CN107145859A - E-book conversion process method, device and computer-readable recording medium - Google Patents
E-book conversion process method, device and computer-readable recording medium Download PDFInfo
- Publication number
- CN107145859A CN107145859A CN201710309205.1A CN201710309205A CN107145859A CN 107145859 A CN107145859 A CN 107145859A CN 201710309205 A CN201710309205 A CN 201710309205A CN 107145859 A CN107145859 A CN 107145859A
- Authority
- CN
- China
- Prior art keywords
- text
- information
- book
- structural information
- electronic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/413—Classification of content, e.g. text, photographs or tables
Abstract
The disclosure is directed to a kind of e-book conversion process method, device and computer-readable recording medium, methods described includes：Obtain the target image of target paper book；Image recognition is carried out to the target image, electronic text information is obtained；By analyzing the electronic text information, the text structural information of the electronic text information is determined；According to the electronic text information and the text structural information, e-book of the generation for the target paper book.The disclosure is recognized by text structural information, obtains the text structural information of target paper book so that when target paper book is converted into e-book, will not lose its text structural information.
Description
Technical field
This disclosure relates to electronic technology field, more particularly to a kind of e-book conversion process method, device and computer can
Read storage medium.
Background technology
Character recognition technology is continued to develop in recent years, has obtained increasingly being widely applied.
For paper book, by the mode such as take pictures, photograph, scan, paper book can be converted to picture；Then text is passed through again
Word is recognized, the word in picture is identified, so that paper book is converted into e-book.
In correlation technique, when only needing to change the partial content (for example, content of tape label) of paper book into e-book, obtain
The e-book arrived is contents fragment, and chapter, paragraph belonging to contents fragment, which such as fall at the layer of structure information, to be lost.
The content of the invention
To overcome problem present in correlation technique, the disclosure provides a kind of e-book conversion process method, device and meter
Calculation machine readable storage medium storing program for executing.
According to the first aspect of the embodiment of the present disclosure there is provided a kind of e-book conversion process method, including：Obtain target basis
The target image of matter book；
Image recognition is carried out to the target image, electronic text information is obtained；
By analyzing the electronic text information, the text structural information of the electronic text information is determined；
According to the electronic text information and the text structural information, electronics of the generation for the target paper book
Book.
Preferably, it is described by analyzing the electronic text information, determine the text structure letter of the electronic text information
Breath, including：
The corresponding text formatting of the electronic text information is obtained, wherein, the text formatting includes font, font size, OK
Away from, whether one or more of overstriking and paragraph format；And
According to pre-set text textural classification model and the text formatting, the text structure of the electronic text information is determined
Information.
Preferably, institute is determined according to pre-set text textural classification model and the text formatting of the word identified described
Before the text structural information for stating electronic text information, methods described also includes：
Training sample set is obtained, the training sample, which is concentrated, to be included：The view data and the figure of the target paper book
Text structural information as corresponding to data；
It is trained using training sample set pair convolutional neural networks, obtains each layer parameter information of convolutional neural networks；
The pre-set text textural classification model according to each layer parameter information architecture.
Preferably, methods described also includes：
Obtain the notes part in the electronic text information；
By the text structural information associated storage corresponding to the notes part and the notes part；And
When meeting preparatory condition, text structure letter of the notes partly and corresponding to the notes part is shown
Breath.
According to the second aspect of the embodiment of the present disclosure there is provided a kind of e-book conversion processing unit, including：
Target image acquisition module, is configured as obtaining the target image of target paper book；
Electronic text information acquisition module, is configured as carrying out image recognition to the target image, obtains e-text
Information；
Structural information determining module, is configured as, by analyzing the electronic text information, determining the e-text letter
The text structural information of breath；
E-book generation module, is configured as, according to the electronic text information and the text structural information, generating pin
To the e-book of the target paper book.
Preferably, the structural information determining module includes：
Text formatting acquisition submodule, is configured as obtaining the corresponding text formatting of the electronic text information, wherein, institute
State text formatting including font, font size, line-spacing, whether one or more of overstriking and paragraph format；And
Structural information determination sub-module, is configured as according to pre-set text textural classification model and the text formatting, really
The text structural information of the fixed electronic text information.
