CN109871770A - Property ownership certificate recognition methods, device, equipment and storage medium - Google Patents
Property ownership certificate recognition methods, device, equipment and storage medium Download PDFInfo
- Publication number
- CN109871770A CN109871770A CN201910047886.8A CN201910047886A CN109871770A CN 109871770 A CN109871770 A CN 109871770A CN 201910047886 A CN201910047886 A CN 201910047886A CN 109871770 A CN109871770 A CN 109871770A
- Authority
- CN
- China
- Prior art keywords
- picture
- property ownership
- ownership certificate
- identified
- model
- 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
Abstract
The present invention relates to image procossing, a kind of property ownership certificate recognition methods, device, equipment and storage medium are disclosed this method comprises: obtaining property ownership certificate picture to be identified, property ownership certificate picture to be identified is input to default picture classification model to obtain picture classification result；Then according to picture classification result determine property ownership certificate picture to be identified belonging to Target Photo type and obtain corresponding picture recognition region；Picture material identification is carried out to property ownership certificate picture to be identified further according to picture recognition region, obtains the corresponding property ownership certificate content of property ownership certificate picture to be identified.Since the present invention is by the way that property ownership certificate picture to be input in preparatory trained picture classification model, then the picture recognition region of picture material identification is determined according to picture classification result, picture material identification is carried out based on picture recognition region again and obtains property ownership certificate content, and then can effectively identify and parse all types of property ownership certificate image contents, it is provided convenience for the typing of property ownership certificate content information.
Description
Technical field
The present invention relates to image identification technical field more particularly to a kind of property ownership certificate recognition methods, device, equipment and storages
Medium.
Background technique
Property ownership certificate (Premises Permit) refers to that house purchaser by transaction, obtains the legal title in house, can be in accordance with the law
Purchased house room is exercised occupy, the certificate of the right of use, income and punishment.In house prosperity transaction, property ownership certificate is indispensable
Certificate, staff and house purchaser can check property ownership certificate information repeatedly especially in second-hand house process of exchange, then will
Property ownership certificate information solicitation is into different orders.
However in traditional real estate industry, there is no the information collection that any a product can accomplish property ownership certificate, rooms
Produce the acquisition that the acquisition of card information still uses " naked eyes identify, are manually entered " to carry out information；Traditional typing mode have with
Lower disadvantage: 1. inefficiency, 2. accuracys rate are low, 3. inconvenient data switch.Simultaneously as house property certificate has in different geographical
The property ownership certificate of the information of different editions, the whole nation is many kinds of, makes troubles to identification, therefore, it is necessary to be directed to different rooms
Card feature is produced, the method for parsing each version property ownership certificate photo content can effectively be identified by designing one kind.
Above content is only used to facilitate the understanding of the technical scheme, and is not represented and is recognized that above content is existing skill
Art.
Summary of the invention
The main purpose of the present invention is to provide a kind of property ownership certificate recognition methods, device, equipment and storage mediums, it is intended to
The technical issues of parsing each version property ownership certificate content can not effectively be identified by solving the prior art.
To achieve the above object, it the present invention provides a kind of property ownership certificate recognition methods, the described method comprises the following steps:
Property ownership certificate picture to be identified is obtained, the property ownership certificate picture to be identified is input to default picture classification model to obtain
Take picture classification result；
According to the picture classification result determine the property ownership certificate picture to be identified belonging to Target Photo type, and obtain
The corresponding picture recognition region of the Target Photo type；
Picture material identification is carried out to the property ownership certificate picture to be identified according to the picture recognition region, obtain it is described to
Identify the corresponding property ownership certificate content of property ownership certificate picture.
Preferably, described to obtain property ownership certificate picture to be identified, the property ownership certificate picture to be identified is input to default picture
Before the step of disaggregated model is to obtain picture classification result, the method also includes:
Corresponding property ownership certificate picture is chosen from picture library by default picture type, and according to the property ownership certificate picture structure of selection
The model training pictures of established model verifying pictures and preset quantity；
Training picture in each model training pictures is input to initial picture disaggregated model and carries out model training, is obtained
The corresponding picture classification model to be verified of each model training pictures；
According to the model verify picture concentrate include verifying picture respectively to the picture classification model to be verified into
Row verifying, and default picture classification model is filtered out from the picture classification model to be verified according to verification result.
Preferably, the verification result includes the picture classification model to be verified to the sorted standard of the verifying picture
True rate and recall rate；
The step for filtering out default picture classification model from the picture classification model to be verified according to verification result
Suddenly, comprising:
It is scored according to the accuracy rate and the recall rate each picture classification model to be verified, obtains scoring knot
Fruit；
Default picture classification model is filtered out from the picture classification model to be verified according to the appraisal result.
