CN110489747A - A kind of image processing method, device, storage medium and electronic equipment - Google Patents
A kind of image processing method, device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN110489747A CN110489747A CN201910704473.2A CN201910704473A CN110489747A CN 110489747 A CN110489747 A CN 110489747A CN 201910704473 A CN201910704473 A CN 201910704473A CN 110489747 A CN110489747 A CN 110489747A
- Authority
- CN
- China
- Prior art keywords
- text
- image
- content
- feature information
- semantic feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003860 storage Methods 0.000 title claims abstract description 21
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 73
- 238000000034 method Methods 0.000 claims abstract description 51
- 238000004458 analytical method Methods 0.000 claims abstract description 40
- 239000000284 extract Substances 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 7
- 238000012550 audit Methods 0.000 abstract description 28
- 230000006870 function Effects 0.000 description 20
- 238000010586 diagram Methods 0.000 description 10
- 230000008451 emotion Effects 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000009826 distribution Methods 0.000 description 8
- 239000000203 mixture Substances 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000012163 sequencing technique Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 208000010086 Hypertelorism Diseases 0.000 description 2
- 206010020771 Hypertelorism of orbit Diseases 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Abstract
The embodiment of the present application discloses a kind of image processing method, device, storage medium and server, wherein, method includes the image that acquisition includes content of text, described image is generated based on service request, described image is identified, obtain the content of text in described image, extract the first semantic feature information of the content of text, the second semantic feature information based on the first semantic feature information received text corresponding with the service request is analyzed and processed, based on analysis and processing result, determine whether the business application passes through.Using the embodiment of the present application, it is ensured that the accuracy rate of business audit improves the treatment effeciency of business audit.
Description
Technical field
This application involves field of computer technology more particularly to a kind of image processing method, device, storage medium and electronics
Equipment.
Background technique
In the application, there can be the case where auditing to content included in image, based on to contained interior in image
The audit of appearance determines corresponding operation, such as: during business approval, service request side can upload related image, be based on
Identification to image decides whether to ratify its request.For another example, during online teaching, student and teacher can by
Line educational system completes the content of remote teaching, such as: teacher can arrange the work of teaching by on-line education system to student
Industry, student can upload to the image (such as operation photo, operation screenshot etc.) comprising this subjob after fulfiling assignment
Line educational system is simultaneously initiated business audit (work correction), is completed with request confirmation operation, is obtained corresponding achievement and/or acquisition
Reward (such as integral, comment)；Etc..Currently, by the way of manual examination and verification, can there is effect when in face of business audit request
The problems such as rate is low, recognition accuracy is low；And it is handled by the way of image comparison (i.e. by screenshot compared with existing picture)
Service request can also have processing result inaccuracy.
Summary of the invention
The embodiment of the present application provides a kind of image processing method, device, storage medium and server, it is ensured that business
The accuracy rate of audit improves the treatment effeciency of business audit.The technical solution is as follows:
In a first aspect, the embodiment of the present application provides a kind of image processing method, which comprises
Acquisition includes the image of content of text, and described image is generated based on service request；
Described image is identified, the content of text in described image is obtained；
Extract the first semantic feature information of the content of text；
The second semantic feature letter based on the first semantic feature information received text corresponding with the service request
Breath is analyzed and processed；
Based on analysis and processing result, determine whether the business application passes through.
Second aspect, the embodiment of the present application provide a kind of image processing apparatus, and described device includes:
Image collection module, for obtain include content of text image, described image based on service request generate；
Picture recognition module obtains the content of text in described image for identifying to described image；
Characteristic extracting module, for extracting the first semantic feature information of the content of text；
Analysis and processing module, for being based on the first semantic feature information received text corresponding with the service request
The second semantic feature information be analyzed and processed；
Apply for determining module, for being based on analysis and processing result, determines whether the business application passes through.
The third aspect, the embodiment of the present application provide a kind of computer storage medium, and the computer storage medium is stored with
A plurality of instruction, described instruction are suitable for being loaded by processor and executing above-mentioned method and step.
Fourth aspect, the embodiment of the present application provide a kind of server, it may include: processor and memory；Wherein, described to deposit
Reservoir is stored with computer program, and the computer program is suitable for being loaded by the processor and executing above-mentioned method and step.
The technical solution bring beneficial effect that some embodiments of the application provide includes at least:
In the application one or more embodiment, service server acquisition includes the image of content of text, the figure
As being generated based on service request, described image is identified, the content of text in described image is obtained, extracts the text
The first semantic feature information of this content is based on the first semantic feature information received text corresponding with the service request
The second semantic feature information be analyzed and processed, be based on analysis and processing result, determine whether the business application passes through.Pass through
Image recognition grade when to service request is extracted to obtain corresponding first semantic feature information, based on the first semantic feature information and
Whether the processing result of the second semantic feature information is passed through with the determination business application, it is ensured that business audit it is accurate
Rate.Described image is audited using service server simultaneously, improves the treatment effeciency of business audit.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, 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
Some embodiments of application for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of scene framework schematic diagram of image procossing provided by the embodiments of the present application；
Fig. 2 is the interface signal for the tip distance adjustment that a kind of image processing method provided by the embodiments of the present application is related to
Figure；
Fig. 3 is a kind of flow diagram of image processing method provided by the embodiments of the present application；
Fig. 4 is the homework uploading scene figure that a kind of image processing method provided by the embodiments of the present application is related to；
Fig. 5 is the interface schematic diagram that the image that a kind of image processing method provided by the embodiments of the present application is related to uploads；
Fig. 6 is the scene figure for the acquisition image that a kind of image processing method provided by the embodiments of the present application is related to；
Fig. 7 is the flow diagram of another image processing method provided by the embodiments of the present application；
Fig. 8 is a kind of structural schematic diagram of image processing apparatus provided by the embodiments of the present application；
Fig. 9 is a kind of structural schematic diagram of picture recognition module provided by the embodiments of the present application；
Figure 10 is the structural schematic diagram of another image processing apparatus provided by the embodiments of the present application；
Figure 11 is a kind of structural schematic diagram of server provided by the embodiments of the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
In the description of the present application, it is to be understood that term " first ", " second " etc. are used for description purposes only, without
It can be interpreted as indication or suggestion relative importance.In the description of the present application, it should be noted that unless otherwise specific regulation
And restriction, " comprising " and " having " and their any deformations, it is intended that cover and non-exclusive include.Such as contain a system
The process, method, system, product or equipment of column step or unit are not limited to listed step or unit, but optional
Ground further includes the steps that not listing or unit, or optionally further comprising intrinsic for these process, methods, product or equipment
Other step or units.For the ordinary skill in the art, above-mentioned term can be understood in the application with concrete condition
In concrete meaning.In addition, unless otherwise indicated, " multiple " refer to two or more in the description of the present application." and/
Or ", the incidence relation of affiliated partner is described, indicates may exist three kinds of relationships, for example, A and/or B, can indicate: individually depositing
In A, A and B, these three situations of individualism B are existed simultaneously.It is a kind of "or" that character "/", which typicallys represent forward-backward correlation object,
Relationship.
The application is described in detail below with reference to specific embodiment.
It referring to Figure 1, is a kind of configuration diagram of image processing system provided by the embodiments of the present application.As shown in Figure 1,
Described image processing system may include service server a and user terminal cluster.The user terminal cluster may include more
A user terminal, as shown in Figure 1, specifically include user terminal 1, user terminal 2 ..., user terminal n, n is the integer greater than 0；
For ease of understanding, the embodiment of the present invention is described by taking the user terminal 1 in Fig. 1 as an example.
