CN108573198A - A kind of method and device identifying vehicle information according to Vehicle Identify Number - Google Patents
A kind of method and device identifying vehicle information according to Vehicle Identify Number Download PDFInfo
- Publication number
- CN108573198A CN108573198A CN201710149934.5A CN201710149934A CN108573198A CN 108573198 A CN108573198 A CN 108573198A CN 201710149934 A CN201710149934 A CN 201710149934A CN 108573198 A CN108573198 A CN 108573198A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- identify number
- information
- vehicle identify
- feature
- 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
- G06V20/56—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
-
- 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
- G06V10/23—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 based on positionally close patterns or neighbourhood relationships
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/015—Detecting movement of traffic to be counted or controlled with provision for distinguishing between two or more types of vehicles, e.g. between motor-cars and cycles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the present application discloses a kind of method and device identifying vehicle information according to Vehicle Identify Number.In this method, the Vehicle Identify Number set that the Vehicle Identify Number of each known vehicle information is constituted is obtained first；The Vehicle Identify Number is split respectively, obtains each feature for building the Vehicle Identify Number, wherein each feature includes a character and the character information the location of in the Vehicle Identify Number；The information parameter of each feature is calculated separately, and decision tree is built according to described information parameter, wherein the decision node in the decision tree is the feature；Vehicle Identify Number to be identified is judged according to the decision node of the decision tree, the vehicle information of the Vehicle Identify Number to be identified is determined according to judging result.With this solution, decision tree can be created, the identification of vehicle information is realized according to decision tree, it is no longer necessary to the correspondence of manual creation Vehicle Identify Number and vehicle information, therefore compared with prior art, the waste of manpower can be reduced.
Description
Technical field
This application involves vehicle cab recognition field more particularly to a kind of methods and dress identifying vehicle information according to Vehicle Identify Number
It sets.
Background technology
Be otherwise known as Vehicle Identify Number (Vehicle Identification Number, VIN) vehicle identification code, in U.S.'s machine
, it is specified that Vehicle Identify Number is made of 17 characters in motor-car Society of Engineers standard, therefore Vehicle Identify Number is commonly called as 17 codes.In Vehicle Identify Number
Contain the vehicle information of vehicle, wherein the vehicle information generally include the manufacturer of vehicle, the age, body model and
Engine code etc..
Assessment of the vehicle information to vehicle performance, and be of great significance to the diagnosis and repair of vehicle, therefore, in reality
In the application of border, it is often necessary to identify vehicle information by Vehicle Identify Number.In the prior art, vehicle information is identified according to Vehicle Identify Number
When, method that generally use artificially carries out data statistics.In this method, the Vehicle Identify Number of each trolley is checked by staff, and
Vehicle information is counted, the correspondence of Vehicle Identify Number and vehicle information and storage are created, when needing to identify vehicle letter by Vehicle Identify Number
When breath, according to the correspondence of storage and Vehicle Identify Number to be identified, you can obtain corresponding vehicle information.
But inventor has found in the research process of the application, when identifying vehicle information using the prior art, due to needing
The correspondence for wanting manual creation Vehicle Identify Number and vehicle information, generally requiring to expend a large amount of manpowers could complete.
Invention content
This application provides a kind of method and devices identifying vehicle information according to Vehicle Identify Number, to solve prior art progress
When vehicle information identifies, the problem of needing to expend a large amount of manpowers.
In a first aspect, this application provides a kind of methods identifying vehicle information according to Vehicle Identify Number, including：
Obtain the Vehicle Identify Number set that the Vehicle Identify Number of each known vehicle information is constituted；
The Vehicle Identify Number is split respectively, obtains each feature for building the Vehicle Identify Number, wherein each feature packet
Include a character and the character information the location of in the Vehicle Identify Number；
The information parameter of each feature is calculated separately, and decision tree is built according to described information parameter, wherein the decision
Decision node in tree is the feature；
Vehicle Identify Number to be identified is judged according to the decision node of the decision tree, according to judging result determine described in wait for
Identify the vehicle information of Vehicle Identify Number.
Optionally, the information parameter for calculating separately each feature, and decision tree, packet are built according to described information parameter
It includes：
21) statistics is in same Vehicle Identify Number set, the quantity of the identical Vehicle Identify Number of vehicle information, and calculates the vehicle letter
The quantity for ceasing identical Vehicle Identify Number accounts for the ratio of the Vehicle Identify Number set；
22) ratio and predetermined threshold value；
If 23) ratio is less than the predetermined threshold value, according to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number
Affiliated Vehicle Identify Number set, calculates the information parameter of each feature；
24) according to described information parameter, target signature is searched；
25) using the target signature as a decision node, and by the decision node to the Vehicle Identify Number set
In each Vehicle Identify Number divided, obtain meet the decision node Vehicle Identify Number set and do not meet the decision node
Vehicle Identify Number set；
26) it obtains respectively in each Vehicle Identify Number set after dividing, removes the residue character set after target signature, if
The residue character set is not sky, returns to step operation 21).
Optionally, described information parameter is conditional entropy or information gain；
If described information parameter is conditional entropy, described according to described information parameter, searching target signature includes：
The feature of search criterion entropy minimum, and using the feature of conditional entropy minimum as target signature；
If described information parameter is information gain, described according to described information parameter, searching target signature includes：
The maximum feature of information gain is searched, and using the maximum feature of described information gain as target signature.
Optionally, if described information parameter is information gain, the vehicle information according to each Vehicle Identify Number and described
Vehicle Identify Number set belonging to Vehicle Identify Number calculates the information parameter of each feature, including：
According to the Vehicle Identify Number set belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number, the vehicle frame is calculated
The comentropy of number corresponding vehicle information of set；
Calculate the conditional entropy of each feature；
According to the comentropy of the corresponding vehicle information of the Vehicle Identify Number set and the conditional entropy of the feature, count successively
Calculate the information gain of each feature.
Optionally, it is calculated by the following formula the comentropy of the corresponding vehicle information of the Vehicle Identify Number set：
Wherein, H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；X is each in the Vehicle Identify Number set
The set of the corresponding vehicle information of a Vehicle Identify Number；X is the element in X；P (x) is the probability that x occurs.
Optionally, it is calculated by the following formula the conditional entropy of each feature：
Wherein, H (X/Y) is the conditional entropy of the feature；X is the corresponding vehicle of each Vehicle Identify Number in the Vehicle Identify Number set
The set of information；X is the element in X；Y is the set of circumstances whether occurred for indicating the feature；Y is in set of circumstances
Element；P (x, y) is for indicating the simultaneous probability of x and y；P (x/y) is used to indicate the probability that x occurs when y occurs.
Optionally, it is calculated by the following formula the information gain of each feature：
G (Y)=H (X)-H (X | Y)；
Wherein, G (Y) is the information gain of the feature；H (X) is the letter of the corresponding vehicle information of the Vehicle Identify Number set
Cease entropy；H (X/Y) is the conditional entropy of the feature.
Second aspect, present invention also provides a kind of devices identifying vehicle information according to Vehicle Identify Number, including：
Acquisition module, the Vehicle Identify Number set that the Vehicle Identify Number for obtaining each known vehicle information is constituted；
Divide module, for being split respectively to the Vehicle Identify Number, acquisition builds each feature of the Vehicle Identify Number,
In, each feature includes a character and the character information the location of in the Vehicle Identify Number；
Module, the information parameter for calculating separately each feature are built, and decision tree is built according to described information parameter,
Wherein, the decision node in the decision tree is the feature；
Judgment module, for judging Vehicle Identify Number to be identified according to the decision node of the decision tree, according to judgement
As a result the vehicle information of the Vehicle Identify Number to be identified is determined.