It is preferably, described to be scored according to the accuracy rate and the recall rate each picture classification model to be verified,
The step of obtaining appraisal result, comprising:
According to the accuracy rate and the recall rate, each picture classification model to be verified is commented by preset formula
Point, obtain appraisal result；
Wherein, the preset formula are as follows:
Fscore=(2*precision*recall)/(precision+recall)
In formula, FscoreFor appraisal result, precision is accuracy rate, and recall is recall rate.
Preferably, described to filter out default picture point from the picture classification model to be verified according to the appraisal result
The step of class model, comprising:
The appraisal result is ranked up by sequence from high to low, and according to ranking results by sort first scoring
As a result it is used as target appraisal result；
Using the corresponding picture classification model to be verified of the target appraisal result as default picture classification model.
Preferably, the figure of different picture types is corresponded in the picture classification result comprising the property ownership certificate picture to be identified
Piece class probability；
It is described according to the picture classification result determine the property ownership certificate picture to be identified belonging to Target Photo type, and
The step of obtaining the Target Photo type corresponding picture recognition region, comprising:
The picture classification probability of maximum probability is chosen from the picture classification probability as Target Photo class probability；
The corresponding Target Photo type of the property ownership certificate picture to be identified is determined according to the Target Photo class probability,
The corresponding picture recognition region of the Target Photo type is searched in the mapping table constructed in advance.
Preferably, described to obtain property ownership certificate picture to be identified, the property ownership certificate picture to be identified is input to default picture
Before the step of disaggregated model is to obtain picture classification result, the method also includes:
Mapping building instruction is received, the picture type to be associated that includes and described wait close is read in the mapping building instruction
Join the corresponding picture recognition region of picture type；
The picture type to be associated picture recognition region corresponding with the picture type to be associated is associated, and
Association results are saved to mapping table.
In addition, to achieve the above object, the present invention also proposes a kind of property ownership certificate identification device, described device includes:
The property ownership certificate picture to be identified is input to pre- by picture classification module for obtaining property ownership certificate picture to be identified
If picture classification model is to obtain picture classification result；
Region obtain module, for according to the picture classification result determine the property ownership certificate picture to be identified belonging to mesh
It marks on a map sheet type, and obtains the corresponding picture recognition region of the Target Photo type；
Content identifier module, for being carried out in image according to the picture recognition region to the property ownership certificate picture to be identified
Hold identification, obtains the corresponding property ownership certificate content of the property ownership certificate picture to be identified.
In addition, to achieve the above object, the present invention also proposes that a kind of property ownership certificate identification equipment, the equipment include: storage
Device, processor and it is stored in the property ownership certificate recognizer that can be run on the memory and on the processor, the house property
The step of card recognizer is arranged for carrying out property ownership certificate recognition methods as described above.
In addition, to achieve the above object, the present invention also proposes a kind of storage medium, house property is stored on the storage medium
Recognizer is demonstrate,proved, the property ownership certificate recognizer realizes the step of property ownership certificate recognition methods as described above when being executed by processor
Suddenly.
Property ownership certificate picture to be identified is input to default picture classification mould by obtaining property ownership certificate picture to be identified by the present invention
Type is to obtain picture classification result；Then according to picture classification result determine property ownership certificate picture to be identified belonging to Target Photo class
Type simultaneously obtains corresponding picture recognition region；Picture material knowledge is carried out to property ownership certificate picture to be identified further according to picture recognition region
Not, the corresponding property ownership certificate content of property ownership certificate picture to be identified is obtained.Since the present invention is pre- by the way that property ownership certificate picture to be input to
First in trained picture classification model, the picture recognition region of picture material identification is then determined according to picture classification result,
Picture material identification is carried out based on picture recognition region again and obtains property ownership certificate content, and then can effectively identify that parsing is all types of
Property ownership certificate image content, provided convenience for the typing of property ownership certificate content information.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the property ownership certificate identification equipment for the hardware running environment that the embodiment of the present invention is related to；
Fig. 2 is the flow diagram of property ownership certificate recognition methods first embodiment of the present invention；
Fig. 3 is the flow diagram of property ownership certificate recognition methods second embodiment of the present invention；
Fig. 4 is the flow diagram of property ownership certificate recognition methods 3rd embodiment of the present invention；
Fig. 5 is the structural block diagram of property ownership certificate identification device first embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that described herein, specific examples are only used to explain the present invention, is not intended to limit the present invention.
Referring to Fig.1, Fig. 1 is that the property ownership certificate for the hardware running environment that the embodiment of the present invention is related to identifies that device structure shows
It is intended to.
As shown in Figure 1, property ownership certificate identification equipment may include: processor 1001, such as central processing unit (Central
Processing Unit, CPU), communication bus 1002, user interface 1003, network interface 1004, memory 1005.Wherein,
Communication bus 1002 is for realizing the connection communication between these components.User interface 1003 may include display screen
(Display), input unit such as keyboard (Keyboard), optional user interface 1003 can also include that the wired of standard connects
Mouth, wireless interface.Network interface 1004 optionally may include standard wireline interface and wireless interface (such as Wireless Fidelity
(WIreless-FIdelity, WI-FI) interface).Memory 1005 can be the random access memory (Random of high speed
Access Memory, RAM) memory, be also possible to stable nonvolatile memory (Non-Volatile Memory,
), such as magnetic disk storage NVM.Memory 1005 optionally can also be the storage device independently of aforementioned processor 1001.