Preferably, described device also includes：
Training sample set acquisition module, is configured as obtaining training sample set, the training sample, which is concentrated, to be included：The mesh
Mark the text structural information corresponding to the view data and described image data of paper book；
Parameter information acquisition module, is configured to, with training sample set pair convolutional neural networks and is trained, rolled up
Each layer parameter information of product neutral net；
Default disaggregated model builds module, is configured as the pre-set text structure according to each layer parameter information architecture
Disaggregated model.
Preferably, described device also includes：
Acquisition module is taken down notes, is configured as obtaining the notes part in the electronic text information；
Memory module, is configured as closing the text structural information corresponding to the notes part and the notes part
Connection storage；And
Display module, is configured as when meeting preparatory condition, shows the notes part and the notes part institute
Corresponding text structural information.
According to the third aspect of the embodiment of the present disclosure there is provided a kind of e-book conversion processing unit, including：
Processor；Memory for storing processor-executable instruction；
Wherein, the processor is configured as：Obtain the target image of target paper book；The target image is schemed
As identification, electronic text information is obtained；By analyzing the electronic text information, the text knot of the electronic text information is determined
Structure information；According to the electronic text information and the text structural information, e-book of the generation for the target paper book.
According to the fourth aspect of the embodiment of the present disclosure there is provided a kind of computer-readable recording medium, calculating is stored thereon with
Machine programmed instruction, the programmed instruction realizes the e-book conversion process side that disclosure first aspect is provided when being executed by processor
The step of method.
The technical scheme provided by this disclosed embodiment can include the following benefits：Be converted to by target paper book
During e-book, text structural information can be retained so that when paper book is converted into e-book, its layer of structure letter will not be lost
Breath；And without manually being changed.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary and explanatory, not
The disclosure can be limited.
Brief description of the drawings
Accompanying drawing herein is merged in specification and constitutes the part of this specification, shows the implementation for meeting the disclosure
Example, and be used to together with specification to explain the principle of the disclosure.
Fig. 1 is a kind of flow chart of e-book conversion process method according to an exemplary embodiment.
Fig. 2 is the schematic flow sheet of acquisition text structural information in the embodiment of the disclosure one.
During Fig. 3 is an embodiment of the disclosure, the handling process when the word in target image belongs to body text
Schematic diagram.
Fig. 4 is that the embodiment of the disclosure one is trained the schematic flow sheet for obtaining pre-set text textural classification model.
Fig. 5 is the flow signal that the embodiment of the disclosure one is tested the pre-set text textural classification model trained
Figure.
Fig. 6 is the schematic flow sheet for carrying out e-book conversion in the embodiment of the disclosure one to notes part.
Fig. 7 is being shown to the document structure information corresponding to notes part and notes part for the embodiment of the disclosure one
Effect diagram.
Fig. 8 is a kind of block diagram of e-book conversion processing unit according to an exemplary embodiment.
Fig. 9 is a kind of block diagram of device for e-book conversion process method according to an exemplary embodiment.
Embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the disclosure.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects be described in detail in claims, the disclosure.
Fig. 1 is a kind of flow chart of e-book conversion process method according to an exemplary embodiment, such as Fig. 1 institutes
Show, the e-book conversion process method is used in terminal, comprises the following steps：
In step s 11, the target image of target paper book is obtained.Target image may include text image and notes figure
Picture.
When user reads paper book, it can be taken notes in paper book by line, figure labeling, hand-written notes etc..When need
When some parts (for example, the word of tape label, word, sentence, paragraph or chapters and sections etc.) of paper book are converted into e-book, pass through
Image collecting device obtains target image to carrying out IMAQ comprising target paper book.
In embodiment of the disclosure, image collecting device can be camera, scanner, mobile phone etc..Target paper book is entered
Row IMAQ, i.e., shot or scanned to target paper book by image collecting device, obtain target image.
In step s 12, image recognition is carried out to target image, obtains electronic text information.
By carrying out the word in image recognition, recognition target image to target image, electronic text information, electronics are obtained
Text message at least includes the text formatting of word and word.Text formatting include font, font size, line-spacing, whether overstriking and section
Fall one or more of form.
, can be by being pre-processed to target image (for example, carrying out ash to target image in an embodiment of the disclosure
Degreeization, noise reduction, binaryzation, character cutting and normalization etc.), and feature extraction is carried out to pretreated target image
After dimension-reduction treatment, Text region and text formatting identification are carried out.
