CN109918984A - Insurance policy number identification method, device, electronic equipment and storage medium - Google Patents
Insurance policy number identification method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109918984A CN109918984A CN201811537708.5A CN201811537708A CN109918984A CN 109918984 A CN109918984 A CN 109918984A CN 201811537708 A CN201811537708 A CN 201811537708A CN 109918984 A CN109918984 A CN 109918984A
- Authority
- CN
- China
- Prior art keywords
- score value
- character
- insurance
- insurance policy
- identification
- 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 71
- 238000012549 training Methods 0.000 claims abstract description 68
- 239000000284 extract Substances 0.000 claims abstract description 11
- 238000004590 computer program Methods 0.000 claims description 19
- 238000000605 extraction Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 abstract description 16
- 238000013527 convolutional neural network Methods 0.000 description 18
- 230000006870 function Effects 0.000 description 14
- 238000012360 testing method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000012790 confirmation Methods 0.000 description 4
- 230000008521 reorganization Effects 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 208000027418 Wounds and injury Diseases 0.000 description 2
- 239000012141 concentrate Substances 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 208000014674 injury Diseases 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
A kind of insurance policy number identification method, comprising: the insurance type of the first insurance policy picture of identification；Extract the first object line character region where the insurance single numbers in the first insurance policy picture；The first identification model that training generates in advance is called to identify first object line character region, and output character recognition result and the corresponding score value of each character identification result；Whether the number for counting the target score value for being greater than default score value threshold value in all score value is 1；When determining number is 1, using the corresponding character identification result of the target score value as the recognition result of the character in first object line character region.The present invention also provides a kind of insurance policy NID number identifier, electronic equipment and storage mediums.The present invention hardly needs artificial participation in the entire identification process of insurance policy, for the different insurers, insurance single numbers can be rapidly obtained from the first insurance policy picture, greatly reduces workload, improves working efficiency.
Description
Technical field
The present invention relates to mode identification technologies, and in particular to a kind of insurance policy number identification method, device, electronics are set
Standby and storage medium.
Background technique
Currently, there are many insurance policies to be stored in insurance company in the form of picture, every portion insurance policy or the first guarantor
Dangerous free hand drawing piece has unique insurance single numbers corresponding.For the first insurance policy picture, insuring single numbers is key message,
Staff generally requires to be operated according to retrieval or the inquiry etc. that insurance single numbers carry out insurance information.If will from multiple first
When obtaining the pictorial informations such as insurance single numbers in insurance policy picture, obtained a sheet by a sheet generally by manually-operated mode, when
The quantity of first insurance policy picture is more, and staff can not be quickly obtained the insurance odd numbers of each the first insurance policy picture
Code, causes workload very big, reduces working efficiency.
Although in the prior art, obtaining the first identification model based on machine learning training to identify insurance single numbers,
But the first identification model of machine learning is also unable to ensure absolutely accuracy rate, and the characteristics of different insurer's hand-written character
Difference, the model identification obtained currently based on machine learning training do not consider that the different insurers' is hand-written when insuring single numbers
Feature.
Summary of the invention
In view of the foregoing, it is necessary to propose that a kind of insurance policy number identification method, device, electronic equipment and storage are situated between
Matter hardly needs artificial participation in the entire identification process of insurance policy, can be rapidly from for the different insurers
Insurance single numbers are obtained in one insurance policy picture, greatly reduce workload, improve working efficiency.
The first aspect of the present invention provides a kind of insurance policy number identification method, which comprises
When receiving the first insurance policy picture, the corresponding insurance type of the first insurance policy picture is identified；
Based on preset insurance type and insure positional relationship extraction institute of the single numbers in the first insurance policy picture
State the first object line character region where the insurance single numbers in the first insurance policy picture；
It calls the first identification model that training generates in advance to identify first object line character region, and exports
The corresponding score value of each of character identification result and the character identification result character identification result；
Whether the number for counting the target score value for being greater than default score value threshold value in all score value is 1；
When determining the number in all score value greater than the target score value of the default score value threshold value is 1, by institute
State recognition result of the corresponding character identification result of target score value as the character in first object line character region.
Preferably, when the number for determining the target score value in all score value greater than the default score value threshold value is greater than
When 1, the method also includes:
The corresponding character identification result of target score value for being greater than the default score value threshold value in the score value is made
For the candidate characters collection of the character in first object line character region；
Show the list of candidate characters recognition result and each candidate characters recognition result pair that the candidate characters are concentrated
The score value answered；
When detect the candidate characters recognition result in the candidate characters recognition result list it is chosen when, will be selected
Recognition result of the candidate characters recognition result as the character in first object line character region.
Preferably, after identifying the character in first object line character region, the method also includes:
It is mentioned based on preset insurance type with positional relationship of insurer's name in the first insurance policy picture
Take the second target line character zone where insurer's name in the first insurance policy picture；
The the two the first identification models for calling training in advance to generate identify to the second target line character zone
To insurer's name in the first insurance policy picture.
Preferably, after identifying insurer's name in the second target line character zone, the method also includes:
The insurance single numbers that will identify that, insurer's name, candidate characters collection, chosen candidate characters recognition result and
The first insurance policy picture is associated storage.
Preferably, the method also includes:
When the second insurance policy picture for receiving the same insurer again, and if call that the preparatory training generates the
One identification model obtains multiple candidates after carrying out character recognition to the first object line character region of the second insurance policy picture
When character identification result, according to the incidence relation of the insurer, candidate characters collection and chosen candidate characters recognition result,
Using selected character identification result as the correspondence character in the first object character zone of the second insurance policy picture
Recognition result.
Preferably, the method also includes:
One display interface is provided；
The candidate characters identification knot that the candidate characters are concentrated is shown according to default display mode on the display interface
Fruit list and each corresponding score value of candidate characters recognition result.
Preferably, the default display mode includes:
Branch shows after the score value is ranked up according to descending sequence；
Show the first insurance policy picture；
Corresponding every row score value shows a tick boxes, to receive staff choose operation after, will be hooked
Recognition result of the candidate characters recognition result of choosing as the character in the target line character zone.
The second aspect of the present invention provides a kind of insurance policy NID number identifier, and described device includes:
Receiving module when for receiving the first insurance policy picture, identifies the corresponding insurance of the first insurance policy picture
Type；
Extraction module, for the position based on preset insurance type and insurance single numbers in the first insurance policy picture
The relationship of setting extracts the first object line character region where the insurance single numbers in the first insurance policy picture；
Identification module, for call in advance training generate the first identification model to first object line character region into
Row identification, and the corresponding score of each of output character recognition result and the character identification result character identification result
Value；
Judgment module, for count the target score value for being greater than default score value threshold value in all score value number whether
It is 1；
Output module, for determining the mesh in all score value greater than the default score value threshold value when the judgment module
When the number for marking score value is 1, using the corresponding character identification result of the target score value as the first object line character
The recognition result of character in region.