Optionally, the structure module includes：
Ratio computing unit, for counting in same Vehicle Identify Number set, the quantity of the identical Vehicle Identify Number of vehicle information, and
The quantity for calculating the identical Vehicle Identify Number of the vehicle information accounts for the ratio of the Vehicle Identify Number set；
Comparing unit is used for the ratio and predetermined threshold value；
Information parameter computing unit, if being less than the predetermined threshold value for the ratio, according to the vehicle of each Vehicle Identify Number
Vehicle Identify Number set belonging to information and the Vehicle Identify Number, calculates the information parameter of each feature；
Target signature searching unit, for according to described information parameter, searching target signature；
Division unit is used for using the target signature as a decision node, and by the decision node to described
Each Vehicle Identify Number in Vehicle Identify Number set is divided, and is obtained the Vehicle Identify Number set for meeting the decision node and is not met described
The Vehicle Identify Number set of decision node；
It is special to remove target for obtaining respectively in each Vehicle Identify Number set after dividing for residue character set acquiring unit
Residue character set after sign returns if the residue character set is not sky and executes operation by the ratio computing unit.
Optionally, described information parameter is conditional entropy or information gain；
If described information parameter is conditional entropy, the target signature searching unit is used for the feature of search criterion entropy minimum,
And using the feature of conditional entropy minimum as target signature；
If described information parameter is information gain, the target signature searching unit is for searching the maximum spy of information gain
Sign, and using the maximum feature of described information gain as target signature.
Optionally, if described information parameter is information gain, described information parameter calculation unit includes：
Comentropy computing unit, for the vehicle frame belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number
Number set, calculate the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；
Conditional entropy computing unit, the conditional entropy for calculating each feature；
Information gain computing unit, for the comentropy according to the corresponding vehicle information of the Vehicle Identify Number set, Yi Jisuo
The conditional entropy of feature is stated, calculates the information gain of each feature successively.
Optionally, described information entropy computing unit is calculated by the following formula the corresponding vehicle information of the Vehicle Identify Number set
Comentropy：
Wherein, H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；X is each in the Vehicle Identify Number set
The set of the corresponding vehicle information of a Vehicle Identify Number；X is the element in X；P (x) is the probability that x occurs.
Optionally, the conditional entropy computing unit is calculated by the following formula the conditional entropy of each feature：
Wherein, H (X/Y) is the conditional entropy of the feature；X is the corresponding vehicle of each Vehicle Identify Number in the Vehicle Identify Number set
The set of information；X is the element in X；Y is the set of circumstances whether occurred for indicating the feature；Y is in set of circumstances
Element；P (x, y) is for indicating the simultaneous probability of x and y；P (x/y) is used to indicate the probability that x occurs when y occurs.
Optionally, described information gain calculating unit is calculated by the following formula the information gain of each feature：
G (Y)=H (X)-H (X | Y)；
Wherein, G (Y) is the information gain of the feature；H (X) is the letter of the corresponding vehicle information of the Vehicle Identify Number set
Cease entropy；H (X/Y) is the conditional entropy of the feature.
The embodiment of the present application discloses a kind of method and device identifying vehicle information according to Vehicle Identify Number, passes through the application, energy
It is enough that decision tree is created according to the Vehicle Identify Number of each known vehicle information, then by Vehicle Identify Number to be identified according to the decision of the decision tree
Node is judged, the vehicle information of the Vehicle Identify Number to be identified is determined according to judging result, realizes the identification of vehicle information.It should
Method no longer needs the correspondence of manual creation Vehicle Identify Number and vehicle information, therefore compared with prior art, can reduce people
The waste of power.
Description of the drawings
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without having to pay creative labor,
Other drawings may also be obtained based on these drawings.
Fig. 1 is a kind of workflow schematic diagram of method identifying vehicle information according to Vehicle Identify Number disclosed in the present application；
Fig. 2 is to be built according to information parameter in a kind of method identifying vehicle information according to Vehicle Identify Number disclosed in the present application
The workflow schematic diagram of decision tree；
Fig. 3 is the schematic diagram of the decision tree in a kind of method identifying vehicle information according to Vehicle Identify Number disclosed in the present application；
Fig. 4 (1) be it is disclosed in the present application another identify that the decision tree in the method for vehicle information is shown according to Vehicle Identify Number
It is intended to；
Fig. 4 (2) be it is disclosed in the present application another identify that the decision tree in the method for vehicle information is shown according to Vehicle Identify Number
It is intended to；
Fig. 5 is a kind of structural schematic diagram of device identifying vehicle information according to Vehicle Identify Number disclosed in the present application.
Specific implementation mode
In order to need to expend a large amount of manpowers when solving the problem of that the prior art carries out the identification of vehicle information, the application discloses
A kind of method and device identifying vehicle information according to Vehicle Identify Number.
The first embodiment of the application discloses a kind of method identifying vehicle information according to Vehicle Identify Number, referring to Fig. 1, described
Include the following steps according to the method that Vehicle Identify Number identifies vehicle information：
Step S11, the Vehicle Identify Number set that the Vehicle Identify Number of each known vehicle information is constituted is obtained.
Step S12, the Vehicle Identify Number is split respectively, obtains each feature for building the Vehicle Identify Number, wherein is every
A feature includes a character and the character information the location of in the Vehicle Identify Number.
Vehicle Identify Number is made of the character of permutation and combination, and the embodiment of the present application is disclosed to identify vehicle information according to Vehicle Identify Number
It in method, is split respectively according to the location of each character in Vehicle Identify Number, obtained each feature after segmentation includes
One character and the character information the location of in the Vehicle Identify Number.
In addition, in the embodiment of the present application, the feature may be configured as diversified forms.It is described in wherein a kind of form
Form characterized by position-character pair.For example, if some Vehicle Identify Number is " LGG7B2D15DZ034463 ", divide to it
After cutting, the set of obtained each feature composition is represented by { 1-L, 2-G, 3-G, 4-7 ... }.Wherein, " 1-L ", " 2-G ", " 3-
G " and " 4-7 " are the form of " position-character to ", and " 1-L " this feature includes the character " L " of structure Vehicle Identify Number, and
Indicate that the character is in the Vehicle Identify Number first position by location information " 1 ", at " 2-G " this character representation character " 2 "
Second position in the Vehicle Identify Number.
Certainly, the feature may be set to be other forms, for example, the form etc. of position/character pair is could be provided as,
The application does not limit this.
Step S13, the information parameter of each feature is calculated separately, and decision tree is built according to described information parameter, wherein
Decision node in the decision tree is the feature.
In the embodiment of the present application, according to the information parameter of each feature, it is determined to the feature as decision node, and lead to
Cross decision node structure decision tree.
Step S14, Vehicle Identify Number to be identified is judged according to the decision node of the decision tree, it is true according to judging result
The vehicle information of the fixed Vehicle Identify Number to be identified.
When needing to carry out the identification of vehicle information to Vehicle Identify Number to be identified, in the embodiment of the present application, by vehicle to be identified
Frame number traverses decision tree, is judged according to the decision node of the decision tree, the vehicle to be identified is determined according to judging result
The matched each decision node of frame number institute, then obtain the Vehicle Identify Number to be identified corresponding to matched each decision node
Vehicle information, which is the vehicle information of the Vehicle Identify Number to be identified, to realize the identification of vehicle information.