In one embodiment, in order to improve the accuracy rate of Text region, obtain after Text region result, then word is known
Other result is optimized, to be corrected to Text region result.Text region result is corrected can be by language model
To be corrected, to correct the word of identification mistake.
In step s 13, by analyzing electronic text information, the text structural information of electronic text information is determined.
Text structural information at least includes one or more of following information：Affiliated part, affiliated chapter, affiliated section and institute
Belong to page.
Referring to Fig. 2, in an embodiment of the disclosure, text structural information is obtained in the following manner：
In the step s 21, the corresponding text formatting of electronic text information is obtained.
In step S22, according to pre-set text textural classification model and text formatting, the text of electronic text information is determined
Structural information.
In embodiment of the disclosure, pre-set text textural classification model is obtained by being trained to training sample set,
Thus, text structural information can be identified.
It should be understood that according to the difference of paper book typesetting format, text structural information can be identified in a different manner.Example
Such as, for some typesetting formats, affiliated part, affiliated chapter and affiliated page can be identified from the information such as header, footer；And institute
Category section can be identified from body part according to text formatting.By being trained to pre-set text textural classification model, it can be achieved
Accurately text structural information is identified.The acquisition of pre-set text textural classification model will subsequently describe in detail.
In step S14, according to electronic text information and text structural information, electronics of the generation for target paper book
Book.
Thus, the e-book conversion process method of the embodiment of the present disclosure, can when target paper book is converted into e-book
Retain text structural information so that when paper book is converted into e-book, its layer of structure information will not be lost；And without carrying out
Artificial conversion.
Referring to Fig. 3, in an embodiment of the disclosure, when the word in target image belongs to body text, do not include
When can react the word of text structural information：
In step S31, the prevpage paper book image of page where collection target image.
In step s 32, the text structural information of the paper book image of collection is identified.
In step S33, if recognize text structural information, this page of last text structural information is regard as target figure
The text structural information of picture；If unidentified arrive text structural information, continue to gather the paper book image of page before target image,
Until recognizing text structural information.
Thus, the acquisition of the affiliated text structural information of target image word can be quickly realized, and ensures recognition success rate.
It is that the embodiment of the disclosure one is trained the schematic flow sheet for obtaining pre-set text textural classification model referring to Fig. 4.
In step S41, training sample set is obtained.
In embodiment of the disclosure, acquisition paper book is shot or scanned to paper book by image collecting device
View data.It should be understood that the accuracy rate in order to improve identification, the paper book that training sample is used can be with target paper book phase
Together；In addition, can also be obtained from the different editions of the paper book, same category of paper book or same paper book of same publishing house
Training sample.
In one embodiment, get after the view data of paper book, image is pre-processed, Text region and excellent
Change.The image that pre-processing to collect carries out gray processing, noise reduction, binaryzation, character cutting and normalization etc..
Thus, training sample set can be obtained, training sample set includes multiple vectors to (Y, P), wherein, Y is paper book
Picture view data after pretreatment, P is the text structural information corresponding to view data.It should be understood that view data institute is right
The text structural information answered can be preset.
In step S42, it is trained using training sample set pair convolutional neural networks, obtains each of convolutional neural networks
Layer parameter information.
When being trained to convolutional neural networks, in propagation stage forward, the paper for the vectorial centering that training sample is concentrated
The view data Y input networks of matter book, calculate corresponding reality output.
The back-propagation stage trained in convolutional neural networks, according to reality output and corresponding preferable output P, adjustment volume
Lamination, the weights of pond layer and full articulamentum and biasing so that reality output and the deviation of preferable output are minimum.
Each layer parameter information includes convolutional layer, the weights of pond layer and full articulamentum and the biasing of convolutional neural networks.
In step S43, according to each layer parameter information architecture pre-set text textural classification model.
In order to ensure the classifying quality of pre-set text textural classification model, referring to Fig. 5 be the embodiment of the disclosure one to training
The schematic flow sheet that good convolutional neural networks are tested.
In step s 51, test sample collection is obtained, test sample, which is concentrated, includes the view data of paper book to be identified.
The view data of paper book to be identified can gather for image collecting device 100, and be carried out according to above-mentioned pretreatment
View data after processing.It should be understood that paper book to be identified here can be target paper book.
In step S52, the view data input for the paper book to be identified that test sample is concentrated is believed by each layer parameter
Cease in the convolutional neural networks built, recognize the text structural information of paper book to be identified.
