CN109784165A - Generation method, device, terminal and the storage medium of poem content - Google Patents
Generation method, device, terminal and the storage medium of poem content Download PDFInfo
- Publication number
- CN109784165A CN109784165A CN201811520555.3A CN201811520555A CN109784165A CN 109784165 A CN109784165 A CN 109784165A CN 201811520555 A CN201811520555 A CN 201811520555A CN 109784165 A CN109784165 A CN 109784165A
- Authority
- CN
- China
- Prior art keywords
- poem
- information
- content
- picture
- generation method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The present invention relates to natural language processing technique fields, provide generation method, device, terminal and the storage medium of a kind of poem content.The generation method of the poem content includes: the picture for obtaining client and uploading;The picture material for extracting the picture analyzes described image content, obtains the context information that described image content is included;The poem information to match with the context information is obtained, and poem content is generated according to the poem information and is back to the client.The present invention realizes the picture according to upload, generates the poem content to match with the image content, enriches poem generating mode, promotes user experience.
Description
Technical field
The present invention relates to natural language processing technique field more particularly to a kind of generation methods of poem content, device, end
End and storage medium.
Background technique
With the fast development of deep learning, using convolutional neural networks and Recognition with Recurrent Neural Network as the deep learning frame of representative
Frame is used widely in computer vision and natural language processing field.In natural language processing field, traditional poem is raw
At a machine translation problem is counted as, using a upper poem as original language, next poem passes through as object language
Statistical machine translation model is translated, and plus the rhymed equal constraint of level and oblique tone, obtains a complete poem of head.
But the text information that current poem content automatic generation method can only be inputted by user, such as keyword, language
The information such as sentence carry out intelligent generation, and generation form is single, lack flexibility, can not adapt to diversified demand.
Summary of the invention
The present invention provides generation method, device, terminal and the storage medium of a kind of poem content, to solve in current poem
The problem of generation form for holding generation method is single, lacks flexibility, can not adapt to diversified demand.
To solve the above problems, the present invention adopts the following technical scheme:
The present invention provides a kind of generation method of poem content, includes the following steps:
Obtain the picture that client uploads;
The picture material for extracting the picture analyzes described image content, obtains described image content and is included
Context information;
The poem information to match with the context information is obtained, and poem content is generated according to the poem information and is returned
To the client.
Optionally, the context information includes Entity Semantics information and relationship semantic information, described to described image content
The step of being analyzed, comprising:
Multi-level mapping is carried out to described image content using convolutional neural networks model, obtains the reality in picture material
Body semantic information and relationship semantic information;
The step of poem information that the acquisition matches with the context information, comprising:
The style of poem content to be generated is generated according to the Entity Semantics information and relationship semantic information;
Obtain the poem information to match with the style.
Optionally, described that multi-level mapping is carried out to the figure content picture using convolutional neural networks model, obtain figure
As the step of Entity Semantics information and relationship semantic information in content, comprising:
Preliminary treatment is carried out to the picture material close to picture boundary using the bottom neuron of convolutional neural networks model;
The local message of each pictorial element in picture material is identified using the middle layer neuron of convolutional neural networks model;
According to preliminary processing results and local message, image is extracted using the high-rise neuron of convolutional neural networks model
The global information of content;
The global information is inputted in LSTM model, the Entity Semantics information and relationship semanteme letter in picture material are obtained
Breath.
Optionally, the generation method of the poem content further include:
According to the Entity Semantics information and relationship semantic information, the base of entity in picture material is extracted using corpus
This information;
According to the essential information, high frequency vocabulary relevant to the essential information is obtained using textRank algorithm;
Obtain the poem information comprising the high frequency vocabulary.
Optionally, described and according to the poem information generate poem content the step of, comprising:
According to the high frequency vocabulary of the poem information, each sentence is generated with the high frequency vocabulary using sequence to series model
The poem content being the theme.