The user terminal can be with the terminal device for uploading image function, including but not limited to: wearable device,
Handheld device, tablet computer, mobile unit, calculates equipment or is connected to other processing of radio modem PC
Equipment etc..
The service server 100 can be individual server apparatus, such as: rack, blade, tower or machine
Cabinet type server apparatus, or have stronger computing capability hardware device using work station, mainframe computer etc.；It is also possible to adopt
The server cluster formed with multiple servers, each server in the service cluster can be to be formed in a symmetrical,
Wherein every server function equivalence, status in service link is of equal value, and each server individually can externally provide service, described
The service of being provided separately can be understood as the auxiliary without other server.
The user terminal is communicated by network with service server, and network can be wireless network, is also possible to
Cable network, wireless network include but is not limited to cellular network, WLAN, infrared network or blueteeth network, cable network
Including but not limited to Ethernet, universal serial bus (universal serial bus, USB) or controller local area network.
User when applying for business is included that the image image of content of text is uploaded to service server by user terminal 1
a。
Described image refers to the description or description to a kind of similitude, vividness of business object, by way of image
Carry out the objective relevant information indicated when the service request generates.Usual described image is exactly the picture with visual effect, In
It can be understood as photo relevant to the business, screenshot, manual draw etc. in the present embodiment.
Specifically, user terminal 1 to service server a send service request by way of, requested service server pair
The user when applying for business include content of text image carry out industry audit.User terminal 1 has camera function, when
User upload application business when including the image of content of text, user terminal 1 open camera function, pass through built-in figure
As acquisition device or external image acquisition device user include content of text image, described image acquisition device
It can be one or more cameras.In the case where the quantity of camera is multiple, multiple cameras can be distributed in difference
Position form camera array, it includes text that it is collected, which to obtain each camera, by camera array for user terminal 1
The image of content, by multiple channels it is collected include content of text image merges to obtain high-fidelity includes text
The image of this content.
Optionally, user terminal 1 acquisition user when applying for business when including the image of content of text, Ke Yifen
Multi collect or acquisition multiple groups user when applying for business include content of text image, then refer to according to the user's choice
Enable from multiple groups include select user in the image of content of text one it is final include that the image of content of text uploads
To service server a.
Optionally, in the image process for including content of text of the user terminal 1 when acquiring user and applying for business, for
For user terminal 1, after user terminal 1 opens camera function, one is had effectively by image acquisition device image
Image Acquisition distance range, the effective image acquisition distance range refer to user terminal in the range to collected object institute
Acquired image can be identified.
In a kind of possible embodiment, user terminal 1 applies for user image real-time monitoring when business, judges quilt
Acquisition target (such as user needs paper, the operation etc. submitted) works as quilt whether within effective image acquisition distance range
For acquisition target when effective image acquires except distance range, user terminal 1 then shows that prompt information, the prompt information are used for
Indicate that user adjusts the relative distance of collected object and user terminal 1.
For example, user terminal 1 effective image acquisition distance range be 0-30cm, be collected object (such as user need
Paper, operation of submission etc.) it is 35cm at a distance from the image collecting device of user terminal 1, user terminal 1 real-time monitors figure
As acquisition hypertelorism, not within effective image acquisition distance range, user terminal 1 is shown as shown in Figure 2 on a display screen
The text prompt information of " hypertelorism please adjust at a distance from camera ", the collected object of prompt user's adjustment (such as user
Paper, operation for needing to submit etc.) at a distance from camera, user can pass through further collected object and camera at this time
Distance, so that user terminal 1 collects more good image.
What service server a reception user terminal 1 uploaded includes the image of content of text, and knows to described image
Not, the content of text in described image is obtained.
The process identified to described image, including but not limited to analyzes described image, processed and is located
Reason, the described image processing technique being related to during being identified to described image can be text detection identification technology,
Image recognition technology based on machine learning method, can image recognition technology based on text identification classifier etc..
Specifically, service server a carries out region division to described image, multiple regions image is obtained, to the multiple
Area image is identified respectively, obtains the corresponding region content of text of each area image, and the region content of text is carried out
Combination, obtains the content of text in described image.
When the quantity of described image is multiple, service server a obtains auxiliary information corresponding to described multiple images
(uplink time, the user gradation for generating user corresponding to time, described image of described image, described image of described image
The service priority of corresponding business), then each image is identified, obtains the content of text in each image.
Finally the content of text in each image is integrated based on the auxiliary information
Service server a based in the content of text keyword and/or critical sentence extract the first semantic feature information,
The second semantic feature information based on the first semantic feature information received text corresponding with the service request is divided
Analysis processing.
The semantic feature refers to the distinctive semantic attribute of the unstructured data of literal expression, is with a paper
Example, semantic feature includes the semantic features such as author's creation intention, Data subject explanation, low-level image feature meaning.The semantic feature
Information is can to express the semantic of object itself and various features semantic in the environment, by taking content of text as an example, institute's predicate
Adopted characteristic information can be composition letter, the sequence of word, the emotion information of word, mutual information etc..The first semantic feature information
In the present embodiment, it can be understood as the semantic feature information of the content of text, the first semantic feature information is for institute
State for content of text, it may include but be not limited to the key word information of text, word frequency distribution information, syntactic level entity letter
Breath, theme of semantic class etc..Second semantic feature information of the received text can be understood as the mark in the present embodiment
The semantic feature information of quasi- text, for the received text, it may include but not the second semantic feature information
It is limited to key word information, word frequency distribution information, the entity information of syntactic level, the theme of semantic class etc. of received text.
Service server a is specifically based on the first semantic feature information and the second semantic feature information, calculates institute
State the semantic similarity of content of text Yu the received text.
When the semantic similarity meets given threshold, it is determined that the business application passes through, and service server a is to user
The 1 successful prompt information of outgoing traffic application of terminal.
When the semantic similarity is unsatisfactory for given threshold, it is determined that the business application does not pass through, service server a
To the prompt information of 1 outgoing traffic application of user terminal failure.
In the embodiment of the present application, service server acquisition includes the image of content of text, and described image is based on business
Request generates, and identifies to described image, obtains the content of text in described image, extracts the of the content of text
One semantic feature information, second based on the first semantic feature information received text corresponding with the service request is semantic
Characteristic information is analyzed and processed, and is based on analysis and processing result, is determined whether the business application passes through.By to service request
When image recognition grade extract to obtain corresponding first semantic feature information, it is semantic special based on the first semantic feature information and second
Whether the processing result of reference breath is passed through with the determination business application, it is ensured that the accuracy rate of business audit.It uses simultaneously
Service server audits described image, and the treatment effeciency of business audit can be improved.
In one embodiment, as shown in figure 3, spy proposes a kind of image processing method, this method can be dependent on calculating
Machine program is realized, can run on the image processing apparatus based on von Neumann system.The computer program can be integrated in application
In, it also can be used as independent tool-class application operation.The image processing apparatus can be server, work station, mainframe computer
Equal electronic equipments.
Specifically, the image processing method includes:
Step 101: acquisition includes the image of content of text, and described image is generated based on service request.
Described image refers to the description or description to a kind of similitude, vividness of business object, by way of image
Carry out the objective relevant information indicated when the service request generates.Usual described image is exactly the picture with visual effect, In
It can be understood as photo relevant to the business, screenshot, manual draw etc. in the present embodiment.