The embodiment of the present application discloses a kind of method identifying vehicle information according to Vehicle Identify Number, in this way, being capable of basis
The Vehicle Identify Number of each known vehicle information creates decision tree, then by Vehicle Identify Number to be identified according to the decision tree decision node into
Row judges, the vehicle information of the Vehicle Identify Number to be identified is determined according to judging result, realizes the identification of vehicle information.This method is not
The correspondence of manual creation Vehicle Identify Number and vehicle information is needed again, therefore compared with prior art, can reduce the wave of manpower
Take.
As the refinement of Fig. 1 methods, in second embodiment provided by the present application, as shown in Fig. 2, described in step S13
The information parameter of each feature is calculated separately, and decision tree is built according to described information parameter, is included the following steps：
Step S21, statistics is in same Vehicle Identify Number set, the quantity of the identical Vehicle Identify Number of vehicle information, and described in calculating
The quantity of the identical Vehicle Identify Number of vehicle information accounts for the ratio of the Vehicle Identify Number set.
For example, in the same Vehicle Identify Number set, there are 10 Vehicle Identify Numbers, wherein the vehicle information of 6 Vehicle Identify Numbers is identical, then
The quantity of the identical Vehicle Identify Number of vehicle information is 0.6 in the ratio of the Vehicle Identify Number set.
Step S22, ratio and predetermined threshold value described in comparison, judge whether the ratio is less than the predetermined threshold value, if so,
The operation of step S23 is executed, if it is not, executing the operation of step S27.
Wherein, the predetermined threshold value is the numerical value less than 1, for example, can 0.7 be set as the predetermined threshold value.
If step S23, the described ratio is less than the predetermined threshold value, according to the vehicle information of each Vehicle Identify Number and described
Vehicle Identify Number set belonging to Vehicle Identify Number, calculates the information parameter of each feature.
Step S24, according to described information parameter, target signature is searched.
In the embodiment of the present application, information parameter is a kind of parameter for weighing sample characteristics importance, can be a variety of
Form.For example, described information parameter can be conditional entropy or information parameter.
Step S25, using the target signature as a decision node, and by the decision node to the Vehicle Identify Number
Each Vehicle Identify Number in set is divided, and is obtained the Vehicle Identify Number set for meeting the decision node and is not met the decision section
The Vehicle Identify Number set of point.
Wherein, the Vehicle Identify Number set for meeting the decision node refers to including the Vehicle Identify Number set of target signature, is not inconsistent
The Vehicle Identify Number set for closing the decision node refers to the Vehicle Identify Number set not comprising the target signature.
Step S26, it obtains respectively in each Vehicle Identify Number set after dividing, removes the residue character collection after target signature
It closes, and judges whether the residue character set is empty, if it is not, the operation of S21 is returned to step, if so, executing step S27
Operation.
If after removing target signature, the residue character set that remaining feature is constituted not is sky, then returns to step
The operation of S21.Due to the operation by step S25, the divided Vehicle Identify Number set for meeting the decision node and it is not inconsistent
Close the Vehicle Identify Number set of the decision node.If the Vehicle Identify Number collection that setting meets the decision node is combined into first set, it is not inconsistent
The Vehicle Identify Number collection for closing the decision node is combined into second set, in this case, when returning to step the operation of S21, then needs
It counts in the first aggregate, the quantity of the identical Vehicle Identify Number of vehicle information, and calculates the number of the identical Vehicle Identify Number of vehicle information
Amount accounts for the ratio of first set, and counts in second set, the quantity of the identical Vehicle Identify Number of vehicle information, and calculates vehicle
The quantity of the identical Vehicle Identify Number of information accounts for the ratio of second set.
Step S27, terminate to be the vehicle frame set search decision node.
It is set in a certain Vehicle Identify Number set, the vehicle information belonging to the identical Vehicle Identify Number of vehicle information is believed for target vehicle
Breath.If by the judgement of step S22, determines that the ratio is not less than the predetermined threshold value, then shows in the Vehicle Identify Number set,
Vehicle Identify Number is that the possibility of target vehicle information is larger, correspondingly, Vehicle Identify Number to be identified belongs to the possibility of target vehicle information
It is larger.In this case, the operation for executing step S27, is no longer the Vehicle Identify Number set search decision node, if what is got waits for
Identification Vehicle Identify Number belongs to the Vehicle Identify Number set, then it is assumed that the vehicle information of the Vehicle Identify Number is target vehicle information.
If in addition, the ratio is less than predetermined threshold value, show in the Vehicle Identify Number set, the distribution of each vehicle information
It is more dispersed, in this case, then the operation of step S23 is further executed, further to determine decision node according to feature.
If by the operation of step S26, determine in a certain Vehicle Identify Number set after dividing, it is surplus after removing target signature
Remaining characteristic set is sky, then illustrates not including the feature that can be used as decision node in the Vehicle Identify Number set, then terminate as the vehicle
Frame set search decision node.
Step S21 to step S27 discloses a kind of information parameter calculating separately each feature, and according to described information parameter
The method for building decision tree.In this method, the quantity that the identical Vehicle Identify Number of vehicle information can be calculated accounts for Vehicle Identify Number set
Ratio, if the ratio be not less than predetermined threshold value, then it is assumed that concentration is compared in the distribution of vehicle information in the Vehicle Identify Number, is no longer this
Vehicle Identify Number set search decision node；If the ratio obtains decision less than predetermined threshold value according to the information parameter of each feature
Node, and build corresponding decision tree.
Wherein, described information parameter is conditional entropy or information gain.If described information parameter is conditional entropy, described according to institute
Information parameter is stated, searching target signature includes：The feature of search criterion entropy minimum, and using the feature of conditional entropy minimum as
Target signature.If described information parameter is information gain, described according to described information parameter, searching target signature includes：It searches
The maximum feature of information gain, and using the maximum feature of described information gain as target signature.
The second embodiment of the application is disclosed according to belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number
Vehicle Identify Number set calculates the operation of the information parameter of each feature, if described information parameter is information gain, which usually wraps
Include following steps：
First, the Vehicle Identify Number set belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number, described in calculating
The comentropy of the corresponding vehicle information of Vehicle Identify Number set.
Then, the conditional entropy of each feature is calculated.
Finally, according to the comentropy of the corresponding vehicle information of the Vehicle Identify Number set and the conditional entropy of the feature, according to
The secondary information gain for calculating each feature.
Inside information theory, entropy is to probabilistic measurement, and entropy is higher, then can transmit more information, and entropy is lower,
Then mean that the information of transmission is fewer, entropy degree has weighed the uncertainty of system.
In the embodiment of the present application, usually it is calculated by the following formula the letter of the corresponding vehicle information of the Vehicle Identify Number set
Cease entropy：
Wherein, H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；X is each in the Vehicle Identify Number set
The set of the corresponding vehicle information of a Vehicle Identify Number；X is the element in X；P (x) is the probability that x occurs.In formula (1), logP
(truth of a matter of (x) is usually 2.