In step S53, when the text structural information of the paper book to be identified got is unsatisfactory for preparatory condition, according to
Convolutional neural networks are re-started training by training sample set, to update each layer parameter information.
Preparatory condition can be：The text structural information of the paper book to be identified identified and actual text structural information phase
Consistent or error is minimized.
Referring to Fig. 6, in an embodiment of the disclosure, rule for user in paper book, figure labeling (example
Such as, draw a circle, change bracket etc.), it is hand-written notes etc. carry out notes generation notes part, by following steps can will notes part with
And the text structural information corresponding to notes part is converted to e-book：
In step S61, the notes part in electronic text information is obtained.
In step S62, the text structural information associated storage partly and corresponding to notes part will be taken down notes.Thus,
The e-book with text structural information of part can be obtained taking down notes.
In step S63, when meeting preparatory condition, the text structure corresponding to display notes part and notes part
Information.Preparatory condition can be shown for user's selection notes part, or other conditions.
In an embodiment of the disclosure, due to the text structure corresponding to the electronic text information corresponding to target image
Information can be determined according to above-described embodiment, and thus, notes part is entered with the electronic text information of text structural information is determined
Row matching, you can obtain taking down notes the text structural information of part.For example, the electronics text that part will be taken down notes with text structure is determined
This information is compared, if the word of continuous predetermined number (for example, 20 or more than 20) is all identical, matching result is matching.
When matching result is matching, the text structural information for taking down notes part and the e-text letter that text structure is determined
Manner of breathing is same.
The e-book conversion process method of the disclosure embodiment, can obtain the text structural information of notes part so that
When notes are partially converted into e-book, its text structural information will not be lost；On the other hand, can be according to the text for taking down notes part
Structural information is shown, both facilitates user to consult, and can remove unwanted extra text information again.
In an embodiment of the disclosure, by the text structural information corresponding to electronic text information and electronic text information
Associated storage, obtains text index information.Text structural information includes：Body structure information and notes structural information, accordingly
Text index information includes text index information and notes index information.Wherein, text index information is used to be based on body structure
Body part in Information locating e-book, notes index information is used for based on the notes in notes structural information positioning electronic book
Part.It should be understood that text index information and notes index information can also synthesize a general index.
Referring to Fig. 7, in an embodiment of the disclosure, according to the method for above-described embodiment, can obtain notes part and
The corresponding document structure information in part is taken down notes, and the text structural information taken down notes part and taken down notes corresponding to part is carried out
Storage.When user thinks note reading part, display notes part, and the corresponding document structure information in notes part.
The e-book conversion process method of the embodiment of the present disclosure, by machine learning, in the word of identification target paper book
While, the text structural information of word is also recognized, word and text structural information are mapped.In one embodiment,
After the text structural information of word and word is identified, text editing is carried out, is obtained comprising the text belonging to word and word
The text of structural information.
In further embodiments, after the text structural information for identifying word and word, i.e., word is inserted belonging to it
Text structure in, obtain the e-book comprising word and text structural information.
Fig. 8 is a kind of block diagram of e-book conversion processing unit according to an exemplary embodiment.The device 800 is wrapped
Include：
Target image acquisition module 801, is configured as obtaining the target image of target paper book；
Electronic text information acquisition module 802, is configured as carrying out image recognition to the target image, obtains electronics text
This information；
Structural information determining module 803, is configured as, by analyzing the electronic text information, determining the e-text
The text structural information of information；
E-book generation module 804, is configured as according to the electronic text information and the text structural information, generation
For the e-book of the target paper book.
In one embodiment, structural information determining module 803 includes：
Text formatting acquisition submodule, is configured as obtaining the corresponding text formatting of the electronic text information, wherein, institute
State text formatting including font, font size, line-spacing, whether one or more of overstriking and paragraph format；And
Structural information determination sub-module, is configured as according to pre-set text textural classification model and the text formatting, really
The text structural information of the fixed electronic text information.
In one embodiment, device 800 also includes：
Training sample set acquisition module, is configured as obtaining training sample set, the training sample, which is concentrated, to be included：The mesh
Mark the text structural information corresponding to the view data and described image data of paper book；
Parameter information acquisition module, is configured to, with training sample set pair convolutional neural networks and is trained, rolled up
Each layer parameter information of product neutral net；
Default disaggregated model builds module, is configured as the pre-set text structure according to each layer parameter information architecture
Disaggregated model.