The service request can be understood as user by user terminal to service server application business when send, industry
Business server is based on the service request and carries out audit or examination & approval processing to the business.Usually when user terminal takes to business
It, can be by the image for including content of text relevant to the application business together with service request when device transmission service request of being engaged in
It is sent to service server.
Specifically, when user upload application business when image when, user terminal 1 can open camera function, by interior
The image of acquisition is sent to by the image of the image collecting device or external image acquisition device user set, user terminal
Service server, service server receive the image that user is uploaded when applying for business by user terminal.
For example, the user terminal can be computer, hand in the business scenario of a specific application " correcting students' papers "
The smart machines such as machine, plate are illustrated by smart phone of user terminal for convenience of description, and the computer can be equipped with
The application program of remote teaching, the application program of the remote teaching have the function of application " correcting students' papers ", and user opens hand
The application program display interface of machine, as shown in figure 4, if user needs to upload the image of the business of this application " correcting students' papers "
When, camera function acquisition user can be opened by " camera " button in finger touch Fig. 4 display interface, mobile phone to be needed to upload
Operation, after acquisition image-operation is completed, mobile phone can pop up prompting frame as shown in Figure 5, prompting frame shown in fig. 5
Middle prompt has " present image has acquired, if uploads ", and user can click ACK button, the industry of above-mentioned application " correcting students' papers "
Business is triggered, and mobile phone monitors the operation for choosing " confirmation " button to upload image, and executes the industry of the application " correcting students' papers "
The corresponding machine-executable instruction of control logic that business is triggered, mobile phone " correct work by executing described instruction, by the application
The corresponding image of the business of industry " is sent to service server, and service server receives the business in application " correcting students' papers "
When the image that is uploaded, save to wait into local cache and handle in next step.
Optionally, service server obtains the image of user terminal uploads, and described image can be user terminal acquisition and work as
Corresponding image in the application business procedure of preceding display interface, it can be understood as the user terminal, which has to touch, writes function
Can, user terminal shows that the corresponding needs of this application business are filled out during user's application business, in current display interface
The prompt information write, user can directly fill in this application industry according to the prompt information on the display interface of user terminal
The content of business, described fill in can be by way of stylus or direct finger touch screen, after completing to fill in, user
The image comprising content of text that terminal will fill in completion is uploaded to service server.
For example, the display interface of Intelligent flat is aobvious as shown in fig. 6, user terminal-Intelligent flat, which has, touches writing function
Being shown with user, this needs the online assignment completed, and user is according to prompt information-topic information on display interface, to the topic
It answers, can be by way of stylus or direct finger touch screen, after completing to answer, user can be clicked
Submitting button, applies for the business of " correcting students' papers ", user terminal by the image of current display interface as shown in FIG. 6 (such as when
The screenshot of front interface) service server is uploaded to apply for the business of " correcting students' papers ".
In a kind of feasible embodiment, user terminal stores the image comprising content of text in advance, when user to
Server upload application business when the image comprising content of text when, can by directly access user terminal storage sky
Between, at least one image comprising content of text relevant to the business is found, and user need to not be called when applying for business
Function (such as camera function) acquisition of the acquisition image of terminal includes the image of content of text, and described image can be and the Shen
Please the relevant screenshot of business, the photo that shoots in advance etc. include content of text image.
For example, user submits to service server about " applying for some project funds " in a business approval scene
Examination & approval, before user is requested by user terminal to service server initiation business approval, meeting storage in advance is good and described " to be applied
The relevant image comprising content of text of some project funds " business, described image, which can be, is related to the project details
Related screenshot, user can initiate asking for business approval to service server by option that touch-control teleservice is examined
Ask, and by the memory space of user terminal, find it is relevant to the business at least one include the project details
Related screenshot, and the related screenshot is sent to service server with service request together.Service server receives described
Service request simultaneously obtains the related screenshot comprising the project details, identify in next step to described image, obtains
The step of content of text in described image.
Step 102: described image being identified, the content of text in described image is obtained.
The content of text refers to the written representation of language, usually has complete, system meaning a sentence
The combination of sub or multiple sentences.The content of text can be a word, a sentence, a paragraph by taking English language as an example,
The content of text can be the practice form of language, usually refer to some spoken and written languages in specific implementation, in this implementation
Content of text described in example can be understood as in described image comprising the corresponding content of text of all spoken and written languages.
Specifically, being located in advance after service server obtains the image uploaded when applying for business to described image
Reason, the pretreatment include the treatment processes such as digitlization, geometric transformation, normalization, smooth, recovery enhancing, eliminate nothing in image
The information of pass is carried out image recovery and enhancing to the image section comprising text information, is then calculated using preset text identification
Method identifies pretreated image, converts content of text for the image information in described image.
Optionally, the preset text recognition algorithms can be partitioning algorithm based on optical character identification, can be
Extraction algorithm based on text filed positioning, the sequence that can be detection algorithm based on scene text identification, view-based access control model object
Column recognizer etc..
It it should be noted that there are many text recognition algorithms, can be above-mentioned one or more, do not make to have herein
Body limits.
Optionally, service server is defeated by the image after the pretreatment after pre-processing to described image
Enter into text identification model, the corresponding content of text of output image.
By obtaining great amount of samples data in advance, characteristic information is extracted, and be labeled to the sample information, the spy
The characteristic information of characteristic information of the reference breath comprising image and the content of text, creates text identification model.The text is known
Other model can be based on convolutional neural networks (Convolutional Neural Network, CNN) model, depth nerve net
Network (Deep Neural Network, DNN) model, Recognition with Recurrent Neural Network (Recurrent Neural Networks, RNN) mould
Type, insertion (embedding) model, gradient promote decision tree (Gradient Boosting Decision Tree, GBDT) mould
What at least one of type, logistic regression (Logistic Regression, LR) model were realized, based on the sample marked
Data are trained text identification model, available trained text identification model.
In a kind of feasible embodiment, service server carries out region division to described image, available multiple
Then area image identifies the multiple area image respectively, obtain the corresponding region content of text of each area image,
The region content of text is combined, the content of text in described image is obtained.
Step 103: extracting the first semantic feature information of the content of text.
The semantic feature refers to the distinctive semantic attribute of the unstructured data of literal expression, is with a paper
Example, semantic feature includes the semantic features such as author's creation intention, Data subject explanation, low-level image feature meaning.The semantic feature
Information is can to express the semantic of object itself and various features semantic in the environment, by taking content of text as an example, institute's predicate
Adopted characteristic information can be composition letter, the sequence of word, the emotion information of word, mutual information etc..
Wherein, composition letter is that a word is made of which letter, the sequencing relationship of these letters.
Word order is the sequencing of expression in short each word of (meaning) composition.
The emotion information of word is word emotion meaning expressed in this sentence, and the emotion meaning is understood that
For word sentence be commendation or derogatory sense, be high or droning, be joyful or sad etc..
Mutual information refers to the statistical iteration relationship between some word or word and classification, and it is right that mutual information is commonly applied to measurement two
Reciprocity as between.
The first semantic feature information is in the present embodiment, it can be understood as the semantic feature of the content of text is believed
Breath, for the content of text, it may include but is not limited to the keyword letter of text the first semantic feature information
Breath, word frequency distribution information, the entity information of syntactic level, the theme of semantic class etc..