It will be understood by those skilled in the art that structure shown in Fig. 1 does not constitute the restriction to property ownership certificate identification equipment,
It may include perhaps combining certain components or different component layouts than illustrating more or fewer components.
As shown in Figure 1, as may include operating system, data storage mould in a kind of memory 1005 of storage medium
Block, network communication module, Subscriber Interface Module SIM and property ownership certificate recognizer.
In property ownership certificate identification equipment shown in Fig. 1, network interface 1004 is mainly used for carrying out data with network server
Communication；User interface 1003 is mainly used for carrying out data interaction with user；Processor in property ownership certificate identification equipment of the present invention
1001, memory 1005 can be set in property ownership certificate identification equipment, and the property ownership certificate identification equipment is adjusted by processor 1001
With the property ownership certificate recognizer stored in memory 1005, and execute property ownership certificate recognition methods provided in an embodiment of the present invention.
The embodiment of the invention provides a kind of property ownership certificate recognition methods, are property ownership certificate identification side of the present invention referring to Fig. 2, Fig. 2
The flow diagram of method first embodiment.
In the present embodiment, the property ownership certificate recognition methods the following steps are included:
Step S10: obtaining property ownership certificate picture to be identified, and the property ownership certificate picture to be identified is input to default picture classification
Model is to obtain picture classification result；
It should be noted that the executing subject of the method for the present invention can be with image real time transfer, program operation and
The computer or terminal device (hereinafter referred to as identification terminal) of network communicating function.The default picture classification model can be pre-
First trained convolutional neural networks (the Convolutional Neural for classifying to property ownership certificate picture
Network, CNN) model.
Further, it is contemplated that in practical application, picture classification training is carried out to a new CNN model completely and needs base
In large-scale image data base (such as ImageNet) Lai Shixian, this mode will undoubtedly expend a large amount of manpower and material resources.Therefore,
Improved Inception model can be directly based upon in the present embodiment to realize image classification.Certainly, improved in the present embodiment
Inception model can be according to the characteristics of image of property ownership certificate and carry out to Inception model (it is in the nature CNN model)
The model (the i.e. described default picture classification model) obtained after special training.
It will be appreciated that the output of Inception model the last layer is the set of softmax probability value, usual situation
Under the corresponding softmax probability value of a certain picture type is higher indicates that the picture may more belong to this picture type.Such as
After property ownership certificate picture A is input to default picture classification model, the picture classification result of acquisition is (picture type 1=60%, figure
Sheet type 2=30%, picture type 3=10%), then show that the picture type maximum probability of property ownership certificate picture A belongs to picture type
1。
In the concrete realization, identification terminal get currently need to carry out the property ownership certificate picture of content recognition when, can will
Property ownership certificate picture to be identified is input in preparatory trained picture classification model and obtains picture classification result.
Step S20: according to the picture classification result determine the property ownership certificate picture to be identified belonging to Target Photo class
Type, and obtain the corresponding picture recognition region of the Target Photo type；
It should be noted that the picture type, i.e., according to region/Time segments division picture type of property ownership certificate, such as
The old card in Shanghai/new card, the old card/new card in Guangzhou, the new card/old card in Nanjing etc..
It should be understood that since the property ownership certificate picture of different region different periods there may be biggish difference, because
This corresponding property ownership certificate information of different property ownership certificate pictures (such as property rights people, identity card title, ID card No., room
Room is located address, shared situation, enrollment time, house situation etc.) region may not also be identical the location of in property ownership certificate,
Staff can configure corresponding picture recognition area previously according to the property ownership certificate picture that the picture type of property ownership certificate is each type
Domain, it is quick according to the picture type in order to which identification terminal is after determining picture type belonging to property ownership certificate picture to be identified
Determine corresponding picture recognition region.
In the concrete realization, identification terminal is after the picture classification result for getting default picture classification model output, i.e.,
Target Photo type belonging to the property ownership certificate picture to be identified can be determined according to the picture classification result, then described in acquisition
The corresponding picture recognition region of Target Photo type.
Step S30: picture material identification is carried out to the property ownership certificate picture to be identified according to the picture recognition region, is obtained
Take the corresponding property ownership certificate content of the property ownership certificate picture to be identified.
It should be understood that described image content recognition (Image content recognition) is one from picture
Identify the computer technology of picture material (i.e. practical judgment), these objects include personage, landscape, building, article, equipment
Deng.