In one embodiment, device 800 also includes：
Acquisition module is taken down notes, is configured as obtaining the notes part in the electronic text information；
Memory module, is configured as closing the text structural information corresponding to the notes part and the notes part
Connection storage；And
Display module, is configured as when meeting preparatory condition, shows the notes part and the notes part institute
Corresponding text structural information.
On the device in above-described embodiment, wherein modules perform the concrete mode of operation in relevant this method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
The disclosure also provides a kind of computer-readable recording medium, is stored thereon with computer program instructions, and the program refers to
The e-book conversion process method that the disclosure is provided is realized in order when being executed by processor the step of.
Fig. 9 is a kind of frame of device 900 for e-book conversion process method according to an exemplary embodiment
Figure.For example, device 900 can be mobile phone, computer, digital broadcast terminal, messaging devices, game console is put down
Board device, Medical Devices, body-building equipment, personal digital assistant etc..
Reference picture 9, device 900 can include following one or more assemblies：Processing assembly 902, memory 904, electric power
Component 906, multimedia groupware 907, audio-frequency assembly 910, the interface 912 of input/output (I/O), sensor cluster 914, and
Communication component 916.
The integrated operation of the usual control device 900 of processing assembly 902, such as with display, call, data communication, phase
Machine operates the operation associated with record operation.Processing assembly 902 can refer to including one or more processors 920 to perform
Order, to complete all or part of step of above-mentioned e-book conversion process method.In addition, processing assembly 902 can include one
Individual or multiple modules, are easy to the interaction between processing assembly 902 and other assemblies.For example, processing assembly 902 can include many matchmakers
Module, to facilitate the interaction between multimedia groupware 907 and processing assembly 902.
Memory 904 is configured as storing various types of data supporting the operation in device 900.These data are shown
Example includes the instruction of any application program or method for being operated on device 900, and contact data, telephone book data disappears
Breath, picture, video etc..Memory 904 can be by any kind of volatibility or non-volatile memory device or their group
Close and realize, such as static RAM (SRAM), Electrically Erasable Read Only Memory (EEPROM) is erasable to compile
Journey read-only storage (EPROM), programmable read only memory (PROM), read-only storage (ROM), magnetic memory, flash
Device, disk or CD.
Electric power assembly 906 provides electric power for the various assemblies of device 900.Electric power assembly 906 can include power management system
System, one or more power supplys, and other components associated with generating, managing and distributing electric power for device 900.
Multimedia groupware 907 is included in the screen of one output interface of offer between described device 900 and user.One
In a little embodiments, screen can include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Curtain may be implemented as touch-screen, to receive the input signal from user.Touch panel includes one or more touch sensings
Device is with the gesture on sensing touch, slip and touch panel.The touch sensor can not only sensing touch or sliding action
Border, but also detection touches or slide related duration and pressure with described.In certain embodiments, many matchmakers
Body component 907 includes a front camera and/or rear camera.When device 900 be in operator scheme, such as screening-mode or
During video mode, front camera and/or rear camera can receive the multi-medium data of outside.Each front camera and
Rear camera can be a fixed optical lens system or with focusing and optical zoom capabilities.
Audio-frequency assembly 910 is configured as output and/or input audio signal.For example, audio-frequency assembly 910 includes a Mike
Wind (MIC), when device 900 be in operator scheme, when such as call model, logging mode and speech recognition mode, microphone by with
It is set to reception external audio signal.The audio signal received can be further stored in memory 904 or via communication set
Part 916 is sent.In certain embodiments, audio-frequency assembly 910 also includes a loudspeaker, for exports audio signal.
I/O interfaces 912 is provide interface between processing assembly 902 and peripheral interface module, above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include but be not limited to：Home button, volume button, start button and lock
Determine button.
Sensor cluster 914 includes one or more sensors, and the state for providing various aspects for device 900 is commented
Estimate.For example, sensor cluster 914 can detect opening/closed mode of device 900, the relative positioning of component is for example described
Component is the display and keypad of device 900, and sensor cluster 914 can be with 900 1 components of detection means 900 or device
Position change, the existence or non-existence that user contacts with device 900, the orientation of device 900 or acceleration/deceleration and device 900
Temperature change.Sensor cluster 914 can include proximity transducer, be configured to detect in not any physical contact
The presence of neighbouring object.Sensor cluster 914 can also include optical sensor, such as CMOS or ccd image sensor, for into
As being used in application.In certain embodiments, the sensor cluster 914 can also include acceleration transducer, gyro sensors
Device, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 916 is configured to facilitate the communication of wired or wireless way between device 900 and other equipment.Device
900 can access the wireless network based on communication standard, such as WiFi, 2G or 3G, or combinations thereof.In an exemplary implementation
In example, communication component 916 receives broadcast singal or broadcast related information from external broadcasting management system via broadcast channel.