Optionally, described and according to the poem information generate poem content after, further includes:
Assessment models are constructed, are scored using poem content of the assessment models to generation;
When the verse score value in poem content is lower than preset value, the verse is re-generated, until in poem
The score value of all verses in appearance stops when being all higher than preset value.
Optionally, described the step of being scored using poem content of the assessment models to generation, comprising:
According to the collocation frequency in poem content between words, using the assessment models to the continuity of poem content into
Row scoring;
According to the correspondence that sound between words is restrained, the verse of front and back two in poem content is carried out using the assessment models
Scoring;
According to rhymed rule, scored using the assessment models the rhymed of poem content.
A kind of generating means of poem content provided by the invention, comprising:
Module is obtained, for obtaining the picture of client upload;
Analysis module analyzes described image content for extracting the picture material of the picture, obtains the figure
The context information for being included as content;
Generation module, for obtaining the poem information to match with the context information, and it is raw according to the poem information
The client is back at poem content.
The present invention provides a kind of terminal, including memory and processor, is stored with computer-readable finger in the memory
It enables, when the computer-readable instruction is executed by the processor, so that the processor executes as above described in any item poems
The step of generation method of word content.
The present invention provides a kind of storage medium, is stored thereon with computer program, and the computer program is held by processor
When row, the generation method of as above described in any item poem contents is realized.
Compared with the existing technology, technical solution of the present invention at least has following advantage:
The generation method of poem content provided by the invention is mentioned from the picture after obtaining the picture that client uploads
Picture material is taken out, and described image content is analyzed, to obtain the context information that described image content is included;Then
The poem information to match with the context information is obtained, and poem content is generated according to the poem information and is back to the visitor
Family end gets rid of the limitation of descriptor so as to generate the poem content to match with the image content according to picture, enriches
Poem generating mode promotes user experience.
The generation method of poem content provided by the invention, also using convolutional neural networks model to described image content
Multi-level mapping is carried out, the Entity Semantics information and relationship semantic information in picture material are obtained, then according to the entity
Semantic information and relationship semantic information generate the style of poem content to be generated, are believed with obtaining the poem to match with the style
Breath improves the generation quality of poem to generate the poem of reflection picture material style.
The generation method of poem content provided by the invention can also utilize the assessment models by constructing assessment models
It scores the poem content of generation;When the verse scoring in poem content is lower than preset value, weight is carried out to the verse
It generates, until the scoring of all verses in poem content stops when being all higher than preset value, it is good to guarantee to have in the poem generated
Good quality.
Detailed description of the invention
Fig. 1 is the implementation environment figure of the generation method of the poem content provided in one embodiment of the invention;
Fig. 2 is a kind of embodiment flow diagram of generation method of poem content of the present invention;
Fig. 3 is a kind of embodiment module frame chart of generating means of poem content of the present invention;
Fig. 4 is the internal structure block diagram of terminal in one embodiment of the invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described.
In some processes of the description in description and claims of this specification and above-mentioned attached drawing, contain according to
Multiple operations that particular order occurs, but it should be clearly understood that these operations can not be what appears in this article suitable according to its
Sequence is executed or is executed parallel, and the serial number of operation such as S11, S12 etc. be only used for distinguishing each different operation, serial number
It itself does not represent and any executes sequence.In addition, these processes may include more or fewer operations, and these operations can
To execute or execute parallel in order.It should be noted that the description such as " first " herein, " second ", is for distinguishing not
Same message, equipment, module etc., does not represent sequencing, does not also limit " first " and " second " and be different type.
It will appreciated by the skilled person that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in specification of the invention
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member
Part is " connected " or when " coupled " to another element, it can be directly connected or coupled to other elements, or there may also be
Intermediary element.In addition, " connection " used herein or " coupling " may include being wirelessly connected or wirelessly coupling.It is used herein to arrange
Diction "and/or" includes one or more associated wholes for listing item or any cell and all combinations.