In the concrete realization, identification terminal can carry out property ownership certificate picture to be identified based on the picture recognition region determined
Picture material identification, so that the corresponding property ownership certificate content of property ownership certificate picture to be identified is obtained, then using the property ownership certificate got
Content carries out subsequent data input.
Property ownership certificate picture to be identified is input to default picture classification by obtaining property ownership certificate picture to be identified by the present embodiment
Model is to obtain picture classification result；Then according to picture classification result determine property ownership certificate picture to be identified belonging to Target Photo
Type simultaneously obtains corresponding picture recognition region；Picture material is carried out to property ownership certificate picture to be identified further according to picture recognition region
Identification, obtains the corresponding property ownership certificate content of property ownership certificate picture to be identified.Since the present embodiment is by inputting property ownership certificate picture
Into preparatory trained picture classification model, the picture recognition area of picture material identification is then determined according to picture classification result
Domain, then picture material identification is carried out based on picture recognition region and obtains property ownership certificate content, and then can effectively identify that parsing is each
The property ownership certificate image content of type provides convenience for the typing of property ownership certificate content information.
With reference to Fig. 3, Fig. 3 is the flow diagram of property ownership certificate recognition methods second embodiment of the present invention.
Based on above-mentioned first embodiment, in the present embodiment, before the step S10, the method also includes:
Step S01: corresponding property ownership certificate picture is chosen from picture library by default picture type, and according to the house property of selection
Demonstrate,prove the model training pictures of picture building model verifying pictures and preset quantity；
It should be understood that storage has the property ownership certificate picture of a large amount of different geographical different periods in advance in the picture library.Institute
State model training pictures have it is multiple, and in each model training pictures each type of property ownership certificate picture quantity it is identical,
But property ownership certificate picture number is different between each model training pictures, for example, in the first model training pictures Shanghai it is new demonstrate,prove/
Old card, the new of Guangzhou demonstrate,prove/demonstrate,prove each 1000 always, the new card in Shanghai in the second model training pictures/demonstrate,prove always, Guangzhou it is new demonstrate,prove/
Each 2000 are demonstrate,proved always.
In addition, the quantity of model verifying pictures is preferably one in the present embodiment, and model verifying picture is concentrated
It contains all types of property ownership certificate pictures and quantity is identical, such as the new card in 500 Shanghai, the old card in 500 Shanghai, 500
The new card of Zhang Guangzhou, the old card etc. in 500 Guangzhou.
In the concrete realization, identification terminal can select respectively different number difference according to default picture type from picture library
The property ownership certificate picture building model verifying pictures and model training pictures of type.
Step S02: the training picture in each model training pictures is input to initial picture disaggregated model and carries out model
Training, obtains the corresponding picture classification model to be verified of each model training pictures；
It should be noted that the initial picture disaggregated model can be CNN model, it is also possible to using above-described embodiment
Described in Inception model.
In the concrete realization, identification terminal can read different types of property ownership certificate picture from model training pictures, so
The picture read is sequentially inputted to afterwards to carry out model training in Inception model, obtains each model training picture
Collect corresponding picture classification model to be verified.Wherein, the quantity of the picture classification model to be verified and the model training figure
The quantity of piece collection is identical.
Step S03: picture is verified according to the model and concentrates the verifying picture for including respectively to the picture to be verified point
Class model is verified, and filters out default picture classification mould from the picture classification model to be verified according to verification result
Type.
It will be appreciated that for guarantee model training result accuracy, identification terminal can from model verify picture concentrate with
Machine chooses verifying picture, then is separately input to verify in the picture classification model to be verified by the picture of selection, and
Obtain verification result.
Further, in this embodiment verification result includes that picture classification model to be verified concentrates model verifying picture
Verify the sorted accuracy rate of picture and recall rate.It should be understood that two, fields such as information retrieval, classification, identification, translation
Most basic index is recall rate (Recall Rate) and accuracy rate (Precision Rate), and recall rate is also recall ratio, accurately
Rate is also precision ratio.
In the concrete realization, identification terminal can according in verification result the accuracy rate and the recall rate to each to be tested
Card picture classification model carries out scoring and obtains appraisal result, then further according to the appraisal result from the picture classification to be verified
Default picture classification model is filtered out in model.
In view of under actual conditions, recall rate and accuracy rate be it is shifting, be difficult to take into account, the present embodiment is considered as
One can integrate the mathematical formulae of two kinds of indexs to measure the superiority and inferiority of picture classification model to be verified.Specifically, identification terminal
It can be scored by preset formula each picture classification model to be verified, be obtained according to the accuracy rate and the recall rate
Appraisal result；Wherein, the preset formula are as follows:
Fscore=(2*precision*recall)/(precision+recall)
In formula, FscoreFor appraisal result, precision is accuracy rate, and recall is recall rate.
It should be noted that F in above-mentioned formulascoreNeither arithmetic mean of instantaneous value, nor geometrical mean, it is possible to understand that
It is geometrical mean square divided by arithmetic mean of instantaneous value.