The third aspect of the present invention provides a kind of electronic equipment, and the electronic equipment includes processor and memory, described
Processor is for realizing the insurance policy number identification method when executing the computer program stored in the memory.
The fourth aspect of the present invention provides a kind of computer readable storage medium, deposits on the computer readable storage medium
Computer program is contained, the computer program realizes the insurance policy number identification method when being executed by processor.
Insurance policy number identification method, device, electronic equipment and storage medium of the present invention identify that first protects first
The insurance type of dangerous free hand drawing piece by insurance type and insures positional relationship of the single numbers in the first insurance policy picture, can be with
Insurance single numbers corresponding first object line character region is extracted, the first identification model that training generates in advance is then recalled
It identifies the character in the first object line character region and provides character identification result and corresponding score value, determining
Have be greater than in score value default score value threshold value target score value only one when, by the corresponding character knowledge of the target score value
Recognition result number of the other result as the character in first object line character region, whole operation process hardly need manually
It participates in, insurance single numbers can be rapidly obtained from a large amount of first insurance policy picture, greatly reduces workload, improves work
Efficiency.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is the flow chart for the insurance policy number identification method that the embodiment of the present invention one provides.
Fig. 2 is the functional block diagram of insurance policy NID number identifier provided by Embodiment 2 of the present invention.
Fig. 3 is the schematic diagram for the electronic equipment that the embodiment of the present invention three provides.
The present invention that the following detailed description will be further explained with reference to the above drawings.
Specific embodiment
To better understand the objects, features and advantages of the present invention, with reference to the accompanying drawing and specific real
Applying example, the present invention will be described in detail.It should be noted that in the absence of conflict, the embodiment of the present invention and embodiment
In feature can be combined with each other.
In the following description, numerous specific details are set forth in order to facilitate a full understanding of the present invention, described embodiment is only
It is only a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill
Personnel's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Unless otherwise defined, all technical and scientific terms used herein and belong to technical field of the invention
The normally understood meaning of technical staff is identical.Term as used herein in the specification of the present invention is intended merely to description tool
The purpose of the embodiment of body, it is not intended that in the limitation present invention.
The insurance policy number identification method of the embodiment of the present invention is applied in one or more electronic equipment.The insurance
Single numbers recognition methods also can be applied to by electronic equipment and the server being attached by network and the electronic equipment
In the hardware environment constituted.Network includes but is not limited to: wide area network, Metropolitan Area Network (MAN) or local area network.The insurance of the embodiment of the present invention
Single numbers recognition methods can be executed by server, can also be executed by electronic equipment；It can also be by server and electricity
Sub- equipment executes jointly.
For needing to carry out the electronic equipment of insurance policy number identification method, it can directly collect cost hair on an electronic device
Insurance policy Number Reorganization function provided by bright method, or installation is for realizing the client of method of the invention.For another example,
Method provided by the present invention can also be in the form of Software Development Kit (Software Development Kit, SDK)
Operate in the equipment such as server, in the form of SDK provide insurance policy Number Reorganization function interface, electronic equipment or other set
Method of the present invention can be realized in the standby interface by providing.
Embodiment one
Fig. 1 is the flow chart for the insurance policy number identification method that the embodiment of the present invention one provides.According to different requirements, should
Execution sequence in flow chart can change, and certain steps can be omitted.
S1 when receiving the first insurance policy picture, identifies the corresponding insurance type of the first insurance policy picture.
In the present embodiment, there are many types of insurance policy or the first insurance policy picture, such as has vehicle insurance insurance policy, life insurance to protect
Dangerous list and accident/injury insurance list etc., each insurance policy are a kind of insurance type.For different types of insurance policy, insurance
The location of single numbers are not identical, such as some insurance single numbers are located at insurance policy upper right corner position on the upper side, some guarantors
Dangerous single numbers are located at insurance policy upper right corner position to the left.After receiving the first insurance policy picture, first guarantor is identified first
Insurance type belonging to dangerous free hand drawing piece, specific identification process are as follows: by size, color and contents and distribution to insurance policy etc. into
The comprehensive identification of row, to judge insurance type belonging to the first insurance policy picture, alternatively, it is also possible to pass through other method identifications
Insurance type belonging to the first insurance policy picture, such as judged belonging to it by identifying the content information of the insurance picture
Insure type etc..
S2 based on preset insurance type and insures positional relationship extraction of the single numbers in the first insurance policy picture
First object line character region where insurance single numbers in the first insurance policy picture.
The present embodiment in advance by different types of insurance policy and insurance single numbers the location of in insurance policy picture into
Row associated storage.After identifying the insurance type belonging to it, based on the insurance type and insurance single numbers in the first insurance policy
Positional relationship in picture extracts insurance single numbers corresponding first object line character in the first insurance policy picture
Region, after being extracted insurance single numbers corresponding first object line character region, it is only necessary to further identify the first object
Insurance single numbers can be obtained in character in line character region.
S3 calls the first identification model that training generates in advance to identify first object line character region, and
The corresponding score value of each of output character recognition result and the character identification result character identification result.
In the present embodiment, training generates the first identification model in advance, and it is relevant that the first identification model can be image procossing
One of a variety of models.
It calls the first identification model to carry out character recognition to first object line character region, which is obtained with identification
Each of line character region character, in general, insurance single numbers are number, can after all character recognition come out
To obtain insurance single numbers.When identifying some character in first object line character region, training in advance is called to generate
First identification model identified, and exports the character identification result and each of character identification result word of the character
Accord with the corresponding score value of recognition result.Illustratively, the insurance policy in the first object line character region of the first insurance policy picture
Number is " 25481 ", then when character " 4 " in single numbers are insured in identification, calls the first identification model that training generates in advance
Identify available 10 character identification results: 0,1,2,3,4,5,6,7,8,9.This 10 character identification results have pair
The score value answered: 0,0,0,0,99%, 0,0,0,0,95%.The score value indicates identified character and insurance single numbers
Character " 4 " similarity, similarity is bigger, it is believed that the accuracy rate of the character identified is higher, and similarity is smaller, it is believed that institute
The accuracy rate of the character identified is lower.
Preferably, the training process of first identification model includes:
1) the insurance policy samples pictures for obtaining preset quantity, using comprising insure single numbers insurance policy samples pictures as the
One pictures, and the insurance policy samples pictures of insurance single numbers will not included as second picture collection；
2) the insurance policy sample graph for extracting the first preset ratio respectively is concentrated from first pictures and second picture
Piece as samples pictures to be trained, and using the first pictures and second picture concentrate remaining insurance policy samples pictures as to
The samples pictures of verifying；
3) model training is carried out using samples pictures respectively to be trained, to generate the convolutional neural networks model, and benefit
Convolutional neural networks model generated is verified with each samples pictures to be verified；
If 4) be verified rate more than or equal to preset threshold, training is completed, and otherwise increases the insurance policy samples pictures
Quantity, with re-start training and verifying.