It will appreciated by the skilled person that unless otherwise defined, all terms used herein (including technology art
Language and scientific term), there is meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Should also
Understand, those terms such as defined in the general dictionary, it should be understood that have in the context of the prior art
The consistent meaning of meaning, and unless idealization or meaning too formal otherwise will not be used by specific definitions as here
To explain.
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description in which the same or similar labels are throughly indicated same or similar element or has same or like function
Element.Obviously, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, those skilled in the art's every other implementation obtained without creative efforts
Example, shall fall within the protection scope of the present invention.
Fig. 1 is the implementation environment figure of the generation method of the poem content provided in one embodiment, as shown in Figure 1, at this
In implementation environment, including server 110, terminal 120.Terminal 120 is connect by network with server, is pacified in the terminal 120
Equipped with corresponding client, user can generate corresponding poem according to the picture by client uploading pictures, server 110
Content is simultaneously back to client.Wherein, above-mentioned network may include internet, 2G/3G/4G, wifi etc..
It should be noted that server 110 can be independent physical server or terminal, it is also possible to multiple physics clothes
The server cluster for device composition of being engaged in can be to provide the basic cloud computing service such as Cloud Server, cloud database, cloud storage and CDN
Cloud Server.
Terminal 120 can be smart phone, tablet computer, laptop, desktop computer, intelligent sound box, intelligent hand
Table etc., however, it is not limited to this.
Referring to Fig. 2, a kind of generation method of poem content provided by the present invention, in one of embodiment, packet
Include following steps:
S21, the picture that client uploads is obtained;
User can be the photo of shooting by terminal to server uploading pictures, the picture, download on the net
Picture or paint.
S22, the picture material for extracting the picture, analyze described image content, obtain described image content institute
The context information for including;
In the present embodiment, after server gets the picture of user's upload, picture material is extracted from the picture, such as
Extract the characteristic information of picture, such as color characteristic, textural characteristics, shape feature, spatial relation characteristics;Then to the figure
As content is analyzed, the personage for including in picture material and object and its correlation between them are analyzed, and obtain institute
State the context information that picture material is included, wherein the context information may include between the entity, tone, entity of picture material
The information such as correlation.
S23, the poem information to match with the context information is obtained, and poem content is generated according to the poem information
It is back to the client.
In the present embodiment, the poem to match with the context information letter can be obtained according to the context information of picture material
Breath, and the verse to be matched according to the poem Automatic generation of information and context information, such as by screening to poem information, replace
It the modes such as changes, delete, integrating, generating new verse, and select the highest verse of matching degree, poem content is obtained, to express figure
The picture material of piece.It wherein, include poem model needed for generating poem content in the poem information, it is such as first to Tang poetry 300
It carries out punctuate to meet, after the deletion of paragraph, generates poem model.
The generation method of poem content provided by the invention is mentioned from the picture after obtaining the picture that client uploads
Picture material is taken out, and described image content is analyzed, to obtain the context information that described image content is included;Then
The poem information to match with the context information is obtained, and poem content is generated according to the poem information and is back to the visitor
Family end gets rid of the limitation of descriptor so as to generate the poem content to match with the image content according to picture, enriches
Poem generating mode promotes user experience.
Optionally, the context information includes Entity Semantics information and relationship semantic information, described to described image content
The step of being analyzed, comprising:
Multi-level mapping is carried out to described image content using convolutional neural networks model, obtains the reality in picture material
Body semantic information and relationship semantic information;
In the present embodiment, different layers are made respectively to picture material using the neuron of convolutional neural networks different levels
Secondary processing, to obtain the Entity Semantics information and relationship semantic information in picture material, the Entity Semantics information includes
Pictorial element in picture material further includes other characteristic informations in pictorial element such as sky, bird or the pictorial element of tree,
As sky be it is blue, tree is straight etc..Relationship of the relationship semantic information between entity and entity, if bird flies in the sky,
Fish such as swims at the mapping relations.