In the concrete realization, identification terminal is calculating each picture classification model correspondence to be verified according to above-mentioned preset formula
Appraisal result after, the appraisal result can be also ranked up by sequence from high to low, and according to ranking results will sort
First appraisal result is as target appraisal result；Using the corresponding picture classification model to be verified of the target appraisal result as
Default picture classification model.For example, the corresponding scoring knot of identification terminal calculated picture classification model a, b, c to be verified
Fruit is 98.8%, 99.1%, 97.9%, learns that the appraisal result of sequence first is after sequence sequence from high to low
99.1%, it may thereby determine that out that optimal picture classification model to be verified is picture classification model b to be verified, can incite somebody to action at this time
Picture classification model b to be verified is as the default picture classification model.
The present embodiment carries out model training to initial picture disaggregated model by model training pictures, then passes through model
Verifying pictures verify the picture classification model to be verified after training, filter out optimal preset further according to verification result
Picture classification model ensure that the accuracy of the model for property ownership certificate picture classification.
With reference to Fig. 4, Fig. 4 is the flow diagram of property ownership certificate recognition methods 3rd embodiment of the present invention.
It include the property ownership certificate picture to be identified in picture classification result described in the present embodiment based on the various embodiments described above
The picture classification probability of corresponding different picture types；
Correspondingly, the step S20 may particularly include:
Step S201: the picture classification probability of maximum probability is chosen from the picture classification probability as Target Photo point
Class probability；
It should be understood that the output of all Inception model the last layeres is softmax probability value set (i.e. institute
Have a set of picture classification probability, and all probability values and for 100%).Under normal conditions, Inception model only takes
The forward preceding 5 progress classification results of probability value are shown.Such as the picture of " jaguar " is opened to Inception mode input one,
After the calculating of Inception category of model, the final probability value set that obtains is (leopard " jaguar "=70%, cheetah
" cheetah "=18%, jaguar " jaguar "=3%, snow leopard " snow leopard "=1%), it at this time can be from the picture
The picture classification probability " 70% " that maximum probability is chosen in class probability is used as Target Photo class probability, then further according to the mesh
Piece class probability of marking on a map determines picture type.
In this step, identification terminal can be obtained from picture classification result property ownership certificate picture to be identified picture classification it is general
Then rate set therefrom chooses the picture classification probability of maximum probability as Target Photo class probability according to the set.
Step S202: the corresponding target figure of the property ownership certificate picture to be identified is determined according to the Target Photo class probability
Sheet type searches the corresponding picture recognition region of the Target Photo type in the mapping table constructed in advance.
It should be noted that before executing this step, the present embodiment property ownership certificate recognition methods further include: receive mapping structure
Instruction is built, the picture type to be associated for including in the mapping building instruction and the corresponding figure of the picture type to be associated are read
Piece identification region；The picture type to be associated picture recognition region corresponding with the picture type to be associated is closed
Connection, and association results are saved to mapping table.
It should be understood that the corresponding picture recognition region of different types of property ownership certificate picture is different, to improve property ownership certificate
Recognition efficiency, staff can first pass through property ownership certificate picture corresponding feature of the identification terminal by each type in advance in the present embodiment
Region is associated with picture type, such as Pekinese is newly demonstrate,proved and Pekinese newly demonstrate,proves corresponding picture recognition region and closes
Connection, then saves association results to mapping table, so that identification terminal is determining that property ownership certificate picture to be identified is corresponding
When picture type, corresponding picture recognition region is searched according to the mapping table.
In the concrete realization, identification terminal determines the property ownership certificate to be identified according to the Target Photo class probability selected
The corresponding Target Photo type of picture, it is corresponding then to search the Target Photo type in the mapping table constructed in advance
Picture recognition region carries out picture material identification to property ownership certificate picture to be identified according to picture recognition region so as to subsequent.
The picture classification that the present embodiment passes through the selection maximum probability from the picture classification probability that picture classification result includes
Probability is as Target Photo class probability；Then the corresponding mesh of property ownership certificate picture to be identified is determined according to Target Photo class probability
It marks on a map sheet type, the corresponding picture recognition region of Target Photo type is searched in the mapping table constructed in advance, is simplified
Property ownership certificate identification process, improves recognition efficiency.
In addition, the embodiment of the present invention also proposes a kind of storage medium, property ownership certificate identification journey is stored on the storage medium
The step of sequence, the property ownership certificate recognizer realizes property ownership certificate recognition methods as described above when being executed by processor.
It is the structural block diagram of property ownership certificate identification device first embodiment of the present invention referring to Fig. 5, Fig. 5.