In addition, in the embodiment of the present application, being usually calculated by the following formula the conditional entropy of each feature：
Wherein, H (X/Y) is the conditional entropy of the feature；X is the corresponding vehicle of each Vehicle Identify Number in the Vehicle Identify Number set
The set of information；X is the element in X；Y is the set of circumstances whether occurred for indicating the feature；Y is in set of circumstances
Element；P (x, y) is for indicating the simultaneous probability of x and y；P (x/y) is used to indicate the probability that x occurs when y occurs.In addition exist
In formula (2), the truth of a matter of logP (x/y) is usually 2.
Information gain is a kind of parameter for weighing sample characteristics importance.In the embodiment of the present application, can usually lead to
Cross the information gain that following formula calculates each feature：
G (Y)=H (X)-H (X | Y) (3)；
Wherein, G (Y) is the information gain of the feature；H (X) is the letter of the corresponding vehicle information of the Vehicle Identify Number set
Cease entropy；H (X/Y) is the conditional entropy of the feature.
In addition, if described information parameter is conditional entropy, the vehicle information according to each Vehicle Identify Number and the vehicle frame
Vehicle Identify Number set belonging to number, calculates the operation of the information parameter of each feature, can directly be realized by formula (2).
For clear scheme disclosed in the present application, below by way of a specific example to basis disclosed in the present application
The method of Vehicle Identify Number identification vehicle information is explained in detail.
In this example, the Vehicle Identify Number set got by step S11 can be indicated by table 1：
Table 1
| Vehicle Identify Number | Vehicle information |
| AABBCCDD | M |
| CAVVDEFC | N |
| ACBCDEFC | M |
| ABBCDEFC | M |
| BAVCDEFC | N |
Wherein, the Vehicle Identify Number of each known vehicle information in the Vehicle Identify Number is respectively：" AABBCCDD " (Vehicle Identify Number
Vehicle information is M), " CAVVDEFC " (the vehicle information of the Vehicle Identify Number is N), " ACBCDEFC " (vehicle information of the Vehicle Identify Number
For M), " ABBCDEFC " (the vehicle information of the Vehicle Identify Number is M) and " BAVCDEFC " (the vehicle information of the Vehicle Identify Number is N).
By upper table it is found that in formula (1), X is the corresponding vehicle information of each Vehicle Identify Number in the Vehicle Identify Number set
Set, then X is the set that vehicle information M and vehicle information N is constituted.And x is the element in X, then x is vehicle information M or vehicle
Type information N.Due to including 5 Vehicle Identify Numbers in upper table, vehicle information M appearance 3 times, vehicle information N appearance 2 times, then x is that vehicle is believed
When ceasing M, P (x) be 0.6, x when being vehicle information N, and P (x) is 0.4.In this case, vehicle information is calculated according to formula (1)
Comentropy is：
In addition, after being split to each Vehicle Identify Number in table 1, each feature difference of the structure Vehicle Identify Number got
For：1-A, 1-B, 1-C, 2-A, 2-B, 2-C, 3-B, 3-V, 4-B, 4-V, 4-C, 5-D, 5-C, 6-C, 6-E, 7-D, 7-F, 8-D and
8-C。
After each feature for obtaining structure Vehicle Identify Number, the conditional entropy of each feature is calculated according to formula (2).Formula
(2) it is：
Wherein, H (X/Y) indicates the conditional entropy of a certain feature；X is described
The set of the corresponding vehicle information of each Vehicle Identify Number in Vehicle Identify Number set, then X is the collection that vehicle information M and vehicle information N are constituted
It closes.And x is the element in X, then x is vehicle information M or vehicle information N.In addition, Y is for indicating whether the feature occurs
Set of circumstances, y be set of circumstances in element, then for this feature occur or this feature do not occur.In addition, P (x, y) is used for table
Show the simultaneous probability of x and y；P (x/y) is used to indicate the probability that x occurs when y occurs.
In this case, it to the conditional entropy of this feature of calculating 1-A, then can be calculated in the following manner：
After conditional entropy is calculated according to above formula, further calculate to obtain the information gain of each feature according to formula (3).
In this example, according to result of calculation, it may be determined that information gain is maximum to be characterized as 1-A, therefore using 1-A this feature as mesh
Feature is marked, using target signature 1-A as a decision node, and by the decision node to each vehicle frame in Vehicle Identify Number set
It number is divided, to obtain meeting target signature 1-A and not meet two groups of Vehicle Identify Number set of target signature 1-A.
Using the Vehicle Identify Number set for meeting target signature 1-A as first set, the Vehicle Identify Number of target signature 1-A will not met
Set is used as second set.In this example, it can be seen that the vehicle information of each Vehicle Identify Number in the first aggregate is vehicle
Type information M, i.e. it is 1 that the quantity of the identical Vehicle Identify Number of vehicle information, which accounts for the ratio of first set, since predetermined threshold value is less than 1, then
The ratio that the quantity of the identical Vehicle Identify Number of vehicle information accounts for first set is certain to be more than predetermined threshold value, then terminates to be described first
Set search decision node.Moreover, the vehicle information of each Vehicle Identify Number in second set is vehicle information N, i.e. vehicle
The ratio that the quantity of the identical Vehicle Identify Number of information accounts for second set is 1, and since predetermined threshold value is less than 1, then vehicle information is identical
The ratio that the quantity of Vehicle Identify Number accounts for second set is certain to be more than predetermined threshold value, then terminates to search decision section for the second set
Point.In the finally obtained decision tree of the example, only comprising this decision node of feature 1-A, the decision tree built accordingly such as Fig. 3 institutes
Show.
If the Vehicle Identify Number to be identified got is CJJKSDDF, it is judged according to the decision node of decision tree,
That is judging whether the Vehicle Identify Number to be identified meets feature 1-A.Since the Vehicle Identify Number first character to be identified is C, then should
Vehicle Identify Number to be identified does not meet feature 1-A.By decision tree it is found that the vehicle information for not meeting the Vehicle Identify Number of feature 1-A is N,
It then can determine that the vehicle information of Vehicle Identify Number to be identified is N, realize the identification to the Vehicle Identify Number to be identified.
In another example, the Vehicle Identify Number set such as table that the Vehicle Identify Number of each known vehicle information got is constituted is set
Shown in 2：
Table 2
Wherein, in upper table latter six of each Vehicle Identify Number be digital number, it is usually unrelated with the identification of vehicle information.Removal
Latter six of each Vehicle Identify Number, and each Vehicle Identify Number is split, the feature got includes：1-L、2-S、3-G、4-J、5-
A, 5-S, 5-T, 6-5,7-2,8-H, 8-U, 9-D, 9-E, 9-C, 9-5,9-6,9-7,9-8,10-H and 10-S.
By calculating, in the Vehicle Identify Number set of upper table, the quantity of the identical Vehicle Identify Number of vehicle information accounts for Vehicle Identify Number set
Ratio is respectively less than predetermined threshold value, then continues the information parameter for calculating each feature.
According to formula (2), the conditional entropy for each feature being calculated is as shown in table 3：
Table 3
| Feature | Conditional entropy | Feature | Conditional entropy |
| 1-L | 2 | 8-U | 1.188721876 |
| 2-S | 2 | 9-D | 1.387710777 |
| 3-G | 2 | 9-E | 1.935268786 |
| 4-J | 2 | 9-C | 1.188721876 |
| 5-A | 1 | 9-5 | 1.886668333 |
| 5-S | 1.188721876 | 9-6 | 1.730730038 |
| 5-T | 1.188721876 | 9-7 | 1.865529464 |
| 6-5 | 2 | 9-8 | 1.619618368 |
| 7-2 | 2 | 10-H | 1.912246333 |
| 8-H | 1.188721876 | 10-S | 1.912246333 |
According to upper table it is found that the conditional entropy of feature 5-A is minimum, then this feature is target signature, and this feature can be used as one
Decision node.Shown in the decision tree built accordingly such as Fig. 4 (1), and Vehicle Identify Number is divided according to the decision tree, to obtain
Meet the Vehicle Identify Number set of feature 5-A, and does not meet the Vehicle Identify Number set of feature 5-A.