The step of poem information that the acquisition matches with the context information, comprising:
The style of poem content to be generated is generated according to the Entity Semantics information and relationship semantic information;
Obtain the poem information to match with the style.
In the present embodiment, the poem information in resources bank is segmented using Chinese lexical analysis tool, is obtained
Corresponding vocabulary, such as to " withered vine on an old tree dusk crow, small bridge over the flowing stream other " in poem information can segment for " withered rattan, veteran, confused crow,
Foot bridge, flowing water, other ", and corresponding poem style is matched to each participle, is stored in vocabulary.When generating poem content,
It can be obtained from vocabulary and picture material according to all Entity Semantics information and relationship semantic information obtained in picture material
The poem style to match, such as include bird, sky, sea, fish Entity Semantics information in picture material, relationship semantic information is
Bird is freeed ballooning in the sky, and fish is played heartily in the sea, then corresponding poem content is then easily cheerful and light-hearted style, is generated similar
In the poem content of " Hai Kuo appoints bird to fly with fish dive, sky ".For another example, when the entity in picture material is sky, sunset clouds, bird, lake
When water, relationship semantic information is that sky joins together with water, and in the sky, then corresponding poem content is then aestheticism for bird and sunset clouds
Style is similar to the poem content of " sunset clouds flies together with lonely duck, and it is of the same colour that autumn waters -- limid eyes are total to the vast of heaven ".
It is in one embodiment, described that multi-level mapping is carried out to the figure content picture using convolutional neural networks model,
The step of obtaining the Entity Semantics information and relationship semantic information in picture material, comprising:
Preliminary treatment is carried out to the picture material close to picture boundary using the bottom neuron of convolutional neural networks model;
In the present embodiment, the bottom neuron of convolutional neural networks model handles the boundary of image, such as can be right
The operation such as blurred or deleted close to the incomplete pictorial element of picture boundary, to prepare for subsequent be further processed.
The local message of each pictorial element in picture material is identified using the middle layer neuron of convolutional neural networks model;
In the present embodiment, the middle layer neuron of convolutional neural networks model can describe more complicated picture material, than
Such as the head of bird or wing in picture, image element attributes are tentatively judged.
According to preliminary processing results and local message, image is extracted using the high-rise neuron of convolutional neural networks model
The global information of content;
The local message that high-rise neuron can be identified according to the result and middle layer neuron of BORDER PROCESSING, to the complete of image
Office's information extracts, such as the pictorial elements such as bird, sky or lake surface in image.
The global information is inputted in LSTM model, the Entity Semantics information and relationship semanteme letter in picture material are obtained
Breath.
In the present embodiment, it is handled using global information of the LSTM model to input, obtains owning in picture material
Entity Semantics information and relationship semantic information, for example, including sky, bird, sea, fish entity, LSTM model in picture material
Frequency is arranged in pairs or groups by the relationship between sky and bird, Yu Yuhai according to entity.Wherein, the LSTM model is one kind of RNN model
Improved model.In LSTM model, the traditional neural member in RNN model, each storage unit and one are replaced using storage unit
A input gate, an out gate and one are associated across the noiseless internal state for being sent into itself of time step.Compared to RNN
Model, LSTM model have stronger memory capability, can remember longer historical information, be widely used in image analysis.
The present invention carries out multi-level processing to the figure content picture by convolutional neural networks model, with comprehensive assurance figure
The picture material of piece generates the poem content for covering all key images information of the picture substantially.
In one embodiment, the generation method of the poem content further include:
According to the Entity Semantics information and relationship semantic information, the base of entity in picture material is extracted using corpus
This information;
According to the essential information, high frequency vocabulary relevant to the essential information is obtained using textRank algorithm;
Obtain the poem information comprising the high frequency vocabulary.