As shown in figure 5, the property ownership certificate identification device that the embodiment of the present invention proposes includes:
The property ownership certificate picture to be identified is input to by picture classification module 501 for obtaining property ownership certificate picture to be identified
Picture classification model is preset to obtain picture classification result；
Region obtains module 502, belonging to determining the property ownership certificate picture to be identified according to the picture classification result
Target Photo type, and obtain the corresponding picture recognition region of the Target Photo type；
Content identifier module 503, for carrying out figure to the property ownership certificate picture to be identified according to the picture recognition region
As content recognition, the corresponding property ownership certificate content of the property ownership certificate picture to be identified is obtained.
Property ownership certificate picture to be identified is input to default picture classification by obtaining property ownership certificate picture to be identified by the present embodiment
Model is to obtain picture classification result；Then according to picture classification result determine property ownership certificate picture to be identified belonging to Target Photo
Type simultaneously obtains corresponding picture recognition region；Picture material is carried out to property ownership certificate picture to be identified further according to picture recognition region
Identification, obtains the corresponding property ownership certificate content of property ownership certificate picture to be identified.Since the present embodiment is by inputting property ownership certificate picture
Into preparatory trained picture classification model, the picture recognition area of picture material identification is then determined according to picture classification result
Domain, then picture material identification is carried out based on picture recognition region and obtains property ownership certificate content, and then can effectively identify that parsing is each
The property ownership certificate image content of type provides convenience for the typing of property ownership certificate content information.
Based on the above-mentioned property ownership certificate identification device first embodiment of the present invention, the second of property ownership certificate identification device of the present invention is proposed
Embodiment.
In the present embodiment, the property ownership certificate identification device further includes model training module, the model training module, is used
In choosing corresponding property ownership certificate picture from picture library by default picture type, and model is constructed according to the property ownership certificate picture of selection
Verify the model training pictures of pictures and preset quantity；Training picture in each model training pictures is input to initially
Picture classification model carries out model training, obtains the corresponding picture classification model to be verified of each model training pictures；According to institute
Stating model verifying picture concentrates the verifying picture for including to verify respectively to the picture classification model to be verified, and according to testing
Card result filters out default picture classification model from the picture classification model to be verified.
Further, the model training module is also used to according to the accuracy rate and the recall rate to each to be verified
Picture classification model scores, and obtains appraisal result；According to the appraisal result from the picture classification model to be verified
Filter out default picture classification model.
Further, the model training module is also used to pass through default public affairs according to the accuracy rate and the recall rate
Formula scores to each picture classification model to be verified, obtains appraisal result；Wherein, the preset formula are as follows:
Fscore=(2*precision*recall)/(precision+recall)
In formula, FscoreFor appraisal result, precision is accuracy rate, and recall is recall rate.
Further, the model training module is also used to arrange the appraisal result by sequence from high to low
Sequence, and according to ranking results using sort first appraisal result as target appraisal result；The target appraisal result is corresponding
Picture classification model to be verified as default picture classification model.
Further, the region obtains module 502, is also used to choose maximum probability from the picture classification probability
Picture classification probability is as Target Photo class probability；The property ownership certificate to be identified is determined according to the Target Photo class probability
The corresponding Target Photo type of picture searches the corresponding picture of the Target Photo type in the mapping table constructed in advance
Identification region.
Further, the property ownership certificate identification device further includes mapping building module, and the mapping constructs module, for connecing
Mapping building instruction is received, the picture type to be associated and the picture type to be associated for including in the mapping building instruction are read
Corresponding picture recognition region；By the picture type to be associated picture recognition region corresponding with the picture type to be associated
It is associated, and association results is saved to mapping table.
The other embodiments or specific implementation of property ownership certificate identification device of the present invention can refer to above-mentioned each method embodiment,
Details are not described herein again.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the system that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in a storage medium
In (such as read-only memory/random access memory, magnetic disk, CD), including some instructions are used so that a terminal device (can
To be mobile phone, computer, server, air conditioner or the network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of property ownership certificate recognition methods, which is characterized in that the described method includes:
Property ownership certificate picture to be identified is obtained, the property ownership certificate picture to be identified is input to default picture classification model to obtain figure
Piece classification results；
According to the picture classification result determine the property ownership certificate picture to be identified belonging to Target Photo type, and obtain described in
The corresponding picture recognition region of Target Photo type；
Picture material identification is carried out to the property ownership certificate picture to be identified according to the picture recognition region, is obtained described to be identified
The corresponding property ownership certificate content of property ownership certificate picture.
2. the method as described in claim 1, which is characterized in that it is described to obtain property ownership certificate picture to be identified, it will be described to be identified
Property ownership certificate picture was input to before the step of default picture classification model is to obtain picture classification result, the method also includes:
Corresponding property ownership certificate picture is chosen from picture library by default picture type, and mould is constructed according to the property ownership certificate picture of selection
The model training pictures of type verifying pictures and preset quantity；
Training picture in each model training pictures is input to initial picture disaggregated model and carries out model training, obtains each mould
The corresponding picture classification model to be verified of type training pictures；
Verifying picture according to the model concentrates the verifying picture for including to test respectively the picture classification model to be verified
Card, and default picture classification model is filtered out from the picture classification model to be verified according to verification result.