In one exemplary embodiment, the communication component 916 also includes near-field communication (NFC) module, to promote junction service.Example
Such as, NFC module can be based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) technology,
Bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, device 900 can be believed by one or more application specific integrated circuits (ASIC), numeral
Number processor (DSP), digital signal processing appts (DSPD), PLD (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for performing above-mentioned e-book conversion process
Method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instructing, example are additionally provided
Such as include the memory 904 of instruction, above-mentioned instruction can be performed to complete above-mentioned e-book conversion by the processor 920 of device 900
Processing method.For example, the non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-
ROM, tape, floppy disk and optical data storage devices etc..
Those skilled in the art will readily occur to other embodiment party of the disclosure after considering specification and putting into practice the disclosure
Case.The application is intended to any modification, purposes or the adaptations of the disclosure, these modifications, purposes or adaptability
Change follows the general principle of the disclosure and including the undocumented common knowledge or usual skill in the art of the disclosure
Art means.Description and embodiments are considered only as exemplary, and the true scope of the disclosure and spirit are by following claim
Point out.
It should be appreciated that the precision architecture that the disclosure is not limited to be described above and is shown in the drawings, and
And various modifications and changes can be being carried out without departing from the scope.The scope of the present disclosure is only limited by appended claim.
Claims (10)
1. a kind of e-book conversion process method, it is characterised in that including：
Obtain the target image of target paper book；
Image recognition is carried out to the target image, electronic text information is obtained；
By analyzing the electronic text information, the text structural information of the electronic text information is determined；
According to the electronic text information and the text structural information, e-book of the generation for the target paper book.
2. according to the method described in claim 1, it is characterised in that described by analyzing the electronic text information, determine institute
The text structural information of electronic text information is stated, including：
The corresponding text formatting of the electronic text information is obtained, wherein, the text formatting includes font, font size, line-spacing, is
One or more of no overstriking and paragraph format；And
According to pre-set text textural classification model and the text formatting, the text structure letter of the electronic text information is determined
Breath.
3. method according to claim 2, it is characterised in that described according to pre-set text textural classification model and identification
Before the text formatting of the word gone out, the text structural information for determining the electronic text information, methods described also includes：
Training sample set is obtained, the training sample, which is concentrated, to be included：The view data and described image number of the target paper book
According to corresponding text structural information；
It is trained using training sample set pair convolutional neural networks, obtains each layer parameter information of convolutional neural networks；
The pre-set text textural classification model according to each layer parameter information architecture.
4. method according to claim 2, it is characterised in that methods described also includes：
Obtain the notes part in the electronic text information；
By the text structural information associated storage corresponding to the notes part and the notes part；And
When meeting preparatory condition, the text structural information corresponding to the display notes part and the notes part.
5. a kind of e-book conversion processing unit, it is characterised in that including：
Target image acquisition module, is configured as obtaining the target image of target paper book；
Electronic text information acquisition module, is configured as carrying out image recognition to the target image, obtains electronic text information；
Structural information determining module, is configured as, by analyzing the electronic text information, determining the electronic text information
Text structural information；
E-book generation module, is configured as according to the electronic text information and the text structural information, generation is directed to institute
State the e-book of target paper book.
6. device according to claim 5, it is characterised in that the structural information determining module includes：
Text formatting acquisition submodule, is configured as obtaining the corresponding text formatting of the electronic text information, wherein, the text
This form include font, font size, line-spacing, whether one or more of overstriking and paragraph format；And
Structural information determination sub-module, is configured as, according to pre-set text textural classification model and the text formatting, determining institute
State the text structural information of electronic text information.
7. device according to claim 5, it is characterised in that described device also includes：
Training sample set acquisition module, is configured as obtaining training sample set, the training sample, which is concentrated, to be included：The target basis
Text structural information corresponding to the view data and described image data of matter book；
Parameter information acquisition module, is configured to, with training sample set pair convolutional neural networks and is trained, and obtains convolution god
Each layer parameter information through network；
Default disaggregated model builds module, is configured as the pre-set text textural classification according to each layer parameter information architecture
Model.