In the present embodiment, according to the Entity Semantics information and relationship semantic information, entity is obtained using corpus
The essential informations such as classification, attribute and relationship, to be obtained by textRank algorithm related to entity according to the essential information
The high frequency vocabulary of connection, such as according to the relationship of bird, sky and bird and sky, the high frequencies such as hovered, circled in the air, soared to great heights
Vocabulary, to generate the poem being the theme with the high frequency vocabulary according to the high frequency vocabulary.In addition, after obtaining high frequency vocabulary, also
Can be according to the existing frequency of high frequency words remittance abroad just, the high frequency vocabulary for selecting the frequency of occurrences forward, as the master for generating poem content
Topic.Wherein, the textRank algorithm is a kind of algorithm for doing keyword extraction, can be used for extracting phrase and automatic
Abstract.The method for extracting keyword phrase is based on keyword extraction, can simply think: if several keywords extracted exist
It is adjacent in text, then constituting a key phrase being extracted.
Optionally, described and according to the poem information generate poem content the step of, comprising:
According to the high frequency vocabulary of the poem information, each sentence is generated with the high frequency vocabulary using sequence to series model
The poem content being the theme.
Specifically, when generating each hexameter, the essential information of acquisition and high frequency words Correspondent are crossed into sequence to sequence
Model generates the verse of every a line reflection high frequency vocabulary, so that the hexameter generated can take into account fluency and take into account meaning.
For the sequence to series model (Seq2Seq) just as a translation model, input is a sequence (such as English sentence
Son), output is also a sequence (such as translator of French corresponding to the english sentence), be can be used for the Sequence Transformed of domain
For the sequence in another domain, and text can be generated at any time, and the list entries of this model and the length of output sequence are variable
Change.
In one embodiment, described and according to the poem information generate poem content after, further includes:
Assessment models are constructed, are scored using poem content of the assessment models to generation;
When the verse score value in poem content is lower than preset value, the verse is re-generated, until in poem
The score value of all verses in appearance stops when being all higher than preset value.
In the present embodiment, by constructing assessment models, continuity, globality and signature can be carried out to the poem content of generation
The assessment of rhythm etc. regenerates the lower verse of score, to guarantee the indices of the poem content generated preferable
In the range of, it improves poem and generates quality.Wherein, when assessing the various aspects of poem content, the scoring of various aspects is obtained
Afterwards, a comprehensive score can be also obtained according to the scoring of various aspects, when the comprehensive score is lower than default score value, to described
Poem content is re-generated;Only this aspect can also accordingly be adjusted when the scoring in poem content in a certain respect is lower
It is whole, such as when the rhymed scoring of the poem of generation is not met the requirements, individually adjust the rhymed of poem content.
Optionally, described the step of being scored using poem content of the assessment models to generation, comprising:
According to the collocation frequency in poem content between words, using the assessment models to the continuity of poem content into
Row scoring;
According to the correspondence that sound between words is restrained, the verse of front and back two in poem content is carried out using the assessment models
Scoring;
According to rhymed rule, scored using the assessment models the rhymed of poem content.
In the present embodiment, the mode that can use high frequency collocation calculates the frequency that front and back is arranged in pairs or groups between word to obtain accordingly
Score with characterize verse continuity evaluation;The matching that poem content is evaluated using the correspondence of corpus sound rule, is not existed
Vocabulary in sound rule initiation, can calculate the similarity of other words in the hexameter of front and back two to judge the correlation of remaining word, into
And the globality of poem content is evaluated;Also using the rhymed rule of poem, commenting for rhymed aspect is carried out to the poem of generation
Valence obtains the preferable poem content of total quality so that the indices to poem are all assessed.
Referring to FIG. 3, the embodiment of the present invention also provides a kind of generating means of poem content, in a kind of the present embodiment,
Including obtaining module 31, analysis module 32 and generation module 33.Wherein,
Module 31 is obtained, for obtaining the picture of client upload;
User can be the photo of shooting by terminal to server uploading pictures, the picture, download on the net
Picture or paint.