In addition, removal target signature, that is, after removing feature 5-A, the left sibling data of the decision tree meet feature 5-A
Vehicle Identify Number set it is as shown in table 4：
Table 4
Correspondingly, removal target signature, that is, after removing feature 5-A, the right node data of the decision tree does not meet feature
The Vehicle Identify Number set of 5-A is as shown in table 5：
Table 5
According to above-mentioned steps, left sibling data (the Vehicle Identify Number set for meeting feature 5-A) and right number of nodes are calculated separately
According to the information parameter of (the Vehicle Identify Number set for not meeting feature 5-A).By calculating it is found that in left sibling data, information increases
Beneficial maximum (i.e. conditional entropy is minimum) is characterized as 8-H, then the decision node of left sibling data is 8-H.In addition, in right node data
In, information gain maximum (i.e. conditional entropy is minimum) is characterized as 5-S, then the decision node of right node data is 5-S.
Continue to divide left sibling data and right node data according to decision node 8-H and decision node 5-S.It divides
Afterwards, it is known that the vehicle information for meeting the Vehicle Identify Number of decision node 8-H is that more/2013 section 1.5 of Buick/triumphant is automatic classic,
That is in the Vehicle Identify Number set for the Vehicle Identify Number structure for meeting decision node 8-H, vehicle information is more/2013 sections of Buick/triumphant
The ratio that the quantity of 1.5 automatic classic Vehicle Identify Numbers accounts for the Vehicle Identify Number set is 1, which is necessarily greater than predetermined threshold value, then
Terminate to be the Vehicle Identify Number set search decision node for meeting the Vehicle Identify Number of decision node 8-H and building.In addition, not met certainly after dividing
The vehicle information of the Vehicle Identify Number of plan node 8-H is Buick/triumphant more/2011 section of 1.6 automatic LX, that is to say, that is not being met certainly
In the Vehicle Identify Number set of the Vehicle Identify Number structure of plan node 8-H, vehicle information is Buick/triumphant more vehicle frame of/2011 section of 1.6 automatic LX
Number quantity account for the Vehicle Identify Number set ratio be 1, which is necessarily greater than predetermined threshold value, then terminates not meet decision node
The Vehicle Identify Number set search decision node of the Vehicle Identify Number structure of 8-H.
In addition, after division, it is known that the vehicle information for meeting the Vehicle Identify Number of decision node 5-S is Buick/triumphant more/2005
1.6 manual LE comfort types of money, that is to say, that in the Vehicle Identify Number set for the Vehicle Identify Number structure for meeting decision node 5-S, vehicle letter
Breath be that account for the ratio of the Vehicle Identify Number set be 1 to the quantity for the Vehicle Identify Number that Buick/triumphant gets over/2005 section of 1.6 manual LE comfort type, the ratio
Example is necessarily greater than predetermined threshold value, then terminates the Vehicle Identify Number set search decision section of the Vehicle Identify Number structure to meet decision node 5-S
Point.In addition, the vehicle information for the Vehicle Identify Number for not meeting decision node 5-S after dividing is that more/2006 section 1.6 of Buick/triumphant is automatic
LE comfort types, that is to say, that in the Vehicle Identify Number set for the Vehicle Identify Number structure for not meeting decision node 5-S, vehicle information is other
Gram/it is triumphant more/2006 section of 1.6 automatic LE comfort type Vehicle Identify Number quantity account for the Vehicle Identify Number set ratio be 1, which must
More than predetermined threshold value, then terminate the Vehicle Identify Number set search decision node of the Vehicle Identify Number structure not meet decision node 5-S.
According to above-mentioned steps, shown in decision tree such as Fig. 4 (2) of structure.After obtaining Vehicle Identify Number to be identified, by that will wait for
Identification Vehicle Identify Number is judged according to the decision node of the decision tree, you can determines the vehicle information of the Vehicle Identify Number to be identified.
Correspondingly, in another embodiment of the application, a kind of device identifying vehicle information according to Vehicle Identify Number is also disclosed.Ginseng
Structural schematic diagram as shown in Figure 5, it is described to identify that the device of vehicle information includes according to Vehicle Identify Number：Acquisition module 100, segmentation mould
Block 200, structure module 300 and judgment module 400.
Wherein, the acquisition module 100, the Vehicle Identify Number collection that the Vehicle Identify Number for obtaining each known vehicle information is constituted
It closes.
The segmentation module 200 obtains for being split respectively to the Vehicle Identify Number and builds each of the Vehicle Identify Number
Feature, wherein each feature includes a character and the character information the location of in the Vehicle Identify Number.
Vehicle Identify Number is made of the character of permutation and combination, and the embodiment of the present application is disclosed to identify vehicle information according to Vehicle Identify Number
It in method, is split respectively according to the location of each character in Vehicle Identify Number, obtained each feature after segmentation includes
One character and the character information the location of in the Vehicle Identify Number.
In addition, in the embodiment of the present application, the feature may be configured as diversified forms.It is described in wherein a kind of form
Form characterized by position-character pair.For example, if some Vehicle Identify Number is " LGG7B2D15DZ034463 ", divide to it
After cutting, the set of obtained each feature composition is represented by { 1-L, 2-G, 3-G, 4-7 ... }.Wherein, " 1-L ", " 2-G ", " 3-
G " and " 4-7 " are the form of " position-character to ", and " 1-L " this feature includes the character " L " of structure Vehicle Identify Number, and
Indicate that the character is in the Vehicle Identify Number first position by location information " 1 ", at " 2-G " this character representation character " 2 "
Second position in the Vehicle Identify Number.
Certainly, the feature may be set to be other forms, for example, the form etc. of position/character pair is could be provided as,
The application does not limit this.
The structure module 300, the information parameter for calculating separately each feature, and built according to described information parameter
Decision tree, wherein the decision node in the decision tree is the feature.
The judgment module 400, for Vehicle Identify Number to be identified to be judged according to the decision node of the decision tree, root
It is judged that result determines the vehicle information of the Vehicle Identify Number to be identified.
When needing to carry out the identification of vehicle information to Vehicle Identify Number to be identified, in the embodiment of the present application, by vehicle to be identified
Frame number traverses decision tree, is judged according to the decision node of the decision tree, the vehicle to be identified is determined according to judging result
The matched each decision node of frame number institute, then obtain the Vehicle Identify Number to be identified corresponding to matched each decision node
Vehicle information, which is the vehicle information of the Vehicle Identify Number to be identified, to realize the identification of vehicle information.
The embodiment of the present application disclose it is a kind of according to Vehicle Identify Number identify vehicle information device being capable of basis by the device
The Vehicle Identify Number of each known vehicle information creates decision tree, then by Vehicle Identify Number to be identified according to the decision tree decision node into
Row judges, the vehicle information of the Vehicle Identify Number to be identified is determined according to judging result, realizes the identification of vehicle information, to no longer
The correspondence of manual creation Vehicle Identify Number and vehicle information is needed, therefore compared with prior art, the waste of manpower can be reduced.