Illustratively, it is assumed that obtain 100,000 insurance single numbers samples pictures, wherein insurance single numbers samples pictures are only wrapped
Containing a line number, which is insurance single numbers, and font is black, and background is white, and can be by each insurance single numbers sample
The name nominating of this picture is contained insurance single numbers.The insurance single numbers samples pictures of the second preset ratio are extracted as instruction
Practice collection, and using remaining insurance single numbers samples pictures in the insurance single numbers samples pictures of preset quantity as test set, instruction
The quantity for practicing the insurance single numbers samples pictures concentrated is greater than the quantity of the insurance single numbers samples pictures in test set, such as will
80% insurance single numbers samples pictures in single numbers samples pictures are insured as training set, by remaining 20% insurance policy
Number samples pictures are as test set.
In first time training convolutional neural networks model, the parameter of the convolutional neural networks model is using the parameter defaulted
It is trained, in the continuous adjusting parameter of training process, after training generates the convolutional neural networks model, using each to be verified
Samples pictures verify convolutional neural networks model generated, and rate is more than or equal to preset threshold, example if the verification passes
If percent of pass is more than or equal to 98%, then training terminates, and the convolutional neural networks model obtained with the training is to carry out identification first
The model of target line character zone；Rate is less than preset threshold if the verification passes, is, for example, less than 98%, then increases insurance policy sample
The quantity of picture, and above-mentioned step is re-executed, until being verified rate more than or equal to preset threshold.
In test, the convolutional neural networks model obtained using training is to the insurance single numbers samples pictures in test set
Insurance policy Number Reorganization is carried out, and the title used of recognition result and the insurance single numbers samples pictures is compared (should
Insurance single numbers samples pictures are named using the insurance single numbers), to assess trained convolutional neural networks model
Recognition effect.
S4: whether the number for counting the target score value for being greater than default score value threshold value in all score value is 1.
In the present embodiment, first object line character region is input in the first identification model that training generates in advance
Character recognition is carried out, is substantially using the first identification model that training generates in advance to the word in first object line character region
Symbol carries out score value calculating.For example, the character 4 in first object line character region, the first identification generated by training in advance
Model is identified to obtain the score value that result is 1 to be 0, and obtaining the score value that result is 2 is 0, obtains the score value that result is 4
It is 99, obtaining the score value that result is 6 is 0, and obtaining the score value that result is 9 is 95.
In the present embodiment, score value threshold value can be preset, score value threshold value indicates the confidence level that recognition result has.
For example, can preset score value threshold value is 85, it is with higher that score value is greater than the result that 85 show that corresponding identification obtains
Confidence level, the result that score value shows that corresponding identification is obtained lower than 85 have lower confidence level.
When determining the number in all score value greater than the target score value of the default score value threshold value is 1, execute
S5；Otherwise, it when determining that the number in all score value greater than the target score value of the default score value threshold value is greater than 1, holds
Row S6.
S5, using the corresponding character identification result of the target score value as the word in first object line character region
The recognition result of symbol.
In the present embodiment, first object line character region is carried out by the first identification model that training generates in advance
Character recognition, only 1 target score value is greater than default score value threshold value in obtained all score value, then shows to identify
As a result there is uniqueness, thus can be using the unique consequence that this is identified as the knowledge of the character in first object line character region
Other result.
It should be understood that carrying out word to first object line character region by the first identification model that training generates
Symbol identification is greater than the target score value of default score value threshold value or unique or be multiple in obtained score value, it is impossible to
It is zero.
S6 will be greater than the corresponding character identification result of target score value of the default score value threshold value in the score value
The candidate characters collection of character as first object line character region.
In the present embodiment, first object line character region is carried out by the first identification model that training generates in advance
Character recognition has in obtained all score value multiple (for example, 2) target score values to be greater than default score value threshold value, then table
The bright result identified does not have uniqueness, multiple recognition result probability having the same as correct recognition result, because
And the multiple candidate characters recognition results that can identify this are as the candidate word of the character in first object line character region
Symbol collection.
Illustratively, the first identification model that training generates in advance is called to carry out first object line character region 4
Character recognition, and export the default corresponding candidate characters recognition result of score value threshold value of score value and the knowledge of each candidate characters
The corresponding score value of other result is respectively as follows: character 4, score value 99；Character 9, score value 95.The multiple score value is all larger than pre-
If score value threshold value 85, thus can candidate by candidate characters recognition result 4 and 9 as first object line character region 4
Character set.
S7: the list of candidate characters recognition result and each candidate characters recognition result that the candidate characters are concentrated are shown
Corresponding score value.
In the present embodiment, a display interface can be provided, institute is shown according to default display mode on the display interface
State the list of candidate characters recognition result and each corresponding score value of candidate characters recognition result of candidate characters concentration.
Preferably, the default display mode includes:
Branch shows after the score value is ranked up according to descending sequence；
Show the first insurance policy picture；
Corresponding every row score value shows a tick boxes, to receive staff choose operation after, will be hooked
Recognition result of the candidate characters recognition result of choosing as the character in the target line character zone.
In the present embodiment, the score value is ranked up according to descending sequence, top score value and correspondence are most
The candidate characters recognition result of high score value is shown in the first row；The candidate characters of secondary high score value and corresponding high score value are known
Not as the result is shown in the second row；And so on；The candidate characters recognition result of minimum score value and corresponding minimum score value is shown
In last line, candidate characters recognition result so can quickly and be intuitively checked convenient for staff.In addition, candidate in display
Described received is also shown while character identification result list and each candidate characters recognition result corresponding score value
One insurance policy picture.It shows the first insurance policy picture, compares the first insurance policy picture convenient for staff and determine institute
Stating which of candidate characters collection candidate characters is final character identification result.
S8: when detect the candidate characters recognition result in the candidate characters recognition result list it is chosen when, will be by
Recognition result of the selected candidate characters recognition result as the character in first object line character region.
Staff can select character identification result as described first by touching or clicking the tick boxes
The recognition result of the character of target line character zone.
Preferably, after identifying the character in first object line character region, the method also includes:
It is mentioned based on preset insurance type with positional relationship of insurer's name in the first insurance policy picture
Take the second target line character zone where insurer's name in the first insurance policy picture；
The second identification model that training generates in advance is called to be identified to obtain institute to the second target line character zone
State insurer's name in the first insurance policy picture.
For different types of insurance policy, the location of insurer's name is not also identical.It can be in advance by inhomogeneity
The insurance policy and insurer's name of type are associated storage the location of in insurance policy picture.Identifying the guarantor belonging to it
After dangerous type, the positional relationship based on the insurance type and insurer's name in the first insurance policy picture extracts the insurer
Name corresponding second target line character zone in the first insurance policy picture is being extracted insurer's name corresponding
After two target line character zones, it is only necessary to the second identification model that training generates in advance be called further to identify second target
Insurer's name can be obtained in character in line character region.
About second identification model training process with first identification model training process, herein not
It elaborates again.
Preferably, after identifying insurer's name in the second target line character zone, the method also includes:
The insurance single numbers that will identify that, insurer's name, candidate characters collection, chosen candidate characters recognition result and
The first insurance policy picture is associated storage.