Specifically, server identifies described image, after the content of text identified, extraction of semantics mould is utilized
Type extracts the first semantic feature information of the content of text.
Optionally, the extraction of semantics model can be the text feature information extracting method based on Context Framework, i.e., first
It determines the extraction element (sentence, word, word, symbol etc.) of content of text, semantic analysis is then incorporated into statistic algorithm to the text
Content carries out extraction processing, obtains the first semantic feature information of the content of text；It can be special based on ontological text
Extracting method is levied, i.e., exports the content of text using the content of text as input using ontology (On-tology) model
The first semantic feature information；It can be the concept characteristic extracting method based on Hownet, i.e., based on the feature extraction of concept characteristic
Method carries out semantic point to the content of text on the basis of vector space model (Vector Space Model, VSM)
Analysis obtains the semantic information of vocabulary using the database of Hownet, and semantic identical vocabulary is mapped to identical concept, is then clustered
Then word after being clustered, and the characteristic item of the text vector as VSM model carry out model calculation etc., etc..It needs
Bright is that the mode of the first semantic feature information for extracting the content of text has and much can be above-mentioned one kind or more
The fitting of kind, is not construed as limiting herein.
Step 104: the second language based on the first semantic feature information received text corresponding with the service request
Adopted characteristic information is analyzed and processed.
The received text refers to the spoken and written languages pre-set, the second semantic feature information of the received text
It can be understood as the semantic feature information of the received text in the present embodiment, the second semantic feature information is for described
For received text, it may include but be not limited to the key word information of received text, word frequency distribution information, syntactic level entity
Information, theme of semantic class etc..
Second semantic feature information of these received texts and the corresponding relationship of the service request are stored in received text
In library, can be understood as service server in the present embodiment can obtain the service request pair in the received text library
Second semantic feature information of the received text answered.
Specifically, service server obtains institute from the received text information bank according to the service identification of the service request
The second semantic feature information of the corresponding received text of service identification is stated, the service identification can be understood as the user and initiate
The identification informations such as the type of the business of business audit or examination & approval, the ID of distribution, business-level, can according to the service identification
To identify the business of the service request instruction, the first semantic feature information is matched with the second semantic feature information
Processing, obtains the matching result of matching treatment.
Optionally, the matching treatment mode can be to the first semantic feature information and the second semantic feature information
Calculate similarity, can be it is similar to the first semantic feature information and the second semantic feature information calculating similarity calculation away from
From can be the calculating difference characteristic information to the first semantic feature information and the second semantic feature information, be then based on
Difference characteristic information is graded or is scored, etc..
Step 105: being based on analysis and processing result, determine whether the business application passes through.
Specifically, service server is after obtaining the analysis and processing result, it is regular to described point according to default judge
Analysis processing result is judged, and when the analysis and processing result, which reaches the judge that application passes through, to be required, determines the business Shen
It please pass through；When the analysis and processing result not up to applies for that the judge passed through requires, determine that the business application does not pass through.
Optionally, when the analysis and processing result is based on the first semantic feature information and the second semantic feature information
Semantic similarity when, the judge rule can be setting similarity threshold, when the analysis and processing result semanteme it is similar
When degree reaches similarity threshold, determine that the business application passes through；When the semantic similarity of the analysis and processing result is not up to
When similarity threshold, determine that the business application does not pass through.
Optionally, when the analysis and processing result is the phase based on the first semantic feature information with the second semantic feature information
Like apart from when, the judges rule can be is arranged similarity distance threshold value, when the similarity of the analysis and processing result reaches phase
When like distance threshold, determine that the business application passes through；When the similarity of the analysis and processing result is not up to similarity distance threshold
When value, determine that the business application does not pass through.
Optionally, when the analysis and processing result is the difference based on the first semantic feature information Yu the second semantic feature information
When the grading or scoring of different characteristic information, the judge rule, which can be, is arranged similar grade threshold value or similar point of threshold value, when described
When the similarity of analysis and processing result reaches similar grade threshold value or similar point of threshold value, determine that the business application passes through；When described
When the similarity of analysis and processing result is not up to similar grade threshold value or similar point of threshold value, determine that the business application does not pass through.
In a kind of specific implement scene, server can be by the first semantic feature information and the second semantic feature information
It is matched, obtains the difference text information of the first semantic feature information Yu the second semantic feature information, the difference text
This information includes difference amount of text (such as quantity of difference word), difference content of text, the part of speech of difference content of text etc..
Then server is based on default difference code of points, and the default difference code of points can be, and setting total score 100 is divided, and is arranged
Deduction of points radix and difference type deduction of points coefficient and difference text part of speech deduction of points coefficient, score to the difference text information,
When appraisal result reaches the scoring score that the business application passes through (such as 80 points), determine that the business application passes through；When
When appraisal result is not up to the scoring score that the business application passes through (such as 79 points), determine that the business application does not pass through.
In the embodiment of the present application, service server acquisition includes the image of content of text, and described image is based on business
Request generates, and identifies to described image, obtains the content of text in described image, extracts the of the content of text
One semantic feature information, second based on the first semantic feature information received text corresponding with the service request is semantic
Characteristic information is analyzed and processed, and is based on analysis and processing result, is determined whether the business application passes through.By to service request
When image recognition grade extract to obtain corresponding first semantic feature information, it is semantic special based on the first semantic feature information and second
Whether the processing result of reference breath is passed through with the determination business application, it is ensured that the accuracy rate of business audit.It uses simultaneously
Service server audits described image, and the treatment effeciency of business audit can be improved.
Fig. 7 is referred to, Fig. 7 is a kind of process signal of another embodiment for image processing method that the application proposes
Figure.It is specific:
Step 201: acquisition includes the image of content of text, and described image is generated based on service request.
For details, reference can be made to steps 101, and details are not described herein again.
Step 202: when the quantity of described image is multiple, obtaining auxiliary information corresponding to described multiple images.
Wherein, the auxiliary information includes at least one following: the uplink time of described image, described image generation when
Between, the service priority of business corresponding to the user gradation of user corresponding to described image, described image.
Uplink time stamp indicate a data it is already existing, complete before the upload of some specific time, can
The data of verifying, usually a character string, for providing a electronic evidence for user, to prove the upload of certain data
Time point.It can be understood as the time point that the user terminal uploads include content of text image in the present embodiment.
The user gradation can be understood as user being divided into different brackets according to the actual situation, such as by the level of consumption it is high,
In net, the time is longer or the user of real-name authentication is divided into VIP user, the level of consumption is low, divide in net time shorter user
For ordinary user etc..The standard that user gradation divides can fixation according to the actual situation, and subsequent adjustable.In this implementation
It is to be understood that service server establishes the mapping relations of user identifier and user gradation in example, and the mapping relations are protected
It deposits into user identifier set.The user identifier refers to the title of identity user identity, such as user's pet name, user name, use
Family account etc. belongs to user identifier.In practical applications, user identifier can be can recognize by series of computation machine system
Character composition, such as letter, Chinese character, Arabic numerals, additional character etc..
The service priority refers to service server when handling the image of each business, and determine each business includes text
This content images receives the parameter of the priority level of computing resource, and the service priority can be understood as in advance to different type
Business configured.For example, can set the service priority of work correction type to 1, classroom is answerred questions the industry of type
Business priority is set as 2, and the service priority of remote test type is set as 3, is arranged with the service priority of hall detection type
It is 4, then the business processing sequence of 4 types of service is successively are as follows: image > classroom of work correction type is answerred questions the figure of type
Picture > remote test type image > with the image of hall detection type.