Analysis module 32 is analyzed described image content, described in acquisition for extracting the picture material of the picture
The context information that picture material is included;
In the present embodiment, after server gets the picture of user's upload, picture material is extracted from the picture, such as
Extract the characteristic information of picture, such as color characteristic, textural characteristics, shape feature, spatial relation characteristics;Then to the figure
As content is analyzed, the personage for including in picture material and object and its correlation between them are analyzed, and obtain institute
State the context information that picture material is included, wherein the context information may include between the entity, tone, entity of picture material
The information such as correlation.
Generation module 33, for obtaining the poem information to match with the context information, and according to the poem information
It generates poem content and is back to the client.
In the present embodiment, the poem to match with the context information letter can be obtained according to the context information of picture material
Breath, and the verse to be matched according to the poem Automatic generation of information and context information, such as by screening to poem information, replace
It the modes such as changes, delete, integrating, generating new verse, and select the highest verse of matching degree, poem content is obtained, to express figure
The picture material of piece.It wherein, include poem model needed for generating poem content in the poem information, it is such as first to Tang poetry 300
It carries out punctuate to meet, after the deletion of paragraph, generates poem model.
The generating means of poem content provided by the invention are mentioned from the picture after obtaining the picture that client uploads
Picture material is taken out, and described image content is analyzed, to obtain the context information that described image content is included;Then
The poem information to match with the context information is obtained, and poem content is generated according to the poem information and is back to the visitor
Family end gets rid of the limitation of descriptor so as to generate the poem content to match with the image content according to picture, enriches
Poem generating mode promotes user experience.
About the device in above-described embodiment, wherein modules execute the concrete mode of operation in related this method
Embodiment in be described in detail, no detailed explanation will be given here.
Fig. 4 shows the block diagram of the part-structure of mobile phone relevant to terminal provided in an embodiment of the present invention.With reference to figure
4, mobile phone include: radio frequency (Radio Frequency, RF) circuit 610, memory 620, input unit 630, display unit 640,
Sensor 650, voicefrequency circuit 660, Wireless Fidelity (wireless fidelity, WiFi) module 670, processor 680 and
The components such as power supply 690.It will be understood by those skilled in the art that handset structure shown in Fig. 4 does not constitute the restriction to mobile phone,
It may include perhaps combining certain components or different component layouts than illustrating more or fewer components.
It is specifically introduced below with reference to each component parts of the Fig. 4 to mobile phone:
RF circuit 610 can be used for receiving and sending messages or communication process in, signal sends and receivees, particularly, by base station
After downlink information receives, handled to processor 680;In addition, the data for designing uplink are sent to base station.In general, RF circuit packet
Include but be not limited to antenna, at least one amplifier, transceiver, coupler, low-noise amplifier (Low Noise
Amplifier, LNA), duplexer etc..In addition, RF circuit 60 can also be communicated with network and other equipment by wireless communication.
Any communication standard or agreement, including but not limited to global system for mobile communications (Global can be used in above-mentioned wireless communication
System of Mobile communication, GSM), general packet radio service (General Packet Radio
Service, GPRS), CDMA (Code Division Multiple Access, CDMA), wideband code division multiple access
(Wideband Code Division Multiple Access, WCDMA), long term evolution (Long Term Evolution,
LTE)), Email, short message service (Short Messaging Service, SMS) etc..
Memory 620 can be used for storing computer-readable store instruction and module, and processor 680 is stored in by operation
The computer-readable store instruction and module of memory 620, thereby executing the various function application and data processing of mobile phone.
Memory 620 can mainly include storing program area and storage data area, wherein storing program area can storage program area, at least
Application program needed for one function (such as sound-playing function, image player function etc.) etc.;Storage data area can store root
Created data (such as audio data, phone directory etc.) etc. are used according to mobile phone.In addition, memory 620 may include high speed
Random access memory, can also include nonvolatile memory, a for example, at least disk memory, flush memory device or
Other volatile solid-state parts.