8. device according to claim 5, it is characterised in that described device also includes：
Acquisition module is taken down notes, is configured as obtaining the notes part in the electronic text information；
Memory module, is configured as depositing the text structural information association corresponding to the notes part and the notes part
Storage；And
Display module, is configured as when meeting preparatory condition, shows that the notes part and the notes part are corresponding
Text structural information.
9. a kind of e-book conversion processing unit, it is characterised in that including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：Obtain the target image of target paper book；Image knowledge is carried out to the target image
Not, electronic text information is obtained；By analyzing the electronic text information, the text structure letter of the electronic text information is determined
Breath；According to the electronic text information and the text structural information, e-book of the generation for the target paper book.
10. a kind of computer-readable recording medium, is stored thereon with computer program instructions, it is characterised in that the programmed instruction
The step of claim 1 methods described is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710309205.1A CN107145859A (en) | 2017-05-04 | 2017-05-04 | E-book conversion process method, device and computer-readable recording medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710309205.1A CN107145859A (en) | 2017-05-04 | 2017-05-04 | E-book conversion process method, device and computer-readable recording medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107145859A true CN107145859A (en) | 2017-09-08 |
Family
ID=59775375
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710309205.1A Pending CN107145859A (en) | 2017-05-04 | 2017-05-04 | E-book conversion process method, device and computer-readable recording medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107145859A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108877764A (en) * | 2018-06-28 | 2018-11-23 | 掌阅科技股份有限公司 | Audio synthetic method, electronic equipment and the computer storage medium of talking e-book |
| CN109376658A (en) * | 2018-10-26 | 2019-02-22 | 信雅达系统工程股份有限公司 | A kind of OCR method based on deep learning |
| CN109522437A (en) * | 2018-11-30 | 2019-03-26 | 珠海格力电器股份有限公司 | A kind of information search method of paper document, device, storage medium and terminal |
| CN110399953A (en) * | 2019-07-29 | 2019-11-01 | 爱布客（武汉）科技有限公司 | A kind of method of paper book and e-book synchronous read |
| CN111290688A (en) * | 2018-12-06 | 2020-06-16 | 中兴通讯股份有限公司 | Multimedia note taking method, terminal and computer readable storage medium |
| CN111493829A (en) * | 2020-04-23 | 2020-08-07 | 四川大学华西医院 | Method, system and equipment for determining mild cognitive impairment recognition parameters |
| CN111556251A (en) * | 2020-05-20 | 2020-08-18 | 深圳前海微众银行股份有限公司 | Electronic book generation method, device and medium |
| CN111753814A (en) * | 2019-03-26 | 2020-10-09 | 杭州海康威视数字技术股份有限公司 | Sample generation method, device and equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101782896A (en) * | 2009-01-21 | 2010-07-21 | 汉王科技股份有限公司 | PDF character extraction method combined with OCR technology |
| CN102782702A (en) * | 2010-03-10 | 2012-11-14 | 微软公司 | Paragraph recognition in an optical character recognition (OCR) process |
| CN104376317A (en) * | 2013-08-12 | 2015-02-25 | 福建福昕软件开发股份有限公司北京分公司 | Method for transforming paper file into electronic file |
| CN105512197A (en) * | 2015-11-27 | 2016-04-20 | 广州宝钢南方贸易有限公司 | Digitized archiving device of documents and archiving and searching device thereof |
| CN105787028A (en) * | 2016-02-24 | 2016-07-20 | 北京橙鑫数据科技有限公司 | Business card proofreading method and system |
| CN106250830A (en) * | 2016-07-22 | 2016-12-21 | 浙江大学 | Digital book structured analysis processing method |
| CN106570105A (en) * | 2016-11-01 | 2017-04-19 | 广西电网有限责任公司电力科学研究院 | Structured processing method for power equipment test report |
-
2017
- 2017-05-04 CN CN201710309205.1A patent/CN107145859A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101782896A (en) * | 2009-01-21 | 2010-07-21 | 汉王科技股份有限公司 | PDF character extraction method combined with OCR technology |
| CN102782702A (en) * | 2010-03-10 | 2012-11-14 | 微软公司 | Paragraph recognition in an optical character recognition (OCR) process |