After identification obtains insurance single numbers and insurer's name, by the insurance single numbers, insurer's name candidate characters
Collection, chosen candidate characters recognition result and the first insurance policy picture are associated storage, so that staff is passing through
The insurance single numbers, which are inquired when perhaps being retrieved, by the insurance policy number inquiry or can retrieve associated with it
One insurance policy picture.In addition, subsequent also can be used insurer's name candidate characters collection, chosen candidate characters recognition result
Incidence relation make further application.
Preferably, the method can also include:
When the second insurance policy picture for receiving the same insurer again, and if call that the preparatory training generates the
One identification model obtains multiple candidates after carrying out character recognition to the first object line character region of the second insurance policy picture
When character identification result, according to the incidence relation of the insurer, candidate characters collection and chosen candidate characters recognition result,
Using selected character identification result as the correspondence character in the first object character zone of the second insurance policy picture
Recognition result.
In the present embodiment, if after subsequent the second insurance policy picture for receiving the same insurer again, even if identifying
There are multiple score value for being greater than default score value threshold value, that is, when having candidate characters collection, without showing what candidate characters were concentrated again
Multiple character identification result lists and corresponding score value, but directly using previously chosen candidate characters recognition result as
The recognition result of the character of the target character region simultaneously directly exports.In this way, it can be further reduced the operation of staff,
The time of staff is saved, and recognition result will not be reduced.
Certainly, the embodiment of the present invention can also first obtain described first after most initially receiving the first insurance policy picture
The insurer in insurance policy picture, it is subsequent when having identified multiple candidate characters, directly export previously chosen candidate word
Accord with recognition result of the recognition result as the character of the target character region.
In conclusion insurance policy number identification method provided in an embodiment of the present invention, identifies the first insurance policy picture first
Insurance type, by insurance type and insure positional relationship of the single numbers in the first insurance policy picture, guarantor can be extracted
Then dangerous single numbers corresponding first object line character region recalls the first identification model that training generates in advance to identify
Character in the first object line character region simultaneously provides character identification result and corresponding score value, is determining all score value
In be greater than default score value threshold value target score value only one when, by the corresponding character identification result work of the target score value
For the recognition result number of the character in first object line character region, whole operation process hardly needs artificial participation, energy
It is enough that insurance single numbers are rapidly obtained from a large amount of first insurance policy picture, greatly reduce workload, improves working efficiency.
Secondly, determining candidate characters collection by score value, and candidate characters collection list is shown, by staff according to simultaneously
First insurance policy picture of display carries out manual confirmation, cannot defect very to make up the first identification model accuracy of identification.
And after first passing through the identification of the first identification model, the amount of candidate characters collection to be confirmed is considerably less, at this time again will not by manual confirmation
It wastes time, and accuracy rate is higher.
The above is only a specific embodiment of the invention, but scope of protection of the present invention is not limited thereto, for
For those skilled in the art, without departing from the concept of the premise of the invention, improvement, but these can also be made
It all belongs to the scope of protection of the present invention.
Below with reference to the 2nd to 3 figure, respectively to the functional module of the electronic equipment of the above-mentioned insurance policy number identification method of realization
And hardware configuration is introduced.
Embodiment two
Fig. 2 is the functional block diagram in insurance policy NID number identifier preferred embodiment of the present invention.
In some embodiments, the insurance policy NID number identifier 20 is run in electronic equipment.The insurance odd numbers
Code recognition device 20 may include multiple functional modules as composed by program code segments.The insurance policy NID number identifier 20
In the program code of each program segment can store in memory, and as performed by least one processor, (in detail with execution
See Fig. 1 and its associated description) insurance policy number identification method.
In the present embodiment, function of the insurance policy NID number identifier 20 according to performed by it can be divided into more
A functional module.The functional module may include: receiving module 201, extraction module 202, identification module 203, judgment module
204, output module 205, display module 206, choose module 207 and relating module 208.The so-called module of the present invention refers to one kind
Performed by least one processor and the series of computation machine program segment of fixed function can be completed, be stored in
In reservoir.In some embodiments, it will be described in detail in subsequent embodiment about the function of each module.
Receiving module 201 when for receiving the first insurance policy picture, identifies the corresponding guarantor of the first insurance policy picture
Dangerous type.
In the present embodiment, there are many types of insurance policy or the first insurance policy picture, such as has vehicle insurance insurance policy, life insurance to protect
Dangerous list and accident/injury insurance list etc., each insurance policy are a kind of insurance type.For different types of insurance policy, insurance
The location of single numbers are not identical, such as some insurance single numbers are located at insurance policy upper right corner position on the upper side, some guarantors
Dangerous single numbers are located at insurance policy upper right corner position to the left.After receiving the first insurance policy picture, first guarantor is identified first
Insurance type belonging to dangerous free hand drawing piece, specific identification process are as follows: by size, color and contents and distribution to insurance policy etc. into
The comprehensive identification of row, to judge insurance type belonging to the first insurance policy picture, alternatively, it is also possible to pass through other method identifications
Insurance type belonging to the first insurance policy picture, such as judged belonging to it by identifying the content information of the insurance picture
Insure type etc..
Extraction module 202, for being based on preset insurance type and insurance single numbers in the first insurance policy picture
Positional relationship extract the first object line character region where the insurance single numbers in the first insurance policy picture.
The present embodiment in advance by different types of insurance policy and insurance single numbers the location of in insurance policy picture into
Row associated storage.After identifying the insurance type belonging to it, based on the insurance type and insurance single numbers in the first insurance policy
Positional relationship in picture extracts insurance single numbers corresponding first object line character in the first insurance policy picture
Region, after being extracted insurance single numbers corresponding first object line character region, it is only necessary to further identify the first object
Insurance single numbers can be obtained in character in line character region.
Identification module 203, for calling the first identification model of training generation in advance to first object line character area
Domain identified, and each of output character recognition result and the character identification result character identification result are corresponding
Score value.
In the present embodiment, training generates the first identification model in advance, and it is relevant that the first identification model can be image procossing
One of a variety of models.
It calls the first identification model to carry out character recognition to first object line character region, which is obtained with identification
Each of line character region character, in general, insurance single numbers are number, can after all character recognition come out
To obtain insurance single numbers.When identifying some character in first object line character region, training in advance is called to generate
First identification model identified, and exports the character identification result and each of character identification result word of the character
Accord with the corresponding score value of recognition result.Illustratively, the insurance policy in the first object line character region of the first insurance policy picture
Number is " 25481 ", then when character " 4 " in single numbers are insured in identification, calls the first identification model that training generates in advance
Identify available 10 character identification results: 0,1,2,3,4,5,6,7,8,9.This 10 character identification results have pair
The score value answered: 0,0,0,0,99%, 0,0,0,0,95%.The score value indicates identified character and insurance single numbers
Character " 4 " similarity, similarity is bigger, it is believed that the accuracy rate of the character identified is higher, and similarity is smaller, it is believed that institute
The accuracy rate of the character identified is lower.