Further, in the device for identifying vehicle information according to Vehicle Identify Number, the structure module 300 includes：
Ratio computing unit, for counting in same Vehicle Identify Number set, the quantity of the identical Vehicle Identify Number of vehicle information, and
The quantity for calculating the identical Vehicle Identify Number of the vehicle information accounts for the ratio of the Vehicle Identify Number set；
Comparing unit is used for the ratio and predetermined threshold value, wherein the predetermined threshold value is a number less than 1
Value, for example, can 0.7 be set as the predetermined threshold value；
Information parameter computing unit, if being less than the predetermined threshold value for the ratio, according to the vehicle of each Vehicle Identify Number
Vehicle Identify Number set belonging to information and the Vehicle Identify Number, calculates the information parameter of each feature；
Target signature searching unit, for according to described information parameter, searching target signature；
Division unit is used for using the target signature as a decision node, and by the decision node to described
Each Vehicle Identify Number in Vehicle Identify Number set is divided, and is obtained the Vehicle Identify Number set for meeting the decision node and is not met described
The Vehicle Identify Number set of decision node, wherein the Vehicle Identify Number set for meeting the decision node refers to including the vehicle of target signature
Frame number is gathered, and the Vehicle Identify Number set for not meeting the decision node refers to the Vehicle Identify Number set not comprising the target signature；
It is special to remove target for obtaining respectively in each Vehicle Identify Number set after dividing for residue character set acquiring unit
Residue character set after sign returns if the residue character set is not sky and executes operation by the ratio computing unit.
By said units, the information parameter of each feature can be calculated separately, and determine according to described information parameter structure
Plan tree.Wherein, the quantity that the identical Vehicle Identify Number of vehicle information can be calculated accounts for the ratio of Vehicle Identify Number set, if the ratio is not
Less than predetermined threshold value, then it is assumed that concentration is compared in the distribution of vehicle information in the Vehicle Identify Number, no longer determines for the Vehicle Identify Number set search
Plan node；If the ratio obtains decision node less than predetermined threshold value according to the information gain information parameter of each feature, and
Build corresponding decision tree.
In the device for identifying vehicle information according to Vehicle Identify Number disclosed in the embodiment of the present application, described information parameter is
Conditional entropy or information gain.
If described information parameter is conditional entropy, the target signature searching unit is used for the feature of search criterion entropy minimum,
And using the feature of conditional entropy minimum as target signature；If described information parameter is information gain, the target signature is looked into
Look for unit for searching the maximum feature of information gain, and using the maximum feature of described information gain as target signature.
Further, in the device for identifying vehicle information disclosed in the embodiment of the present application according to Vehicle Identify Number, if the letter
Breath parameter is information gain, and described information parameter calculation unit includes：
Comentropy computing unit, for the vehicle frame belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number
Number set, calculate the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；
Conditional entropy computing unit, the conditional entropy for calculating each feature；
Information gain computing unit, for the comentropy according to the corresponding vehicle information of the Vehicle Identify Number set, Yi Jisuo
The conditional entropy of feature is stated, calculates the information gain of each feature successively.
In the embodiment of the present application, described information entropy computing unit is calculated by the following formula the Vehicle Identify Number set and corresponds to
Vehicle information comentropy：
Wherein, H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；X is each in the Vehicle Identify Number set
The set of the corresponding vehicle information of a Vehicle Identify Number；X is the element in X；P (x) is the probability that x occurs.
In addition, in the embodiment of the present application, the conditional entropy computing unit is calculated by the following formula the item of each feature
Part entropy：
Wherein, H (X/Y) is the conditional entropy of the feature；X is the corresponding vehicle of each Vehicle Identify Number in the Vehicle Identify Number set
The set of information；X is the element in X；Y is the set of circumstances whether occurred for indicating the feature；Y is in set of circumstances
Element；P (x, y) is for indicating the simultaneous probability of x and y；P (x/y) is used to indicate the probability that x occurs when y occurs.
In addition, in the embodiment of the present application, described information gain calculating unit is calculated by the following formula each feature
Information gain：
G (Y)=H (X)-H (X | Y) (3)；
Wherein, G (Y) is the information gain of the feature；H (X) is the letter of the corresponding vehicle information of the Vehicle Identify Number set
Cease entropy；H (X/Y) is the conditional entropy of the feature.
Further, if described information parameter is conditional entropy, it is single that a kind of information parameter calculating is also disclosed in the embodiment of the present application
Member, the information parameter computing unit pass through formula (2) design conditions entropy.
In the specific implementation, the present invention also provides a kind of computer storage medias, wherein the computer storage media can store
There is program, step that some or all of which may include when executing in each embodiment of method of calling provided by the invention.Institute
The storage medium stated can be magnetic disc, CD, read-only memory (English：Read-only memory, referred to as：ROM) or with
Machine storage memory (English：Random access memory, referred to as：RAM) etc..
It is required that those skilled in the art can be understood that the technology in the embodiment of the present invention can add by software
The mode of general hardware platform realize.Based on this understanding, the technical solution in the embodiment of the present invention substantially or
Say that the part that contributes to existing technology can be expressed in the form of software products, which can deposit
Storage is in storage medium, such as ROM/RAM, magnetic disc, CD, including some instructions are used so that computer equipment (can be with
Be personal computer, server either network equipment etc.) execute certain part institutes of each embodiment of the present invention or embodiment
The method stated.
The same or similar parts between the embodiments can be referred to each other in this specification.Especially for ... implement
For example, since it is substantially similar to the method embodiment, so description is fairly simple, related place is referring in embodiment of the method
Explanation.
Invention described above embodiment is not intended to limit the scope of the present invention..
Claims (14)
1. a kind of method identifying vehicle information according to Vehicle Identify Number, which is characterized in that including：
Obtain the Vehicle Identify Number set that the Vehicle Identify Number of each known vehicle information is constituted；
The Vehicle Identify Number is split respectively, obtains each feature for building the Vehicle Identify Number, wherein each feature includes one
A character and the character information the location of in the Vehicle Identify Number；
The information parameter of each feature is calculated separately, and decision tree is built according to described information parameter, wherein in the decision tree
Decision node be the feature；
Vehicle Identify Number to be identified is judged according to the decision node of the decision tree, is determined according to judging result described to be identified
The vehicle information of Vehicle Identify Number.
2. the method according to claim 1 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that described to calculate separately
The information parameter of each feature, and decision tree is built according to described information parameter, including：
21) statistics is in same Vehicle Identify Number set, the quantity of the identical Vehicle Identify Number of vehicle information, and calculates the vehicle information phase
The quantity of same Vehicle Identify Number accounts for the ratio of the Vehicle Identify Number set；
22) ratio and predetermined threshold value；
If 23) ratio is less than the predetermined threshold value, belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number
Vehicle Identify Number set, calculate the information parameter of each feature；
24) according to described information parameter, target signature is searched；
25) using the target signature as a decision node, and by the decision node in the Vehicle Identify Number set
Each Vehicle Identify Number is divided, and acquisition meets the Vehicle Identify Number set of the decision node and do not meet the vehicle frame of the decision node
Number set；
26) it obtains respectively in each Vehicle Identify Number set after dividing, the residue character set after target signature is removed, if described
Residue character set is not sky, returns to step operation 21).
