognition methods of vehicle position, device, equipment and computer-readable medium |
| KR20190024324A (en) * | 2017-08-31 | 2019-03-08 | 이수연 | Method for providing insurance using deep learning |
| CN109508731A (en) * | 2018-10-09 | 2019-03-22 | 中山大学 | A kind of vehicle based on fusion feature recognition methods, system and device again |
-
2019
- 2019-05-22 CN CN201910429460.9A patent/CN110334586A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130054832A (en) * | 2011-11-17 | 2013-05-27 | 현대자동차주식회사 | Wireless operating apparatus for vehicle |
| CN106600421A (en) * | 2016-11-21 | 2017-04-26 | 中国平安财产保险股份有限公司 | Intelligent car insurance loss assessment method and system based on image recognition |
| CN108573198A (en) * | 2017-03-14 | 2018-09-25 | 优信互联（北京）信息技术有限公司 | A kind of method and device identifying vehicle information according to Vehicle Identify Number |
| CN108733719A (en) * | 2017-04-24 | 2018-11-02 | 优信拍（北京）信息科技有限公司 | A kind of recognition methods of vehicle position, device, equipment and computer-readable medium |
| CN107423732A (en) * | 2017-07-26 | 2017-12-01 | 大连交通大学 | Vehicle VIN recognition methods based on Android platform |
| KR20190024324A (en) * | 2017-08-31 | 2019-03-08 | 이수연 | Method for providing insurance using deep learning |
| CN108038423A (en) * | 2017-11-22 | 2018-05-15 | 广东数相智能科技有限公司 | The recognition methods of automotive type based on image recognition and device |
| CN108197626A (en) * | 2017-12-28 | 2018-06-22 | 深圳开思时代科技有限公司 | information identifying method, system and computer readable storage medium |
| CN108319907A (en) * | 2018-01-26 | 2018-07-24 | 腾讯科技（深圳）有限公司 | A kind of vehicle identification method, device and storage medium |
| CN109508731A (en) * | 2018-10-09 | 2019-03-22 | 中山大学 | A kind of vehicle based on fusion feature recognition methods, system and device again |
Non-Patent Citations (1)
| Title |
|---|
| 茅正冲;韩毅;: "计算机视觉技术的车型识别应用研究", 单片机与嵌入式系统应用, no. 06, 1 June 2017 (2017-06-01) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110796709A (en) * | 2019-10-29 | 2020-02-14 | 上海眼控科技股份有限公司 | Method and device for acquiring size of frame number, computer equipment and storage medium |
| CN112634066A (en) * | 2020-12-25 | 2021-04-09 | 明觉科技（北京）有限公司 | Method and device for analyzing sales vehicle type through vehicle identification number |
| CN112634066B (en) * | 2020-12-25 | 2021-12-10 | 明觉科技（北京）有限公司 | Method and device for analyzing sales vehicle type through vehicle identification number |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10657370B2 (en) | Determining images having unidentifiable license plates | |
| CN107092922B (en) | Vehicle damages recognition methods and server | |
| CN109685100A (en) | Character identifying method, server and computer readable storage medium | |
| CN107909661A (en) | Stop field control method, system and computer-readable recording medium | |
| KR102198296B1 (en) | Apparatus, method and computer program for automatically calculating the damage | |
| CN112862702A (en) | Image enhancement method, device, equipment and storage medium | |
| CN110334586A (en) | A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing | |
| CN112418360B (en) | Convolutional neural network training method, pedestrian attribute identification method and related equipment | |
| CN110705361B (en) | Intelligent damage assessment method and device for vehicle, computer system and readable storage medium | |
| CN109034086A (en) | Vehicle recognition methods, apparatus and system again | |
| CN107730485A (en) | Car damage identification method, electronic equipment and computer-readable recording medium | |
| CN110490127A (en) | Close rule recognition methods and device, storage medium, electronic device in position | |
| CN111311176A (en) | Vehicle maintenance auditing method and device, maintenance end equipment, server and medium | |
| CN110781381A (en) | Data verification method, device and equipment based on neural network and storage medium | |
| CN108510673B (en) | Automatic radio frequency quick checkout method and device and electronic equipment | |
| CN113239910A (en) | Certificate identification method, device, equipment and storage medium | |
| CN112567398A (en) | Techniques for matching different input data | |
| CN115049151A (en) | Navigation route recommendation method, device, equipment and storage medium | |
| CN114663726A (en) | Training method of target type detection model, target detection method and electronic equipment | |
| CN106228179A (en) | The method and system of vehicle comparison | |
| CN111985448A (en) | Vehicle image recognition method and device, computer equipment and readable storage medium | |
| CN110597831A (en) | Method and device for updating article information and computer-readable storage medium | |
| US7519237B2 (en) | Method for characterizing stored information | |
| CN109086762A (en) | A kind of fuzzy license plate identification method based on more mesh cameras | |
| CN111695587A (en) | Vehicle brand and model identification method and 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 |