Specifically, service server and user terminal cluster form service link, each service section on the service link
Point (service server, user terminal etc.) constantly works, when multiple user terminals are concurrently or consecutively to next service node
When service server sends image, or when a user terminal continues described image being transmitted to next service through service link
When node traffic server, server can receive multiple images simultaneously at this time.
Optionally, when described image includes multiple, service server parses the data of the multiple images of acquisition,
The timestamp for obtaining each image in described multiple images, then converts the stamp, obtains the upload of described multiple images
Time, according to the uplink time, the described image earliest to current uplink time is identified.
Optionally, for user terminal when sending service request to server, user is applied for institute when business by user terminal
It states image and the user identifier is packaged, be uploaded to service server in the form of data packet, service server is receiving
When to multiple data packets, the multiple data packet is parsed, obtain in the multiple data packet the image of each data packet and
The user identifier obtains the corresponding user gradation of the user identifier according to the user identifier in user identifier set,
Service server, can be according to the height of user gradation, to current user's level highest when identifying to each image
User upload described image identify.
Optionally, user is applied for that described image and the type of service when business are packaged by user terminal, with number
Be uploaded to service server according to the form of packet, service server when receiving multiple data packets, to the multiple data packet into
Row parsing, obtains the image of each data packet and the type of service in the multiple data packet, and it is corresponding to obtain the type of service
Priority.Service server, can be according to the business of the type of service when identifying to each image
Priority, to identifying for the highest priority of current business.
Step 203: each image being identified, the content of text in each image is obtained, is based on the auxiliary
Information integrates the content of text in each image.
Specifically, when described multiple images are that same user is applying for the upload of some business (such as certain asset examination and approval business)
When, user can be the described image uploaded several times when applying for the business, and server can be based on described image at this time
Priority parameters, the priority is arranged according to successive each image in described multiple images of uplink time in uplink time
Parameter refers to that computer time sharing system when handling multiple operation procedures, determines that each operation procedure receives system resource
The parameter of priority level.Give each images match one priority parameters, for indicating the priority level of these images.At this point,
Server handles the highest image of current priority, successively using the highest image of current priority as target image,
The target image is input in text identification model trained in advance, to obtain the content of text of the target image.
It is then based on the priority orders and obtains the image of next highest priority as target image, execute the target image
It is input in text identification model trained in advance, thus the step of obtaining the content of text of the target image.Until obtaining
The corresponding content of text of each image in all images.Then the corresponding content of text of each image in all images is carried out whole
It closes, the integration can be understood as the corresponding content of text of each image being integrated into a content of text.
In a kind of specific implement scene, described multiple images can be different user and apply for the same business upload
, such as by taking certain project asset examination and approval business as an example, the project funds Shen batch needs user 1 (common office worker), 2 (department of user
Manager), user 3 (general manager) the project application book is carried out filling in upload, 3 users user gradation are as follows: user 3 > use
2 > user of family 1.For user terminal when sending service request to server, user is applied for described image when business by user terminal
And the user identifier is packaged, and is uploaded to service server in the form of data packet, service server receive it is multiple
When data packet, the multiple data packet is parsed, obtains the image of each data packet and the use in the multiple data packet
Family mark obtains the corresponding user gradation of the user identifier, business clothes according to the user identifier in user identifier set
Device be engaged in when identifying to each image, it can be according to the height of user gradation, successively on user 3, user 2, user 1
The described image of biography is identified.Then the correspondence content of text of the described image of 3 users is respectively obtained, and by text
Content is integrated, and the corresponding content of text of each image can be integrated into a content of text by the integration.
Step 204: based on the keyword and/or critical sentence extraction the first semantic feature letter in the content of text
Breath.
The semantic feature refers to the distinctive semantic attribute of the unstructured data of literal expression, is with a paper
Example, semantic feature includes the semantic features such as author's creation intention, Data subject explanation, low-level image feature meaning.The semantic feature
Information is can to express the semantic of object itself and various features semantic in the environment, by taking content of text as an example, institute's predicate
Adopted characteristic information can be the composition sequence of word, the emotion information of word, word/sentence mutual information etc.
Wherein, composition letter is that a word is made of which letter, the sequencing relationship of these letters.
Word order is the sequencing of expression in short each word of (meaning) composition.
The emotion information of word is word emotion meaning expressed in this sentence, and the emotion meaning is understood that
For word sentence be commendation or derogatory sense, be high or droning, be joyful or sad etc..
Mutual information refers to the statistical iteration relationship between some word or word and classification, and it is right that mutual information is commonly applied to measurement two
Reciprocity as between.
The first semantic feature information is in the present embodiment, it can be understood as the semantic feature of the content of text is believed
Breath, for the content of text, it may include but is not limited to keyword/sentence of text the first semantic feature information
Information, word/distributed intelligence of sentence frequency, the entity information of syntactic level, theme of semantic class etc..
Specifically, server identifies described image, after the content of text identified, extraction of semantics mould is utilized
Type is based on the keyword and/or critical sentence extraction the first semantic feature information in the content of text.
Optionally, the extraction of semantics model can be the text feature information extracting method based on Context Framework, i.e., first
It determines the extraction element (critical sentence, keyword, keyword, key symbol etc.) of content of text, then semantic analysis is incorporated and is united
Calculating method carries out extraction processing to the content of text, obtains the first semantic feature information of the content of text；It can be base
In ontological text feature, i.e., using ontology (On-tology) model using the content of text as input,
Export the first semantic feature information of the content of text；It can be the concept characteristic extracting method based on Hownet, i.e., based on general
The feature extracting method for reading feature, on the basis of vector space model (Vector Space Model, VSM), to the text
This content carries out semantic analysis, and the semantic information of vocabulary is obtained using the database of Hownet, and semantic identical vocabulary is mapped to
Identical concept, the word after then being clustered, and the characteristic item of the text vector as VSM model, then carry out model
Operation etc., etc..It should be noted that the mode of the first semantic feature information for extracting the content of text has very much, it can
To be above-mentioned one or more fittings, it is not construed as limiting herein.
Step 205: based on the first semantic feature information and the second semantic feature information, calculating in the text
Hold the semantic similarity with the received text.
Specifically, the received text refers to the spoken and written languages pre-set, the second of the received text is semantic
Characteristic information can be understood as the semantic feature information of the received text, the second semantic feature information in the present embodiment
For the received text, it may include but be not limited to the key word information of received text, word frequency distribution information, grammer
Entity information, the theme of semantic class etc. of grade.
Second semantic feature information of these received texts and the corresponding relationship of the service request are stored in received text
In library, can be understood as service server in the present embodiment can obtain the service request pair in the received text library
Second semantic feature information of the received text answered.
Specifically, service server obtains institute from the received text information bank according to the service identification of the service request
The second semantic feature information of the corresponding received text of service identification is stated, the service identification can be understood as the user and initiate
The identification informations such as the type of the business of business audit or examination & approval, the ID of distribution, business-level, can according to the service identification
To identify the business of the service request instruction, the first semantic feature information is matched with the second semantic feature information
Processing, to calculate the semantic similarity of the content of text Yu the received text.
Optional, the semantic similarity calculation method can be maximum weight matching algorithm based on bigraph (bipartite graph), can be base
The matching algorithm of distance between seeking vector, can be k recently another calculating method, can be bayes arithmetic etc..