Preferably, the training process of first identification model includes:
1) the insurance policy samples pictures for obtaining preset quantity, using comprising insure single numbers insurance policy samples pictures as the
One pictures, and the insurance policy samples pictures of insurance single numbers will not included as second picture collection；
2) the insurance policy sample graph for extracting the first preset ratio respectively is concentrated from first pictures and second picture
Piece as samples pictures to be trained, and using the first pictures and second picture concentrate remaining insurance policy samples pictures as to
The samples pictures of verifying；
3) model training is carried out using samples pictures respectively to be trained, to generate the convolutional neural networks model, and benefit
Convolutional neural networks model generated is verified with each samples pictures to be verified；
If 4) be verified rate more than or equal to preset threshold, training is completed, and otherwise increases the insurance policy samples pictures
Quantity, with re-start training and verifying.
Illustratively, it is assumed that obtain 100,000 insurance single numbers samples pictures, wherein insurance single numbers samples pictures are only wrapped
Containing a line number, which is insurance single numbers, and font is black, and background is white, and can be by each insurance single numbers sample
The name nominating of this picture is contained insurance single numbers.The insurance single numbers samples pictures of the second preset ratio are extracted as instruction
Practice collection, and using remaining insurance single numbers samples pictures in the insurance single numbers samples pictures of preset quantity as test set, instruction
The quantity for practicing the insurance single numbers samples pictures concentrated is greater than the quantity of the insurance single numbers samples pictures in test set, such as will
80% insurance single numbers samples pictures in single numbers samples pictures are insured as training set, by remaining 20% insurance policy
Number samples pictures are as test set.
In first time training convolutional neural networks model, the parameter of the convolutional neural networks model is using the parameter defaulted
It is trained, in the continuous adjusting parameter of training process, after training generates the convolutional neural networks model, using each to be verified
Samples pictures verify convolutional neural networks model generated, and rate is more than or equal to preset threshold, example if the verification passes
If percent of pass is more than or equal to 98%, then training terminates, and the convolutional neural networks model obtained with the training is to carry out identification first
The model of target line character zone；Rate is less than preset threshold if the verification passes, is, for example, less than 98%, then increases insurance policy sample
The quantity of picture, and above-mentioned step is re-executed, until being verified rate more than or equal to preset threshold.
In test, the convolutional neural networks model obtained using training is to the insurance single numbers samples pictures in test set
Insurance policy Number Reorganization is carried out, and the title used of recognition result and the insurance single numbers samples pictures is compared (should
Insurance single numbers samples pictures are named using the insurance single numbers), to assess trained convolutional neural networks model
Recognition effect.
Judgment module 204, for counting the number for being greater than the target score value of default score value threshold value in all score value
It whether is 1.
In the present embodiment, first object line character region is input in the first identification model that training generates in advance
Character recognition is carried out, is substantially using the first identification model that training generates in advance to the word in first object line character region
Symbol carries out score value calculating.For example, the character 4 in first object line character region, the first identification generated by training in advance
Model is identified to obtain the score value that result is 1 to be 0, and obtaining the score value that result is 2 is 0, obtains the score value that result is 4
It is 99, obtaining the score value that result is 6 is 0, and obtaining the score value that result is 9 is 95.
In the present embodiment, score value threshold value can be preset, score value threshold value indicates the confidence level that recognition result has.
For example, can preset score value threshold value is 85, it is with higher that score value is greater than the result that 85 show that corresponding identification obtains
Confidence level, the result that score value shows that corresponding identification is obtained lower than 85 have lower confidence level.
Output module 205, for being greater than the default score value threshold in all score value when the judgment module 204 determines
When the number of the target score value of value is 1, using the corresponding character identification result of the target score value as the first object
The recognition result of character in line character region.
In the present embodiment, first object line character region is carried out by the first identification model that training generates in advance
Character recognition, only 1 target score value is greater than default score value threshold value in obtained all score value, then shows to identify
As a result there is uniqueness, thus can be using the unique consequence that this is identified as the knowledge of the character in first object line character region
Other result.
It should be understood that carrying out word to first object line character region by the first identification model that training generates
Symbol identification is greater than the target score value of default score value threshold value or unique or be multiple in obtained score value, it is impossible to
It is zero.
The output module 205 is also used to when the target for being greater than the default score value threshold value in determining all score value
When the number of score value is greater than 1, the corresponding word of target score value of the default score value threshold value will be greater than in the score value
Accord with candidate characters collection of the recognition result as the character in first object line character region.
In the present embodiment, first object line character region is carried out by the first identification model that training generates in advance
Character recognition has in obtained all score value multiple (for example, 2) target score values to be greater than default score value threshold value, then table
The bright result identified does not have uniqueness, multiple recognition result probability having the same as correct recognition result, because
And the multiple candidate characters recognition results that can identify this are as the candidate word of the character in first object line character region
Symbol collection.
Illustratively, the first identification model that training generates in advance is called to carry out first object line character region 4
Character recognition, and export the default corresponding candidate characters recognition result of score value threshold value of score value and the knowledge of each candidate characters
The corresponding score value of other result is respectively as follows: character 4, score value 99；Character 9, score value 95.The multiple score value is all larger than pre-
If score value threshold value 85, thus can candidate by candidate characters recognition result 4 and 9 as first object line character region 4
Character set.
Display module 206, for showing the list of candidate characters recognition result and each time of the candidate characters concentration
Select the corresponding score value of character identification result.
In the present embodiment, a display interface can be provided, institute is shown according to default display mode on the display interface
State the list of candidate characters recognition result and each corresponding score value of candidate characters recognition result of candidate characters concentration.
Preferably, the default display mode includes:
Branch shows after the score value is ranked up according to descending sequence；
Show the first insurance policy picture；
Corresponding every row score value shows a tick boxes, to receive staff choose operation after, will be hooked
Recognition result of the candidate characters recognition result of choosing as the character in the target line character zone.
In the present embodiment, the score value is ranked up according to descending sequence, top score value and correspondence are most
The candidate characters recognition result of high score value is shown in the first row；The candidate characters of secondary high score value and corresponding high score value are known
Not as the result is shown in the second row；And so on；The candidate characters recognition result of minimum score value and corresponding minimum score value is shown
In last line, candidate characters recognition result so can quickly and be intuitively checked convenient for staff.In addition, candidate in display
Described received is also shown while character identification result list and each candidate characters recognition result corresponding score value
One insurance policy picture.It shows the first insurance policy picture, compares the first insurance policy picture convenient for staff and determine institute
Stating which of candidate characters collection candidate characters is final character identification result.
Module 207 is chosen, for working as the candidate characters recognition result detected in the candidate characters recognition result list
When chosen, using chosen candidate characters recognition result as the identification knot of the character in first object line character region
Fruit.
Staff can select character identification result as described first by touching or clicking the tick boxes
The recognition result of the character of target line character zone.