| CN104376317A (en) * | 2013-08-12 | 2015-02-25 | 福建福昕软件开发股份有限公司北京分公司 | Method for transforming paper file into electronic file |
| CN105512197A (en) * | 2015-11-27 | 2016-04-20 | 广州宝钢南方贸易有限公司 | Digitized archiving device of documents and archiving and searching device thereof |
| CN105787028A (en) * | 2016-02-24 | 2016-07-20 | 北京橙鑫数据科技有限公司 | Business card proofreading method and system |
| CN106250830A (en) * | 2016-07-22 | 2016-12-21 | 浙江大学 | Digital book structured analysis processing method |
| CN106570105A (en) * | 2016-11-01 | 2017-04-19 | 广西电网有限责任公司电力科学研究院 | Structured processing method for power equipment test report |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108877764A (en) * | 2018-06-28 | 2018-11-23 | 掌阅科技股份有限公司 | Audio synthetic method, electronic equipment and the computer storage medium of talking e-book |
| CN108877764B (en) * | 2018-06-28 | 2019-06-07 | 掌阅科技股份有限公司 | Audio synthetic method, electronic equipment and the computer storage medium of talking e-book |
| CN109376658A (en) * | 2018-10-26 | 2019-02-22 | 信雅达系统工程股份有限公司 | A kind of OCR method based on deep learning |
| CN109522437A (en) * | 2018-11-30 | 2019-03-26 | 珠海格力电器股份有限公司 | A kind of information search method of paper document, device, storage medium and terminal |
| CN111290688A (en) * | 2018-12-06 | 2020-06-16 | 中兴通讯股份有限公司 | Multimedia note taking method, terminal and computer readable storage medium |
| CN111753814A (en) * | 2019-03-26 | 2020-10-09 | 杭州海康威视数字技术股份有限公司 | Sample generation method, device and equipment |
| CN111753814B (en) * | 2019-03-26 | 2023-07-25 | 杭州海康威视数字技术股份有限公司 | Sample generation method, device and equipment |
| CN110399953A (en) * | 2019-07-29 | 2019-11-01 | 爱布客（武汉）科技有限公司 | A kind of method of paper book and e-book synchronous read |
| CN111493829A (en) * | 2020-04-23 | 2020-08-07 | 四川大学华西医院 | Method, system and equipment for determining mild cognitive impairment recognition parameters |
| CN111556251A (en) * | 2020-05-20 | 2020-08-18 | 深圳前海微众银行股份有限公司 | Electronic book generation method, device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107145859A (en) | E-book conversion process method, device and computer-readable recording medium | |
| CN107944447A (en) | Image classification method and device | |
| CN107798327A (en) | Character identifying method and device | |
| JP5120777B2 (en) | Electronic data editing apparatus, electronic data editing method and program | |
| CN107527059A (en) | Character recognition method, device and terminal | |
| CN106651955A (en) | Method and device for positioning object in picture | |
| CN107679483A (en) | Number plate recognition methods and device | |
| CN107609560A (en) | Character recognition method and device | |
| CN107679533A (en) | Character recognition method and device | |
| CN105095366B (en) | Word message treating method and apparatus | |
| CN114240882A (en) | Defect detection method and device, electronic equipment and storage medium | |
| CN108121736A (en) | A kind of descriptor determines the method for building up, device and electronic equipment of model | |
| CN107665354A (en) | Identify the method and device of identity card | |
| CN107133354A (en) | The acquisition methods and device of description information of image | |
| CN109819288A (en) | Determination method, apparatus, electronic equipment and the storage medium of advertisement dispensing video | |
| CN107247936A (en) | Image-recognizing method and device | |
| CN107105517A (en) | Method for connecting network and device | |
| US11263447B2 (en) | Information processing method, information processing device, mobile terminal, and storage medium | |
| CN111209354A (en) | Method and device for judging repetition of map interest points and electronic equipment | |
| CN110717399A (en) | Face recognition method and electronic terminal equipment | |
| CN106921958A (en) | The method and apparatus for quitting the subscription of business | |
| CN106485246A (en) | Character identifying method and device | |
| CN108509406A (en) | A kind of language material abstracting method, device and electronic equipment | |
| CN109002184A (en) | A kind of association method and device of input method candidate word | |
| CN105243051A (en) | Creation method, extraction method and apparatus for hotel order short message template |
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: 20170908 |