3. the method according to claim 2 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that
Described information parameter is conditional entropy or information gain；
If described information parameter is conditional entropy, described according to described information parameter, searching target signature includes：
The feature of search criterion entropy minimum, and using the feature of conditional entropy minimum as target signature；
If described information parameter is information gain, described according to described information parameter, searching target signature includes：
The maximum feature of information gain is searched, and using the maximum feature of described information gain as target signature.
4. the method according to claim 3 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that if described information is joined
Number is information gain, and the Vehicle Identify Number set belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number calculates
The information parameter of each feature, including：
According to the Vehicle Identify Number set belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number, the Vehicle Identify Number collection is calculated
Close the comentropy of corresponding vehicle information；
Calculate the conditional entropy of each feature；
According to the comentropy of the corresponding vehicle information of the Vehicle Identify Number set and the conditional entropy of the feature, calculate successively each
The information gain of a feature.
5. the method according to claim 4 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that pass through following formula
Calculate the comentropy of the corresponding vehicle information of the Vehicle Identify Number set：
Wherein, H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；X is each vehicle in the Vehicle Identify Number set
The set of the corresponding vehicle information of frame number；X is the element in X；P (x) is the probability that x occurs.
6. the method according to claim 4 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that pass through following formula
Calculate the conditional entropy of each feature：
Wherein, H (X/Y) is the conditional entropy of the feature；X is the corresponding vehicle information of each Vehicle Identify Number in the Vehicle Identify Number set
Set；X is the element in X；Y is the set of circumstances whether occurred for indicating the feature；Y is the member in set of circumstances
Element；P (x, y) is for indicating the simultaneous probability of x and y；P (x/y) is used to indicate the probability that x occurs when y occurs.
7. the method according to claim 4 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that pass through following formula
Calculate the information gain of each feature：
G (Y)=H (X)-H (X | Y)；
Wherein, G (Y) is the information gain of the feature；H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；
H (X/Y) is the conditional entropy of the feature.
8. a kind of device identifying vehicle information according to Vehicle Identify Number, which is characterized in that including：
Acquisition module, the Vehicle Identify Number set that the Vehicle Identify Number for obtaining each known vehicle information is constituted；
Divide module, for being split respectively to the Vehicle Identify Number, obtain each feature for building the Vehicle Identify Number, wherein
Each feature includes a character and the character information the location of in the Vehicle Identify Number；
Module, the information parameter for calculating separately each feature are built, and decision tree is built according to described information parameter,
In, the decision node in the decision tree is the feature；
Judgment module, for judging Vehicle Identify Number to be identified according to the decision node of the decision tree, according to judging result
Determine the vehicle information of the Vehicle Identify Number to be identified.
9. the device according to claim 8 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that the structure module
Including：
Ratio computing unit, for counting in same Vehicle Identify Number set, the quantity of the identical Vehicle Identify Number of vehicle information, and calculate
The quantity of the identical Vehicle Identify Number of the vehicle information accounts for the ratio of the Vehicle Identify Number set；
Comparing unit is used for the ratio and predetermined threshold value；
Information parameter computing unit, if being less than the predetermined threshold value for the ratio, according to the vehicle information of each Vehicle Identify Number,
And the Vehicle Identify Number set belonging to the Vehicle Identify Number, calculate the information parameter of each feature；
Target signature searching unit, for according to described information parameter, searching target signature；
Division unit is used for using the target signature as a decision node, and by the decision node to the vehicle frame
Number set in each Vehicle Identify Number divided, obtain meet the decision node Vehicle Identify Number set and do not meet the decision
The Vehicle Identify Number set of node；
Residue character set acquiring unit, for respectively obtain divide after each Vehicle Identify Number set in, remove target signature it
Residue character set afterwards returns if the residue character set is not sky and executes operation by the ratio computing unit.
10. the device according to claim 9 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that
Described information parameter is conditional entropy or information gain；
If described information parameter is conditional entropy, the target signature searching unit is used for the feature of search criterion entropy minimum, and will
The feature of the conditional entropy minimum is as target signature；
If described information parameter is information gain, the target signature searching unit is used to search the maximum feature of information gain,
And using the maximum feature of described information gain as target signature.
11. the device according to claim 9 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that if described information
Parameter is information gain, and described information parameter calculation unit includes：
Comentropy computing unit, for the Vehicle Identify Number collection belonging to the vehicle information of each Vehicle Identify Number and the Vehicle Identify Number
It closes, calculates the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；
Conditional entropy computing unit, the conditional entropy for calculating each feature；
Information gain computing unit is used for the comentropy according to the corresponding vehicle information of the Vehicle Identify Number set and the spy
The conditional entropy of sign calculates the information gain of each feature successively.
12. the device according to claim 11 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that described information entropy
Computing unit is calculated by the following formula the comentropy of the corresponding vehicle information of the Vehicle Identify Number set：
Wherein, H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；X is each vehicle in the Vehicle Identify Number set
The set of the corresponding vehicle information of frame number；X is the element in X；P (x) is the probability that x occurs.
13. the device according to claim 11 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that the conditional entropy
Computing unit is calculated by the following formula the conditional entropy of each feature：
Wherein, H (X/Y) is the conditional entropy of the feature；X is the corresponding vehicle information of each Vehicle Identify Number in the Vehicle Identify Number set
Set；X is the element in X；Y is the set of circumstances whether occurred for indicating the feature；Y is the member in set of circumstances
Element；P (x, y) is for indicating the simultaneous probability of x and y；P (x/y) is used to indicate the probability that x occurs when y occurs.