The extraction module 202 is also used to protect based on preset insurance type and insurer's name described first
Positional relationship in dangerous free hand drawing piece extracts the second target line character where insurer's name in the first insurance policy picture
Region.
The identification module 203 is also used to call the second identification model that training generates in advance to second target line
Character zone is identified to obtain insurer's name in the first insurance policy picture.
For different types of insurance policy, the location of insurer's name is not also identical.It can be in advance by inhomogeneity
The insurance policy and insurer's name of type are associated storage the location of in insurance policy picture.Identifying the guarantor belonging to it
After dangerous type, the positional relationship based on the insurance type and insurer's name in the first insurance policy picture extracts the insurer
Name corresponding second target line character zone in the first insurance policy picture is being extracted insurer's name corresponding
After two target line character zones, it is only necessary to the second identification model that training generates in advance be called further to identify second target
Insurer's name can be obtained in character in line character region.
About second identification model training process with first identification model training process, herein not
It elaborates again.
Relating module 208, insurance single numbers, insurer's name, candidate characters collection, chosen time for will identify that
Character identification result and the first insurance policy picture is selected to be associated storage.
After identification obtains insurance single numbers and insurer's name, by the insurance single numbers, insurer's name candidate characters
Collection, chosen candidate characters recognition result and the first insurance policy picture are associated storage, so that staff is passing through
The insurance single numbers, which are inquired when perhaps being retrieved, by the insurance policy number inquiry or can retrieve associated with it
One insurance policy picture.In addition, subsequent also can be used insurer's name candidate characters collection, chosen candidate characters recognition result
Incidence relation make further application.
Preferably, the insurance policy NID number identifier 20 can also include:
When the second insurance policy picture for receiving the same insurer again, and if call that the preparatory training generates the
One identification model obtains multiple candidates after carrying out character recognition to the first object line character region of the second insurance policy picture
When character identification result, according to the incidence relation of the insurer, candidate characters collection and chosen candidate characters recognition result,
Using selected character identification result as the correspondence character in the first object character zone of the second insurance policy picture
Recognition result.
In the present embodiment, if after subsequent the second insurance policy picture for receiving the same insurer again, even if identifying
There are multiple score value for being greater than default score value threshold value, that is, when having candidate characters collection, without showing what candidate characters were concentrated again
Multiple character identification result lists and corresponding score value, but directly using previously chosen candidate characters recognition result as
The recognition result of the character of the target character region simultaneously directly exports.In this way, it can be further reduced the operation of staff,
The time of staff is saved, and recognition result will not be reduced.
Certainly, the embodiment of the present invention can also first obtain described first after most initially receiving the first insurance policy picture
The insurer in insurance policy picture, it is subsequent when having identified multiple candidate characters, directly export previously chosen candidate word
Accord with recognition result of the recognition result as the character of the target character region.
In conclusion insurance policy NID number identifier provided in an embodiment of the present invention, identifies the first insurance policy picture first
Insurance type, by insurance type and insure positional relationship of the single numbers in the first insurance policy picture, guarantor can be extracted
Then dangerous single numbers corresponding first object line character region recalls the first identification model that training generates in advance to identify
Character in the first object line character region simultaneously provides character identification result and corresponding score value, is determining all score value
In be greater than default score value threshold value target score value only one when, by the corresponding character identification result work of the target score value
For the recognition result number of the character in first object line character region, whole operation process hardly needs artificial participation, energy
It is enough that insurance single numbers are rapidly obtained from a large amount of first insurance policy picture, greatly reduce workload, improves working efficiency.
Secondly, determining candidate characters collection by score value, and candidate characters collection list is shown, by staff according to simultaneously
First insurance policy picture of display carries out manual confirmation, cannot defect very to make up the first identification model accuracy of identification.
And after first passing through the identification of the first identification model, the amount of candidate characters collection to be confirmed is considerably less, at this time again will not by manual confirmation
It wastes time, and accuracy rate is higher.
The above-mentioned integrated unit realized in the form of software function module, can store and computer-readable deposit at one
In storage media.Above-mentioned software function module is stored in a storage medium, including some instructions are used so that a computer
It is each that equipment (can be personal computer, double screen equipment or the network equipment etc.) or processor (processor) execute the present invention
The part of a embodiment the method.
Embodiment three
Fig. 3 is the schematic diagram for the electronic equipment that the embodiment of the present invention three provides.
The electronic equipment 3 includes: memory 31, at least one processor 32, is stored in the memory 31 and can
The computer program 33 and at least one communication bus 34 run at least one described processor 32.
At least one described processor 32 realizes the step in above method embodiment when executing the computer program 33.
Illustratively, the computer program 33 can be divided into one or more module/units, it is one or
Multiple module/units are stored in the memory 31, and are executed by least one described processor 32, to complete the present invention
Step in above method embodiment.One or more of module/units, which can be, can complete a series of of specific function
Computer program instructions section, the instruction segment is for describing implementation procedure of the computer program 33 in the electronic equipment 3.
The electronic equipment 3 can be the calculating such as desktop PC, notebook, palm PC and cloud server and set
It is standby.It will be understood by those skilled in the art that the schematic diagram 4 is only the example of electronic equipment 3, do not constitute to electronic equipment
3 restriction may include perhaps combining certain components or different components, such as institute than illustrating more or fewer components
Stating electronic equipment 3 can also include input-output equipment, network access equipment, bus etc..
At least one described processor 32 can be central processing unit (Central Processing Unit, CPU),
It can also be other general processors, digital signal processor (Digital Signal Processor, DSP), dedicated integrated
Circuit (Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..The processor 32 can be microprocessor or the processor 32 is also possible to any conventional processor
Deng the processor 32 is the control centre of the electronic equipment 3, utilizes various interfaces and the entire electronic equipment 3 of connection
Various pieces.
The memory 31 can be used for storing the computer program 33 and/or module/unit, and the processor 32 passes through
Operation executes the computer program and/or module/unit being stored in the memory 31, and calls and be stored in memory
Data in 31 realize the various functions of the electronic equipment 3.The memory 31 can mainly include storing program area and storage
Data field, wherein storing program area can application program needed for storage program area, at least one function (for example sound plays
Function, image player function etc.) etc.；Storage data area, which can be stored, uses created data (such as sound according to electronic equipment 3
Frequency evidence, phone directory etc.) etc..In addition, memory 31 may include high-speed random access memory, it can also include non-volatile
Memory, such as hard disk, memory, plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital
(Secure Digital, SD) card, flash card (Flash Card), at least one disk memory, flush memory device or other
Volatile solid-state part.