In a kind of feasible embodiment, the semantic similarity calculation method be can be based on extraction of semantics model
Co sinus vector included angle matching, i.e., indicate the semantic similarity of the content of text Yu the received text with included angle cosine.
After through extraction of semantics model the first semantic feature information and the processing of the second semantic feature information vectorization, adopt
Similarity S is calculated with following formula:
Wherein, COS () is cosine function, and V is the first semantic feature information, and U is the second semantic feature information, and Vi indicates the
I-th of vector in one semantic feature information, Ui indicate i-th of vector in the first semantic feature information.
The semantic similarity of content of text Yu the received text can be calculated by above-mentioned formula.
Step 206: when the semantic similarity meets given threshold, it is determined that the business application passes through, outgoing traffic
Apply for successful prompt information.
Specifically, server is after the semantic similarity for obtaining the content of text and the received text, root
The semantic similarity is judged according to default judge rule, when the semantic similarity reaches given threshold, determines institute
The business application of stating passes through.
For example, the given threshold is 0.85, semantic similarity obtained by calculation is 0.89, and semantic similarity is greater than
Given threshold, it is determined that the business application passes through.
Specifically, can be mentioned to what the user terminal transmission service application passed through after the business application passes through
Show that information, the prompt information can be in the form of PUSH message, broadcasting video or animation etc..
Step 207: when the semantic similarity is unsatisfactory for given threshold, it is determined that the business application does not pass through, defeated
The prompt information of business application failure out.
Specifically, server is after the semantic similarity for obtaining the content of text and the received text, root
The semantic similarity is judged according to default judge rule, when the semantic similarity is not up to given threshold, is determined
The business application does not pass through.
For example, the given threshold is 0.85, semantic similarity obtained by calculation is 0.8, and semantic similarity is less than
Given threshold, it is determined that the business application does not pass through.
Specifically, when the business application does not pass through, it can be unsanctioned to the user terminal transmission service application
Prompt information, the prompt information can be in the form of PUSH message, broadcasting video or animation etc..
Optionally, user terminal can also provide a user again in the unsanctioned prompt information of transmission service application
The option of upload, user can click the option, upload the image comprising content of text again.
In the embodiment of the present application, service server acquisition includes the image of content of text, and described image is based on business
Request generates, and identifies to described image, obtains the content of text in described image, extracts the of the content of text
One semantic feature information, second based on the first semantic feature information received text corresponding with the service request is semantic
Characteristic information is analyzed and processed, and is based on analysis and processing result, is determined whether the business application passes through.By to service request
When image recognition grade extract to obtain corresponding first semantic feature information, it is semantic special based on the first semantic feature information and second
Whether the processing result of reference breath is passed through with the determination business application, it is ensured that the accuracy rate of business audit.It uses simultaneously
Service server audits described image, and the treatment effeciency of business audit can be improved.
Following is the application Installation practice, can be used for executing the application embodiment of the method.It is real for the application device
Undisclosed details in example is applied, the application embodiment of the method is please referred to.
Fig. 8 is referred to, it illustrates the structural representations for the image processing apparatus that one exemplary embodiment of the application provides
Figure.The image processing apparatus can by software, hardware or both be implemented in combination with as device all or part of.It should
Device 1 includes image collection module module 11, picture recognition module 12, characteristic extracting module 13, analysis and processing module 14 and Shen
It please determining module 15.
Image collection module 11, for obtain include content of text image, described image based on service request generate；
Picture recognition module 12 obtains the content of text in described image for identifying to described image；
Characteristic extracting module 13, for extracting the first semantic feature information of the content of text；
Analysis and processing module 14, for based on the first semantic feature information standard text corresponding with the service request
This second semantic feature information is analyzed and processed；
Apply for determining module 15, for being based on analysis and processing result, determines whether the business application passes through.
Optionally, the analysis and processing module 14, is specifically used for:
Based on the first semantic feature information and the second semantic feature information, calculate the content of text with it is described
The semantic similarity of received text.
Optionally, the application determining module 15, is specifically used for:
When the semantic similarity meets given threshold, it is determined that the business application passes through, otherwise, it determines not lead to
It crosses.
Optionally, the characteristic extracting module 13, is specifically used for:
Based on the keyword and/or critical sentence extraction the first semantic feature information in the content of text.
Optionally, as shown in figure 9, described image identification module 12, further includes:
Information acquisition unit 121, for obtaining corresponding to described multiple images when the quantity of described image is multiple
Auxiliary information；
Image identification unit 122 obtains the content of text in each image for identifying to each image；
Content integral unit 123, for being integrated the content of text in each image based on the auxiliary information.
Optionally, described image identification module 12, is specifically used for:
Region division is carried out to described image, multiple regions image is obtained, the multiple area image is known respectively
Not, the corresponding region content of text of each area image is obtained, the region content of text is combined, is obtained in described image
The content of text.
Optionally, as shown in Figure 10, described device 1, further includes:
Prompt information module 16, for when determining that the business application passes through, outgoing traffic application successfully to prompt to believe
Breath；And/or when determining that the business application does not pass through, the prompt information of outgoing traffic application failure.
It should be noted that image processing apparatus provided by the above embodiment is when executing image processing method, only more than
The division progress of each functional module is stated for example, can according to need and in practical application by above-mentioned function distribution by difference
Functional module complete, i.e., the internal structure of equipment is divided into different functional modules, with complete it is described above whole or
Person's partial function.In addition, image processing apparatus provided by the above embodiment and image processing method embodiment belong to same design,
It embodies realization process and is detailed in embodiment of the method, and which is not described herein again.
Above-mentioned the embodiment of the present application serial number is for illustration only, does not represent the advantages or disadvantages of the embodiments.
In the embodiment of the present application, service server acquisition includes the image of content of text, and described image is based on business
Request generates, and identifies to described image, obtains the content of text in described image, extracts the of the content of text
One semantic feature information, second based on the first semantic feature information received text corresponding with the service request is semantic
Characteristic information is analyzed and processed, and is based on analysis and processing result, is determined whether the business application passes through.By to service request
When image recognition grade extract to obtain corresponding first semantic feature information, it is semantic special based on the first semantic feature information and second
Whether the processing result of reference breath is passed through with the determination business application, it is ensured that the accuracy rate of business audit.It uses simultaneously
Service server audits described image, and the treatment effeciency of business audit can be improved.
The embodiment of the present application also provides a kind of computer storage medium, the computer storage medium can store more
Item instruction, described instruction are suitable for being loaded by processor and being executed the described image processing side such as above-mentioned Fig. 1-embodiment illustrated in fig. 7
Method, specific implementation procedure may refer to Fig. 1-embodiment illustrated in fig. 7 and illustrate, herein without repeating.
Present invention also provides a kind of computer program product, which is stored at least one instruction,
At least one instruction is loaded by the processor and is executed the described image processing side such as above-mentioned Fig. 1-embodiment illustrated in fig. 6
Method, specific implementation procedure may refer to Fig. 1-embodiment illustrated in fig. 6 and illustrate, herein without repeating.
Referring to Figure 11, a kind of structural schematic diagram of server is provided for the embodiment of the present application.As shown in figure 11, described
Server 1000 may include: at least one processor 1001, at least one network interface 1004, user interface 1003, storage
Device 1005, at least one communication bus 1002.
Wherein, communication bus 1002 is for realizing the connection communication between these components.