3. method according to claim 2, which is characterized in that the verification result includes the picture classification model to be verified
To the sorted accuracy rate of the verifying picture and recall rate；
Described the step of default picture classification model is filtered out from the picture classification model to be verified according to verification result, packet
It includes:
It is scored according to the accuracy rate and the recall rate each picture classification model to be verified, obtains appraisal result；
Default picture classification model is filtered out from the picture classification model to be verified according to the appraisal result.
4. method as claimed in claim 3, which is characterized in that it is described according to the accuracy rate and the recall rate to each to be tested
The step of card picture classification model scores, obtains appraisal result, comprising:
According to the accuracy rate and the recall rate, is scored by preset formula each picture classification model to be verified, obtained
Take appraisal result；
Wherein, the preset formula are as follows:
Fscore=(2*precision*recall)/(precision+recall)
In formula, FscoreFor appraisal result, precision is accuracy rate, and recall is recall rate.
5. method as claimed in claim 4, which is characterized in that described to be divided according to the appraisal result from the picture to be verified
The step of default picture classification model is filtered out in class model, comprising:
The appraisal result is ranked up by sequence from high to low, and according to ranking results by sort first appraisal result
As target appraisal result；
Using the corresponding picture classification model to be verified of the target appraisal result as default picture classification model.
6. such as method described in any one of claim 1 to 5, which is characterized in that in the picture classification result comprising it is described to
Identification property ownership certificate picture corresponds to the picture classification probability of different picture types；
It is described according to the picture classification result determine the property ownership certificate picture to be identified belonging to Target Photo type, and obtain
The step of Target Photo type corresponding picture recognition region, comprising:
The picture classification probability of maximum probability is chosen from the picture classification probability as Target Photo class probability；
The corresponding Target Photo type of the property ownership certificate picture to be identified is determined according to the Target Photo class probability, preparatory
The corresponding picture recognition region of the Target Photo type is searched in the mapping table of building.
7. method as claimed in claim 6, which is characterized in that it is described to obtain property ownership certificate picture to be identified, it will be described to be identified
Property ownership certificate picture was input to before the step of default picture classification model is to obtain picture classification result, the method also includes:
Mapping building instruction is received, the picture type to be associated for including in the mapping building instruction and the figure to be associated are read
The corresponding picture recognition region of sheet type；
The picture type to be associated picture recognition region corresponding with the picture type to be associated is associated, and will be closed
It is coupled fruit to save to mapping table.
8. a kind of property ownership certificate identification device, which is characterized in that described device includes:
The property ownership certificate picture to be identified is input to default figure for obtaining property ownership certificate picture to be identified by picture classification module
Piece disaggregated model is to obtain picture classification result；
Region obtain module, for according to the picture classification result determine the property ownership certificate picture to be identified belonging to target figure
Sheet type, and obtain the corresponding picture recognition region of the Target Photo type；
Content identifier module, for carrying out picture material knowledge to the property ownership certificate picture to be identified according to the picture recognition region
Not, the corresponding property ownership certificate content of the property ownership certificate picture to be identified is obtained.
9. a kind of property ownership certificate identifies equipment, which is characterized in that the equipment includes: memory, processor and is stored in described deposit
On reservoir and the property ownership certificate recognizer that can run on the processor, the property ownership certificate recognizer are arranged for carrying out such as power
Benefit require any one of 1 to 7 described in property ownership certificate recognition methods the step of.