If the integrated module/unit of the electronic equipment 3 is realized in the form of SFU software functional unit and as independent
Product when selling or using, can store in a computer readable storage medium.Based on this understanding, the present invention is real
All or part of the process in existing above-described embodiment method, can also instruct relevant hardware come complete by computer program
At the computer program can be stored in a computer readable storage medium, which is being executed by processor
When, it can be achieved that the step of above-mentioned each embodiment of the method.Wherein, the computer program includes computer program code, described
Computer program code can be source code form, object identification code form, executable file or certain intermediate forms etc..The meter
Calculation machine readable medium may include: can carry the computer program code any entity or device, recording medium, USB flash disk,
Mobile hard disk, magnetic disk, CD, computer storage, read-only memory (ROM, Read-Only Memory), random access memory
Device (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It needs to illustrate
It is that the content that the computer-readable medium includes can be fitted according to the requirement made laws in jurisdiction with patent practice
When increase and decrease, such as in certain jurisdictions, according to legislation and patent practice, computer-readable medium does not include electric carrier wave letter
Number and telecommunication signal.
In several embodiments provided by the present invention, it should be understood that disclosed electronic equipment and method, Ke Yitong
Other modes are crossed to realize.For example, electronic equipment embodiment described above is only schematical, for example, the unit
Division, only a kind of logical function partition, there may be another division manner in actual implementation.
It, can also be in addition, each functional unit in each embodiment of the present invention can integrate in same treatment unit
It is that each unit physically exists alone, can also be integrated in same unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of hardware adds software function module.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included in the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.This
Outside, it is clear that one word of " comprising " is not excluded for other units or, odd number is not excluded for plural number.The multiple units stated in system claims
Or device can also be implemented through software or hardware by a unit or device.The first, the second equal words are used to indicate name
Claim, and does not indicate any particular order.
Finally it should be noted that the above examples are only used to illustrate the technical scheme of the present invention and are not limiting, although reference
Preferred embodiment describes the invention in detail, those skilled in the art should understand that, it can be to of the invention
Technical solution is modified or equivalent replacement, without departing from the spirit of the technical scheme of the invention range.
Claims (10)
1. a kind of insurance policy number identification method, which is characterized in that the described method includes:
When receiving the first insurance policy picture, the corresponding insurance type of the first insurance policy picture is identified；
Based on preset insurance type and insures positional relationship of the single numbers in the first insurance policy picture and extract described the
First object line character region where insurance single numbers in one insurance policy picture；
The first identification model that training generates in advance is called to identify first object line character region, and output character
The corresponding score value of each of recognition result and the character identification result character identification result；
Whether the number for counting the target score value for being greater than default score value threshold value in all score value is 1；
When determining the number in all score value greater than the target score value of the default score value threshold value is 1, by the mesh
Mark recognition result of the corresponding character identification result of score value as the character in first object line character region.
2. the method as described in claim 1, which is characterized in that be greater than the default score value threshold in all score value when determining
When the number of the target score value of value is greater than 1, the method also includes:
The corresponding character identification result of target score value of the default score value threshold value will be greater than in the score value as institute
State the candidate characters collection of the character in first object line character region；
Show that the list of candidate characters recognition result and each candidate characters recognition result of the candidate characters concentration are corresponding
Score value；
When detect the candidate characters recognition result in the candidate characters recognition result list it is chosen when, time that will be chosen
Select character identification result as the recognition result of the character in first object line character region.
3. method according to claim 2, which is characterized in that identifying the character in first object line character region
Afterwards, the method also includes:
Positional relationship based on preset insurance type and insurer's name in the first insurance policy picture extracts institute
State the second target line character zone where insurer's name in the first insurance policy picture；
The the two the first identification models for calling training in advance to generate are identified to obtain institute to the second target line character zone
State insurer's name in the first insurance policy picture.
4. method as claimed in claim 3, which is characterized in that identifying the insurance in the second target line character zone
After people's name, the method also includes:
The insurance single numbers that will identify that, insurer's name, candidate characters collection, chosen candidate characters recognition result and described
First insurance policy picture is associated storage.
5. method as claimed in claim 4, which is characterized in that the method also includes:
When the second insurance policy picture for receiving the same insurer again, and if calling the first of the preparatory training generation to know
Other model obtains multiple candidate characters after carrying out character recognition to the first object line character region of the second insurance policy picture
When recognition result, according to the incidence relation of the insurer, candidate characters collection and chosen candidate characters recognition result, by institute
Identification of the selected character identification result as the correspondence character in the first object character zone of the second insurance policy picture
As a result.
6. method as claimed in claim 4, which is characterized in that the method also includes:
One display interface is provided；
The candidate characters recognition result column that the candidate characters are concentrated are shown according to default display mode on the display interface
Table and each corresponding score value of candidate characters recognition result.
7. the method as described in claim 1, which is characterized in that the default display mode includes:
Branch shows after the score value is ranked up according to descending sequence；
Show the first insurance policy picture；
Corresponding every row score value shows a tick boxes, to receive staff choose operation after, by what is be checked
Recognition result of the candidate characters recognition result as the character in the target line character zone.
8. a kind of insurance policy NID number identifier, which is characterized in that described device includes:
Receiving module when for receiving the first insurance policy picture, identifies the corresponding insurance type of the first insurance policy picture；
Extraction module, for being closed based on the position of preset insurance type and insurance single numbers in the first insurance policy picture
System extracts the first object line character region where the insurance single numbers in the first insurance policy picture；
Identification module, for calling the first identification model that training generates in advance to know first object line character region
Not, and the corresponding score value of each of output character recognition result and the character identification result character identification result；
Judgment module, whether the number for counting the target score value for being greater than default score value threshold value in all score value is 1；
Output module, for determining that the target in all score value greater than the default score value threshold value obtains when the judgment module
When the number of score value is 1, using the corresponding character identification result of the target score value as first object line character region
In character recognition result.
9. a kind of electronic equipment, which is characterized in that the electronic equipment includes processor and memory, and the processor is for holding
Insurance single numbers as claimed in any of claims 1 to 7 in one of claims are realized when the computer program stored in the row memory
Recognition methods.