Input unit 630 can be used for receiving the number or character information of input, and generate the user setting with mobile phone 600
And the related key signals input of function control.Specifically, input unit 630 may include touch panel 631 and other inputs
Equipment 632.Touch panel 631, also referred to as touch screen, collecting the touch operation of user on it or nearby, (for example user makes
With the operation of any suitable object or attachment such as finger, stylus on touch panel 631 or near touch panel 631), and
Corresponding attachment device is driven according to preset formula.Optionally, touch panel 631 may include touch detecting apparatus and touching
Touch two parts of controller.Wherein, the touch orientation of touch detecting apparatus detection user, and detect touch operation bring letter
Number, transmit a signal to touch controller;Touch controller receives touch information from touch detecting apparatus, and is converted into
Contact coordinate, then give processor 680, and order that processor 680 is sent can be received and executed.Furthermore, it is possible to using
The multiple types such as resistance-type, condenser type, infrared ray and surface acoustic wave realize touch panel 631.It is defeated in addition to touch panel 631
Entering unit 630 can also include other input equipments 632.Specifically, other input equipments 632 can include but is not limited to physics
One of keyboard, function key (such as volume control button, switch key etc.), trace ball, mouse, operating stick etc. are a variety of.
Display unit 640 can be used for showing information input by user or be supplied to user information and mobile phone it is various
Menu.Display unit 640 may include display panel 641, optionally, can use liquid crystal display (Liquid Crystal
Display, LCD), the forms such as Organic Light Emitting Diode (Organic Light-Emitting Diode, OLED) it is aobvious to configure
Show panel 641.Further, touch panel 631 can cover display panel 641, when touch panel 631 detect it is on it or attached
After close touch operation, processor 680 is sent to determine the type of touch event, is followed by subsequent processing device 680 according to touch event
Type corresponding visual output is provided on display panel 641.Although in Fig. 4, touch panel 631 and display panel 641
It is that the input and input function of mobile phone are realized as two independent components, but in some embodiments it is possible to by touch-control
Panel 631 and display panel 641 are integrated and that realizes mobile phone output and input function.
Mobile phone 600 may also include at least one sensor 650, such as optical sensor, motion sensor and other sensings
Device.Specifically, optical sensor may include ambient light sensor and proximity sensor, wherein ambient light sensor can be according to environment
The light and shade of light adjusts the brightness of display panel 641, and proximity sensor can close display panel when mobile phone is moved in one's ear
641 and/or backlight.As a kind of motion sensor, accelerometer sensor can detect in all directions (generally three axis) and add
The size of speed can detect that size and the direction of gravity when static, can be used to identify application (such as the horizontal/vertical screen of mobile phone posture
Switching, dependent game, magnetometer pose calibrating), Vibration identification correlation function (such as pedometer, tap) etc.;Also as mobile phone
The other sensors such as configurable gyroscope, barometer, hygrometer, thermometer, infrared sensor, details are not described herein.
Voicefrequency circuit 660, loudspeaker 661, microphone 662 can provide the audio interface between user and mobile phone.Audio-frequency electric
Electric signal after the audio data received conversion can be transferred to loudspeaker 661, be converted to sound by loudspeaker 661 by road 660
Signal output;On the other hand, the voice signal of collection is converted to electric signal by microphone 662, is turned after being received by voicefrequency circuit 660
It is changed to audio data, then by after the processing of audio data output processor 680, such as another mobile phone is sent to through RF circuit 610,
Or audio data is exported to memory 620 to be further processed.
WiFi belongs to short range wireless transmission technology, and mobile phone can help user's transceiver electronics postal by WiFi module 670
Part, browsing webpage and access streaming video etc., it provides wireless broadband internet access for user.Although Fig. 4 is shown
WiFi module 670, but it is understood that, and it is not belonging to must be configured into for mobile phone 600, it is can according to need completely not
Change in the range of the essence of invention and omits.