Wherein, user interface 1003 may include display screen (Display), camera (Camera), optional user interface
1003 can also include standard wireline interface and wireless interface.
Wherein, network interface 1004 optionally may include standard wireline interface and wireless interface (such as WI-FI interface).
Wherein, processor 1001 may include one or more processing core.Processor 1001 using it is various excuse and
Various pieces in the entire server 1000 of connection, by running or executing the instruction being stored in memory 1005, journey
Sequence, code set or instruction set, and call the data that are stored in memory 1005, the various functions of execute server 1000 and
Handle data.Optionally, processor 1001 can using Digital Signal Processing (Digital Signal Processing,
DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), programmable logic array
At least one of (Programmable Logic Array, PLA) example, in hardware is realized.Processor 1001 can integrating central
Processor (Central Processing Unit, CPU), image processor (Graphics Processing Unit, GPU)
With the combination of one or more of modem etc..Wherein, the main processing operation system of CPU, user interface and apply journey
Sequence etc.；GPU is used to be responsible for the rendering and drafting of content to be shown needed for display screen；Modem is for handling channel radio
Letter.It is understood that above-mentioned modem can not also be integrated into processor 1001, carried out separately through chip piece
It realizes.
Wherein, memory 1005 may include random access memory (Random Access Memory, RAM), also can wrap
Include read-only memory (Read-Only Memory).Optionally, which includes non-transient computer-readable medium
(non-transitory computer-readable storage medium).Memory 1005 can be used for store instruction, journey
Sequence, code, code set or instruction set.Memory 1005 may include storing program area and storage data area, wherein storing program area
Can store the instruction for realizing operating system, the instruction at least one function (such as touch function, sound play function
Energy, image player function etc.), for realizing instruction of above-mentioned each embodiment of the method etc.；Storage data area can store each above
The data etc. being related in a embodiment of the method.Memory 1005 optionally can also be that at least one is located remotely from aforementioned processing
The storage device of device 1001.As shown in figure 11, as may include in a kind of memory 1005 of computer storage medium operation
System, network communication module, Subscriber Interface Module SIM and image processing application program.
In the server 1000 shown in Figure 11, user interface 1003 is mainly used for providing the interface of input for user, obtains
Take the data of family input；And processor 1001 can be used for calling the image processing application program stored in memory 1005,
And specifically execute following operation:
Acquisition includes the image of content of text, and described image is generated based on service request；
Described image is identified, the content of text in described image is obtained；
Extract the first semantic feature information of the content of text；
The second semantic feature letter based on the first semantic feature information received text corresponding with the service request
Breath is analyzed and processed；
Based on analysis and processing result, determine whether the business application passes through.
In one embodiment, the processor 1001 execute it is described based on the first semantic feature information with it is described
Second semantic feature information of the corresponding received text of service request is analyzed and processed, specific to execute following operation:
Based on the first semantic feature information and the second semantic feature information, calculate the content of text with it is described
The semantic similarity of received text.
In one embodiment, the processor 1001 is described based on analysis and processing result in execution, determines the business
Whether application passes through, specific to execute following operation:
When the semantic similarity meets given threshold, it is determined that the business application passes through, otherwise, it determines not lead to
It crosses.
In one embodiment, the processor 1001 is executing first semantic feature for extracting the content of text
Information is specific to execute following operation:
Based on the keyword and/or critical sentence extraction the first semantic feature information in the content of text.
In one embodiment, the processor 1001 execute it is described described image is identified, obtain the figure
The content of text as in, specific to execute following operation:
When the quantity of described image is multiple, auxiliary information corresponding to described multiple images is obtained；
Each image is identified, the content of text in each image is obtained；
The content of text in each image is integrated based on the auxiliary information；
The auxiliary information includes at least one following: the uplink time of described image, the generation time of described image, institute
State the service priority of business corresponding to the user gradation of user corresponding to image, described image.
In one embodiment, the processor 1001 execute it is described described image is identified, obtain the figure
The content of text as in, specific to execute following operation:
Region division is carried out to described image, obtains multiple regions image；
The multiple area image is identified respectively, obtains the corresponding region content of text of each area image；
The region content of text is combined, the content of text in described image is obtained.
In one embodiment, the processor 1001 is executing described image processing method, also executes following operation:
When determining that the business application passes through, the successful prompt information of outgoing traffic application；And/or described in the determination
When business application does not pass through, the prompt information of outgoing traffic application failure.
In the present embodiment, service server acquisition includes the image of content of text, and described image is based on service request
It generates, described image is identified, the content of text in described image is obtained, extracts the first language of the content of text
Adopted characteristic information, the second semantic feature based on the first semantic feature information received text corresponding with the service request
Information is analyzed and processed, and is based on analysis and processing result, is determined whether the business application passes through.When by service request
Image recognition grade is extracted to obtain corresponding first semantic feature information, is believed based on the first semantic feature information and the second semantic feature
Whether the processing result of breath is passed through with the determination business application, it is ensured that the accuracy rate of business audit.Business is used simultaneously
Server audits described image, and the treatment effeciency of business audit can be improved.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory or random access memory etc..
Above disclosed is only the application preferred embodiment, cannot limit the right model of the application with this certainly
It encloses, therefore according to equivalent variations made by the claim of this application, still belongs to the range that the application is covered.
Claims (10)
1. a kind of image processing method, which is characterized in that the described method includes:
Acquisition includes the image of content of text, and described image is generated based on service request；
Described image is identified, the content of text in described image is obtained；
Extract the first semantic feature information of the content of text；
The second semantic feature information based on the first semantic feature information received text corresponding with the service request into
Row analysis processing；
Based on analysis and processing result, determine whether the business application passes through.
2. the method according to claim 1, wherein described be based on the first semantic feature information and the industry
Business requests the second semantic feature information of corresponding received text to be analyzed and processed, comprising: is based on first semantic feature
Information and the second semantic feature information, calculate the semantic similarity of the content of text Yu the received text.
3. according to the method described in claim 2, it is characterized in that, it is described be based on analysis and processing result, determine the business Shen
Please whether passing through, comprising: when the semantic similarity meets given threshold, it is determined that the business application passes through, otherwise, it determines
Not pass through.
4. method according to claim 1 to 3, which is characterized in that first semanteme for extracting the content of text
Characteristic information, comprising: based on the keyword and/or critical sentence extraction the first semantic feature information in the content of text.
5. method according to claim 1 to 3, which is characterized in that it is described that described image is identified, obtain institute
State the content of text in image, comprising:
When the quantity of described image is multiple, auxiliary information corresponding to described multiple images is obtained；
Each image is identified, the content of text in each image is obtained；
The content of text in each image is integrated based on the auxiliary information；
The auxiliary information includes at least one following: the uplink time of described image, the generation time of described image, the figure
The service priority of the business as corresponding to the user gradation of corresponding user, described image.
6. method according to claim 1 to 3, which is characterized in that it is described that described image is identified, obtain institute
State the content of text in image, comprising:
Region division is carried out to described image, obtains multiple regions image；
The multiple area image is identified respectively, obtains the corresponding region content of text of each area image；
The region content of text is combined, the content of text in described image is obtained.
7. the method according to claim 1, wherein further include:
When determining that the business application passes through, the successful prompt information of outgoing traffic application；And/or when determining the business
When application does not pass through, the prompt information of outgoing traffic application failure.