14. the device according to claim 11 for identifying vehicle information according to Vehicle Identify Number, which is characterized in that described information increases
Beneficial computing unit is calculated by the following formula the information gain of each feature：
G (Y)=H (X)-H (X | Y)；
Wherein, G (Y) is the information gain of the feature；H (X) is the comentropy of the corresponding vehicle information of the Vehicle Identify Number set；
H (X/Y) is the conditional entropy of the feature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710149934.5A CN108573198A (en) | 2017-03-14 | 2017-03-14 | A kind of method and device identifying vehicle information according to Vehicle Identify Number |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710149934.5A CN108573198A (en) | 2017-03-14 | 2017-03-14 | A kind of method and device identifying vehicle information according to Vehicle Identify Number |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108573198A true CN108573198A (en) | 2018-09-25 |
Family
ID=63578344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710149934.5A Pending CN108573198A (en) | 2017-03-14 | 2017-03-14 | A kind of method and device identifying vehicle information according to Vehicle Identify Number |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108573198A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110070092A (en) * | 2019-03-15 | 2019-07-30 | 平安科技（深圳）有限公司 | Model recognizing method, device, computer equipment and storage medium |
| CN110334586A (en) * | 2019-05-22 | 2019-10-15 | 深圳壹账通智能科技有限公司 | A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing |
| 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 |
| CN115408580A (en) * | 2022-08-31 | 2022-11-29 | 广东数鼎科技有限公司 | Vehicle source model identification method and device |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101261722A (en) * | 2008-01-17 | 2008-09-10 | 北京航空航天大学 | Electronic police background intelligent management and automatic implementation system |
| CN102509090A (en) * | 2011-11-29 | 2012-06-20 | 冷明 | Vehicle feature recognition device based on public security video images in skynet engineering |
| CN202563521U (en) * | 2011-11-29 | 2012-11-28 | 冷明 | Vehicle characteristic identification device based on public security video image in skynet engineering |
| CN103049498A (en) * | 2012-12-07 | 2013-04-17 | 王志欣 | Method for recognizing vehicle type through vehicle identification number (VIN) |
| CN104216391A (en) * | 2013-05-31 | 2014-12-17 | 广州汽车集团股份有限公司 | Automobile decoder and automotive type recognition method thereof |
| CN104488004A (en) * | 2012-05-23 | 2015-04-01 | 实耐宝公司 | Methods and systems for providing vehicle repair information |
| CN104537010A (en) * | 2014-12-17 | 2015-04-22 | 温州大学 | Component classifying method based on net establishing software of decision tree |
| CN104572854A (en) * | 2014-12-17 | 2015-04-29 | 语联网（武汉）信息技术有限公司 | Decision-tree-based translator classification method |
| CN105205468A (en) * | 2015-09-24 | 2015-12-30 | 上海车音网络科技有限公司 | Vehicle washing device, system and method capable of automatically identifying vehicle types |
| CN105320708A (en) * | 2014-08-05 | 2016-02-10 | 北京大学 | Establishment method of vehicle model database and server |
| CN105335760A (en) * | 2015-11-16 | 2016-02-17 | 南京邮电大学 | Image number character recognition method |
| CN106021545A (en) * | 2016-05-27 | 2016-10-12 | 大连楼兰科技股份有限公司 | Method for remote diagnoses of cars and retrieval of spare parts |
| CN106056221A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | FP-Tree sequence pattern mining and fault code classification-based vehicle remote diagnosis and spare part retrieval method |
| CN106054858A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | Decision tree classification and fault code classification-based vehicle remote diagnosis and spare part retrieval method |
-
2017
- 2017-03-14 CN CN201710149934.5A patent/CN108573198A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101261722A (en) * | 2008-01-17 | 2008-09-10 | 北京航空航天大学 | Electronic police background intelligent management and automatic implementation system |
| CN102509090A (en) * | 2011-11-29 | 2012-06-20 | 冷明 | Vehicle feature recognition device based on public security video images in skynet engineering |
| CN202563521U (en) * | 2011-11-29 | 2012-11-28 | 冷明 | Vehicle characteristic identification device based on public security video image in skynet engineering |
| CN104488004A (en) * | 2012-05-23 | 2015-04-01 | 实耐宝公司 | Methods and systems for providing vehicle repair information |
| CN103049498A (en) * | 2012-12-07 | 2013-04-17 | 王志欣 | Method for recognizing vehicle type through vehicle identification number (VIN) |
| CN104216391A (en) * | 2013-05-31 | 2014-12-17 | 广州汽车集团股份有限公司 | Automobile decoder and automotive type recognition method thereof |
| CN105320708A (en) * | 2014-08-05 | 2016-02-10 | 北京大学 | Establishment method of vehicle model database and server |
| CN104572854A (en) * | 2014-12-17 | 2015-04-29 | 语联网（武汉）信息技术有限公司 | Decision-tree-based translator classification method |
| CN104537010A (en) * | 2014-12-17 | 2015-04-22 | 温州大学 | Component classifying method based on net establishing software of decision tree |
| CN105205468A (en) * | 2015-09-24 | 2015-12-30 | 上海车音网络科技有限公司 | Vehicle washing device, system and method capable of automatically identifying vehicle types |
| CN105335760A (en) * | 2015-11-16 | 2016-02-17 | 南京邮电大学 | Image number character recognition method |
| CN106021545A (en) * | 2016-05-27 | 2016-10-12 | 大连楼兰科技股份有限公司 | Method for remote diagnoses of cars and retrieval of spare parts |
| CN106056221A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | FP-Tree sequence pattern mining and fault code classification-based vehicle remote diagnosis and spare part retrieval method |
| CN106054858A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | Decision tree classification and fault code classification-based vehicle remote diagnosis and spare part retrieval method |
Non-Patent Citations (4)
| Title |
|---|
| K. YING等: "Decision tree-based machine learning algorithm for in-node vehicle classification", 《2015 IEEE GREEN ENERGY AND SYSTEMS CONFERENCE (IGESC)》 * |
| WAN HUI SONG等: "Vehicle Identification and Classification System Based on Decision Tree", 《APPLIED MECHANICS AND MATERIALS》 * |
| 张洪明等: "基于决策树的车型识别技术研究", 《数字技术与应用》 * |
| 风筝疯了: "决策树", 《CSDN: HTTPS://BLOG.CSDN.NET/KITE_CRAZY/ARTICLE/DETAILS/45250791》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110070092A (en) * | 2019-03-15 | 2019-07-30 | 平安科技（深圳）有限公司 | Model recognizing method, device, computer equipment and storage medium |
| CN110334586A (en) * | 2019-05-22 | 2019-10-15 | 深圳壹账通智能科技有限公司 | A kind of automobile recognition methods, device, computer system and readable storage medium storing program for executing |
| 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 |
| CN115408580A (en) * | 2022-08-31 | 2022-11-29 | 广东数鼎科技有限公司 | Vehicle source model identification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108573198A (en) | A kind of method and device identifying vehicle information according to Vehicle Identify Number | |
| CN103514201B (en) | Method and device for querying data in non-relational database | |
| CN106503148B (en) | A kind of table entity link method based on multiple knowledge base | |
| CN106096727A (en) | A kind of network model based on machine learning building method and device | |
| CN104346444B (en) | A kind of the best site selection method based on the anti-spatial key inquiry of road network | |
| CN102750286B (en) | A kind of Novel decision tree classifier method processing missing data | |
| CN103064970A (en) | Search method for optimizing translators | |
| CN104392010A (en) | Subgraph matching query method | |
| CN106600067A (en) | Method and device for optimizing multidimensional cube model | |
| CN101482876B (en) | Weight-based link multi-attribute entity recognition method | |
| CN106294661A (en) | A kind of extended search method and device | |
| CN110502509A (en) | A kind of traffic big data cleaning method and relevant apparatus based on Hadoop Yu Spark frame | |
| CN107145516A (en) | A kind of Text Clustering Method and system | |
| CN105005584A (en) | Multi-subspace Skyline query computation method | |
| CN112200471B (en) | Charging pile quantity configuration method and device and charging pile distribution system | |
| CN113467851B (en) | Dynamic vehicle computing task unloading method and device based on vehicle clustering | |
| CN113886587A (en) | Data classification method based on deep learning and map building method | |
| CN101894129A (en) | Video topic finding method based on online video-sharing website structure and video description text information | |
| CN106874506A (en) | community mining method and system based on statistical model | |
| CN110162716A (en) | A kind of influence power community search method and system based on community's retrieval | |
| CN110275868A (en) | A kind of multi-modal pretreated method of manufaturing data in intelligent plant | |
| CN108171538A (en) | User data processing method and system | |
| CN109657060B (en) | Safety production accident case pushing method and system | |
| CN113902220B (en) | Vehicle track prediction method based on self-adaptive density clustering algorithm | |
| CN107403386A (en) | Insure management method and system based on bank outlets |
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 | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20180925 Assignee: Beijing May 8th clapping Information Technology Co.,Ltd. Assignor: YOUXIN HULIAN (BEIJING) INFORMATION TECHNOLOGY Co.,Ltd. Contract record no.: X2020990000157 Denomination of invention: Method and device for identifying vehicle type information according to vehicle frame numbers License type: Common License Record date: 20200402 |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180925 |