10. a kind of storage medium, which is characterized in that be stored with property ownership certificate recognizer, the property ownership certificate on the storage medium
The step of property ownership certificate recognition methods as described in any one of claim 1 to 7 is realized when recognizer is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910047886.8A CN109871770A (en) | 2019-01-17 | 2019-01-17 | Property ownership certificate recognition methods, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910047886.8A CN109871770A (en) | 2019-01-17 | 2019-01-17 | Property ownership certificate recognition methods, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109871770A true CN109871770A (en) | 2019-06-11 |
Family
ID=66917758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910047886.8A Pending CN109871770A (en) | 2019-01-17 | 2019-01-17 | Property ownership certificate recognition methods, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109871770A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111243159A (en) * | 2020-01-20 | 2020-06-05 | 支付宝实验室(新加坡)有限公司 | Counterfeit certificate identification method and device and electronic equipment |
| CN111476113A (en) * | 2020-03-20 | 2020-07-31 | 中保车服科技服务股份有限公司 | Card identification method, device and equipment based on transfer learning and readable medium |
| CN112733916A (en) * | 2020-12-31 | 2021-04-30 | 五八有限公司 | False certificate picture identification method and device, electronic equipment and storage medium |
| CN114241485A (en) * | 2022-02-24 | 2022-03-25 | 深圳大道云科技有限公司 | Information identification method, device, equipment and storage medium of property certificate |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107341508A (en) * | 2017-06-22 | 2017-11-10 | 北京飞搜科技有限公司 | A kind of quick cuisines image identification method and system |
| CN107633255A (en) * | 2017-08-11 | 2018-01-26 | 天津大学 | A kind of rock lithology automatic recognition classification method under deep learning pattern |
| CN108038504A (en) * | 2017-12-11 | 2018-05-15 | 深圳房讯通信息技术有限公司 | A kind of method for parsing property ownership certificate photo content |
| CN108491885A (en) * | 2018-03-28 | 2018-09-04 | 广东工业大学 | A kind of autoCAD graphic blocks identifying method and devices based on Naive Bayes Classifier |
| WO2018191421A1 (en) * | 2017-04-11 | 2018-10-18 | Alibaba Group Holding Limited | Image-based vehicle damage determining method, apparatus, and electronic device |
-
2019
- 2019-01-17 CN CN201910047886.8A patent/CN109871770A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018191421A1 (en) * | 2017-04-11 | 2018-10-18 | Alibaba Group Holding Limited | Image-based vehicle damage determining method, apparatus, and electronic device |
| CN107341508A (en) * | 2017-06-22 | 2017-11-10 | 北京飞搜科技有限公司 | A kind of quick cuisines image identification method and system |
| CN107633255A (en) * | 2017-08-11 | 2018-01-26 | 天津大学 | A kind of rock lithology automatic recognition classification method under deep learning pattern |
| CN108038504A (en) * | 2017-12-11 | 2018-05-15 | 深圳房讯通信息技术有限公司 | A kind of method for parsing property ownership certificate photo content |
| CN108491885A (en) * | 2018-03-28 | 2018-09-04 | 广东工业大学 | A kind of autoCAD graphic blocks identifying method and devices based on Naive Bayes Classifier |
Non-Patent Citations (1)
| Title |
|---|
| 常参参: "基于OCR技术的通用证件识别系统", 《信息科技》, no. 12, pages 8 - 55 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111243159A (en) * | 2020-01-20 | 2020-06-05 | 支付宝实验室(新加坡)有限公司 | Counterfeit certificate identification method and device and electronic equipment |
| CN111476113A (en) * | 2020-03-20 | 2020-07-31 | 中保车服科技服务股份有限公司 | Card identification method, device and equipment based on transfer learning and readable medium |
| CN112733916A (en) * | 2020-12-31 | 2021-04-30 | 五八有限公司 | False certificate picture identification method and device, electronic equipment and storage medium |
| CN112733916B (en) * | 2020-12-31 | 2023-04-07 | 五八有限公司 | False certificate picture identification method and device, electronic equipment and storage medium |
| CN114241485A (en) * | 2022-02-24 | 2022-03-25 | 深圳大道云科技有限公司 | Information identification method, device, equipment and storage medium of property certificate |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109871770A (en) | Property ownership certificate recognition methods, device, equipment and storage medium | |
| CN105306495B (en) | user identification method and device | |
| CN103916436B (en) | Information-pushing method, device, terminal and server | |
| CN104598518B (en) | Content pushing method and device | |
| CN107679189A (en) | A kind of point of interest update method, device, server and medium | |
| CN108763519A (en) | The recommendation method, apparatus and readable storage medium storing program for executing of reading | |
| CN107291715A (en) | Resume appraisal procedure and device | |
| CN108416003A (en) | A kind of picture classification method and device, terminal, storage medium | |
| CN106874279A (en) | Generate the method and device of applicating category label | |
| CN106503006A (en) | The sort method and device of application App neutron applications | |
| CN106484766B (en) | Searching method and device based on artificial intelligence | |
| CN104838420A (en) | Rotation of image based on image content to correct image orientation | |
| CN108230113A (en) | User's portrait generation method, device, equipment and readable storage medium storing program for executing | |
| CN109947949A (en) | Knowledge information intelligent management, device and server | |
| CN104346408B (en) | A kind of method and apparatus being labeled to the network user | |
| CN107729361A (en) | It is automatically synthesized picture method for pushing, device and storage medium | |
| CN107818491A (en) | Electronic installation, Products Show method and storage medium based on user's Internet data | |
| CN108132963A (en) | Resource recommendation method and device, computing device and storage medium | |
| CN107526846A (en) | Generation, sort method, device, server and the medium of channel sequencing model | |
| CN109783624A (en) | Answer generation method, device and the intelligent conversational system in knowledge based library | |
| CN110347777A (en) | A kind of classification method, device, server and the storage medium of point of interest POI | |
| CN106776760A (en) | Be applied to intelligent terminal searches topic method and device | |
| CN106569691A (en) | Mobile terminal application icon arrangement method and system | |
| CN106843941A (en) | Information processing method, device and computer equipment | |
| CN108566666A (en) | Wi-Fi Hotspot recommends method, apparatus and storage medium |
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 |