8. a kind of image processing apparatus, which is characterized in that the described method includes:
Image collection module, for obtain include content of text image, described image based on service request generate；
Picture recognition module obtains the content of text in described image for identifying to described image；
Characteristic extracting module, for extracting the first semantic feature information of the content of text；
Analysis and processing module, for based on the first semantic feature information received text corresponding with the service request
Two semantic feature information are analyzed and processed；
Apply for determining module, for being based on analysis and processing result, determines whether the business application passes through.
9. a kind of computer storage medium, which is characterized in that the computer storage medium is stored with a plurality of instruction, described instruction
Suitable for being loaded by processor and being executed the method and step such as claim 1~7 any one.
10. a kind of server characterized by comprising processor and memory；Wherein, the memory is stored with computer
Program, the computer program are suitable for being loaded by the processor and being executed the method step such as claim 1~7 any one
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910704473.2A CN110489747A (en) | 2019-07-31 | 2019-07-31 | A kind of image processing method, device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910704473.2A CN110489747A (en) | 2019-07-31 | 2019-07-31 | A kind of image processing method, device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110489747A true CN110489747A (en) | 2019-11-22 |
Family
ID=68547703
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910704473.2A Pending CN110489747A (en) | 2019-07-31 | 2019-07-31 | A kind of image processing method, device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110489747A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111353422A (en) * | 2020-02-27 | 2020-06-30 | 维沃移动通信有限公司 | Information extraction method and device and electronic equipment |
| CN111445545A (en) * | 2020-02-27 | 2020-07-24 | 北京大米未来科技有限公司 | Text-to-map method, device, storage medium and electronic equipment |
| CN113255836A (en) * | 2021-06-28 | 2021-08-13 | 北京乐学帮网络技术有限公司 | Job data processing method and device, computer equipment and storage medium |
| CN113849249A (en) * | 2020-06-28 | 2021-12-28 | Oppo（重庆）智能科技有限公司 | Text information display method and device, storage medium and electronic equipment |
| WO2022134588A1 (en) * | 2020-12-21 | 2022-06-30 | 深圳壹账通智能科技有限公司 | Method for constructing information review classification model, and information review method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107545210A (en) * | 2016-06-27 | 2018-01-05 | 北京新岸线网络技术有限公司 | A kind of method of video text extraction |
| CN108256591A (en) * | 2018-02-26 | 2018-07-06 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of output information |
| CN108595410A (en) * | 2018-03-19 | 2018-09-28 | 小船出海教育科技（北京）有限公司 | The automatic of hand-written composition corrects method and device |
| CN108596759A (en) * | 2018-05-09 | 2018-09-28 | 平安普惠企业管理有限公司 | loan application information detecting method and server |
| CN109919014A (en) * | 2019-01-28 | 2019-06-21 | 平安科技（深圳）有限公司 | OCR recognition methods and its electronic equipment |
| CN109993417A (en) * | 2019-03-12 | 2019-07-09 | 平安普惠企业管理有限公司 | A kind of service condition mark adding method, device and storage medium |
| CN109993040A (en) * | 2018-01-03 | 2019-07-09 | 北京世纪好未来教育科技有限公司 | Text recognition method and device |
-
2019
- 2019-07-31 CN CN201910704473.2A patent/CN110489747A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107545210A (en) * | 2016-06-27 | 2018-01-05 | 北京新岸线网络技术有限公司 | A kind of method of video text extraction |
| CN109993040A (en) * | 2018-01-03 | 2019-07-09 | 北京世纪好未来教育科技有限公司 | Text recognition method and device |
| CN108256591A (en) * | 2018-02-26 | 2018-07-06 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of output information |
| CN108595410A (en) * | 2018-03-19 | 2018-09-28 | 小船出海教育科技（北京）有限公司 | The automatic of hand-written composition corrects method and device |
| CN108596759A (en) * | 2018-05-09 | 2018-09-28 | 平安普惠企业管理有限公司 | loan application information detecting method and server |
| CN109919014A (en) * | 2019-01-28 | 2019-06-21 | 平安科技（深圳）有限公司 | OCR recognition methods and its electronic equipment |
| CN109993417A (en) * | 2019-03-12 | 2019-07-09 | 平安普惠企业管理有限公司 | A kind of service condition mark adding method, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 韩红旗, 北京：科学技术文献出版社 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111353422A (en) * | 2020-02-27 | 2020-06-30 | 维沃移动通信有限公司 | Information extraction method and device and electronic equipment |
| CN111445545A (en) * | 2020-02-27 | 2020-07-24 | 北京大米未来科技有限公司 | Text-to-map method, device, storage medium and electronic equipment |
| CN111445545B (en) * | 2020-02-27 | 2023-08-18 | 北京大米未来科技有限公司 | Text transfer mapping method and device, storage medium and electronic equipment |
| CN111353422B (en) * | 2020-02-27 | 2023-08-22 | 维沃移动通信有限公司 | Information extraction method and device and electronic equipment |
| CN113849249A (en) * | 2020-06-28 | 2021-12-28 | Oppo（重庆）智能科技有限公司 | Text information display method and device, storage medium and electronic equipment |
| WO2022134588A1 (en) * | 2020-12-21 | 2022-06-30 | 深圳壹账通智能科技有限公司 | Method for constructing information review classification model, and information review method |
| CN113255836A (en) * | 2021-06-28 | 2021-08-13 | 北京乐学帮网络技术有限公司 | Job data processing method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7122341B2 (en) | Method and apparatus for evaluating translation quality | |
| CN109241524B (en) | Semantic analysis method and device, computer-readable storage medium and electronic equipment | |
| CN107391760B (en) | User interest recognition methods, device and computer readable storage medium | |
| CN110489747A (en) | A kind of image processing method, device, storage medium and electronic equipment | |
| CN108536679A (en) | Name entity recognition method, device, equipment and computer readable storage medium | |
| CN110162593A (en) | A kind of processing of search result, similarity model training method and device | |
| US20220284327A1 (en) | Resource pushing method and apparatus, device, and storage medium | |
| CN107704495A (en) | Training method, device and the computer-readable recording medium of subject classification device | |
| CN109271493A (en) | A kind of language text processing method, device and storage medium | |
| CN110377900A (en) | Checking method, device, computer equipment and the storage medium of Web content publication | |
| CN110781668B (en) | Text information type identification method and device | |
| CN110674255B (en) | Text content auditing method and device | |
| US20180124437A1 (en) | System and method for video data collection | |
| WO2021218028A1 (en) | Artificial intelligence-based interview content refining method, apparatus and device, and medium | |
| CN109992781B (en) | Text feature processing method and device and storage medium | |
| CN112417158A (en) | Training method, classification method, device and equipment of text data classification model | |
| CN110309114A (en) | Processing method, device, storage medium and the electronic device of media information | |
| CN110287341A (en) | A kind of data processing method, device and readable storage medium storing program for executing | |
| CN108345612A (en) | A kind of question processing method and device, a kind of device for issue handling | |
| CN115130711A (en) | Data processing method and device, computer and readable storage medium | |
| CN111192170B (en) | Question pushing method, device, equipment and computer readable storage medium | |
| CN114065720A (en) | Conference summary generation method and device, storage medium and electronic equipment | |
| CN114528851B (en) | Reply sentence determination method, reply sentence determination device, electronic equipment and storage medium | |
| CN115934891A (en) | Question understanding method and device | |
| CN111414609B (en) | Object verification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191122 |