10. a kind of computer readable storage medium, computer program, feature are stored on the computer readable storage medium
It is, insurance single numbers as claimed in any of claims 1 to 7 in one of claims is realized when the computer program is executed by processor
Recognition methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811537708.5A CN109918984A (en) | 2018-12-15 | 2018-12-15 | Insurance policy number identification method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811537708.5A CN109918984A (en) | 2018-12-15 | 2018-12-15 | Insurance policy number identification method, device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109918984A true CN109918984A (en) | 2019-06-21 |
Family
ID=66959984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811537708.5A Pending CN109918984A (en) | 2018-12-15 | 2018-12-15 | Insurance policy number identification method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109918984A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458184A (en) * | 2019-06-26 | 2019-11-15 | 平安科技（深圳）有限公司 | Optical character identification householder method, device, computer equipment and storage medium |
| CN110598527A (en) * | 2019-07-24 | 2019-12-20 | 平安科技（深圳）有限公司 | Machine learning-based claims insurance policy number identification method and related equipment |
| CN110807139A (en) * | 2019-10-23 | 2020-02-18 | 腾讯科技（深圳）有限公司 | Picture identification method and device, computer readable storage medium and computer equipment |
| CN110852896A (en) * | 2019-12-22 | 2020-02-28 | 上海眼控科技股份有限公司 | Date validity judgment method, date validity judgment device, date validity judgment equipment and storage medium |
| CN110942069A (en) * | 2019-10-22 | 2020-03-31 | 云南大学 | Method, device, storage medium and equipment for extracting key information of vehicle insurance policy |
| CN111144373A (en) * | 2019-12-31 | 2020-05-12 | 广州市昊链信息科技股份有限公司 | Information identification method and device, computer equipment and storage medium |
| CN111666868A (en) * | 2020-06-03 | 2020-09-15 | 阳光保险集团股份有限公司 | Insurance policy identification method and device and computer equipment |
| CN110598527B (en) * | 2019-07-24 | 2024-05-28 | 平安科技（深圳）有限公司 | Method and related equipment for identifying claim warranty number based on machine learning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101149805A (en) * | 2006-09-19 | 2008-03-26 | 北京三星通信技术研究有限公司 | Method and device for hand writing identification using character structural information for post treatment |
| CN102750272A (en) * | 2012-07-02 | 2012-10-24 | 安徽科大讯飞信息科技股份有限公司 | Method and system for optimizing hand-input candidate item of character |
| CN106557747A (en) * | 2016-11-15 | 2017-04-05 | 平安科技（深圳）有限公司 | The method and device of identification insurance single numbers |
| CN107958201A (en) * | 2017-10-13 | 2018-04-24 | 上海眼控科技股份有限公司 | A kind of intelligent checking system and method for vehicle annual test insurance policy form |
| CN108288078A (en) * | 2017-12-07 | 2018-07-17 | 腾讯科技（深圳）有限公司 | Character identifying method, device and medium in a kind of image |
| CN108875722A (en) * | 2017-12-27 | 2018-11-23 | 北京旷视科技有限公司 | Character recognition and identification model training method, device and system and storage medium |
-
2018
- 2018-12-15 CN CN201811537708.5A patent/CN109918984A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101149805A (en) * | 2006-09-19 | 2008-03-26 | 北京三星通信技术研究有限公司 | Method and device for hand writing identification using character structural information for post treatment |
| CN102750272A (en) * | 2012-07-02 | 2012-10-24 | 安徽科大讯飞信息科技股份有限公司 | Method and system for optimizing hand-input candidate item of character |
| CN106557747A (en) * | 2016-11-15 | 2017-04-05 | 平安科技（深圳）有限公司 | The method and device of identification insurance single numbers |
| CN107958201A (en) * | 2017-10-13 | 2018-04-24 | 上海眼控科技股份有限公司 | A kind of intelligent checking system and method for vehicle annual test insurance policy form |
| CN108288078A (en) * | 2017-12-07 | 2018-07-17 | 腾讯科技（深圳）有限公司 | Character identifying method, device and medium in a kind of image |
| CN108875722A (en) * | 2017-12-27 | 2018-11-23 | 北京旷视科技有限公司 | Character recognition and identification model training method, device and system and storage medium |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458184A (en) * | 2019-06-26 | 2019-11-15 | 平安科技（深圳）有限公司 | Optical character identification householder method, device, computer equipment and storage medium |
| WO2020258500A1 (en) * | 2019-06-26 | 2020-12-30 | 平安科技（深圳）有限公司 | Optical character recognition assisting method and apparatus, computer device and storage medium |
| CN110458184B (en) * | 2019-06-26 | 2023-06-30 | 平安科技（深圳）有限公司 | Optical character recognition assistance method, device, computer equipment and storage medium |
| CN110598527A (en) * | 2019-07-24 | 2019-12-20 | 平安科技（深圳）有限公司 | Machine learning-based claims insurance policy number identification method and related equipment |
| CN110598527B (en) * | 2019-07-24 | 2024-05-28 | 平安科技（深圳）有限公司 | Method and related equipment for identifying claim warranty number based on machine learning |
| CN110942069A (en) * | 2019-10-22 | 2020-03-31 | 云南大学 | Method, device, storage medium and equipment for extracting key information of vehicle insurance policy |
| CN110807139A (en) * | 2019-10-23 | 2020-02-18 | 腾讯科技（深圳）有限公司 | Picture identification method and device, computer readable storage medium and computer equipment |
| CN110807139B (en) * | 2019-10-23 | 2023-09-01 | 腾讯科技（深圳）有限公司 | Picture identification method, device, computer readable storage medium and computer equipment |
| CN110852896A (en) * | 2019-12-22 | 2020-02-28 | 上海眼控科技股份有限公司 | Date validity judgment method, date validity judgment device, date validity judgment equipment and storage medium |
| CN111144373A (en) * | 2019-12-31 | 2020-05-12 | 广州市昊链信息科技股份有限公司 | Information identification method and device, computer equipment and storage medium |
| CN111666868A (en) * | 2020-06-03 | 2020-09-15 | 阳光保险集团股份有限公司 | Insurance policy identification method and device and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109918984A (en) | Insurance policy number identification method, device, electronic equipment and storage medium | |
| CN108427953A (en) | A kind of character recognition method and device | |
| US20190370291A1 (en) | Node task data display method and apparatus, storage medium and computer equipment | |
| CN112507936B (en) | Image information auditing method and device, electronic equipment and readable storage medium | |
| CN107679997A (en) | Method, apparatus, terminal device and storage medium are refused to pay in medical treatment Claims Resolution | |
| CN111476324B (en) | Traffic data labeling method, device, equipment and medium based on artificial intelligence | |
| CN107784517A (en) | A kind of customer information screening method and terminal device | |
| CN105095415A (en) | Method and apparatus for confirming network emotion | |
| CN109828906A (en) | UI automated testing method, device, electronic equipment and storage medium | |
| CN115640074A (en) | Service data processing method and device and intelligent counter terminal | |
| CN113127621A (en) | Dialogue module pushing method, device, equipment and storage medium | |
| CN112783825B (en) | Data archiving method, device, computer device and storage medium | |
| CN112307464A (en) | Fraud identification method and device and electronic equipment | |
| CN114638501A (en) | Business data processing method and device, computer equipment and storage medium | |
| CN110363667A (en) | Order financing processing method, apparatus, computer equipment and storage medium based on AI | |
| CN110472246A (en) | Work order classification method, device and storage medium | |
| CN113553431A (en) | User label extraction method, device, equipment and medium | |
| CN113010785A (en) | User recommendation method and device | |
| CN113591900A (en) | Identification method and device for high-demand response potential user and terminal equipment | |
| CN117520503A (en) | Financial customer service dialogue generation method, device, equipment and medium based on LLM model | |
| CN109993381B (en) | Demand management application method, device, equipment and medium based on knowledge graph | |
| CN110929519B (en) | Entity attribute extraction method and device | |
| CN108898167A (en) | It breaks one's promise the display methods and device of number | |
| CN113377980A (en) | Information labeling method and device, electronic equipment and storage medium | |
| CN107071553A (en) | A kind of method, device and computer-readable recording medium for changing video speech |
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 |