Processor 680 is the control centre of mobile phone, using the various pieces of various interfaces and connection whole mobile phone, is led to
It crosses operation or executes computer-readable store instruction and/or the module being stored in memory 620, and call and be stored in storage
Data in device 620 execute the various functions and processing data of mobile phone, to carry out integral monitoring to mobile phone.Optionally, it handles
Device 680 may include one or more processing units;Preferably, processor 680 can integrate application processor and modulation /demodulation processing
Device, wherein the main processing operation system of application processor, user interface and application program etc., modem processor is mainly located
Reason wireless communication.It is understood that above-mentioned modem processor can not also be integrated into processor 680.
Mobile phone 600 further includes the power supply 690 (such as battery) powered to all parts, it is preferred that power supply can pass through electricity
Management system and processor 680 are logically contiguous, to realize management charging, electric discharge and power consumption by power-supply management system
The functions such as management.
Although being not shown, mobile phone 600 can also include camera, bluetooth module etc., and details are not described herein.
In embodiments of the present invention, when the computer-readable store instruction and module that are stored in memory 620 are processed
When device executes, processor 680 may make to realize the generation method of above-mentioned poem content, and realizes in illustrated embodiment in poem
The function of corresponding module in the generating means of appearance.
In one embodiment, the invention also provides a kind of storage medium for being stored with computer-readable instruction, the meters
When calculation machine readable instruction is executed by one or more processors, so that one or more processors execute the life of above-mentioned poem content
At method.For example, the storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and light number
According to storage equipment etc..
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, which can be stored in a storage medium, the program
When being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, storage medium above-mentioned can for magnetic disk, CD, only
Read non-volatile memory mediums or random access memory (Random such as storage memory (Read-Only Memory, ROM)
Access Memory, RAM) etc..
Based on the above embodiments it is found that the maximum beneficial effect of the present invention is:
Generation method, device, terminal and the storage medium of poem content provided by the invention obtain the figure that client uploads
After piece, picture material is extracted from the picture, and analyze described image content, to obtain described image content institute
The context information for including;Then the poem information to match with the context information is obtained, and is generated according to the poem information
Poem content is back to the client, rich so as to generate the poem content to match with the image content according to picture
Rich poem generating mode, promotes user experience.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
Limitations on the scope of the patent of the present invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (10)
1. a kind of generation method of poem content, which comprises the steps of:
Obtain the picture that client uploads;
The picture material for extracting the picture analyzes described image content, obtains the feelings that described image content is included
Scape information;
The poem information to match with the context information is obtained, and poem content is generated according to the poem information and is back to institute
State client.
2. the generation method of poem content according to claim 1, which is characterized in that the context information includes entity language
Adopted information and relationship semantic information, described the step of described image content is analyzed, comprising:
Multi-level mapping is carried out to described image content using convolutional neural networks model, obtains the entity language in picture material
Adopted information and relationship semantic information;
The step of poem information that the acquisition matches with the context information, comprising:
The style of poem content to be generated is generated according to the Entity Semantics information and relationship semantic information;
Obtain the poem information to match with the style.
3. the generation method of poem content according to claim 2, which is characterized in that described to utilize convolutional neural networks mould
Type carries out multi-level mapping to the figure content picture, obtains Entity Semantics information in picture material and relationship semantic information
Step, comprising:
Preliminary treatment is carried out to the picture material close to picture boundary using the bottom neuron of convolutional neural networks model;
The local message of each pictorial element in picture material is identified using the middle layer neuron of convolutional neural networks model;
According to preliminary processing results and local message, picture material is extracted using the high-rise neuron of convolutional neural networks model
Global information;
The global information is inputted in LSTM model, the Entity Semantics information and relationship semantic information in picture material are obtained.
4. the generation method of poem content according to claim 2, which is characterized in that further include:
According to the Entity Semantics information and relationship semantic information, the basic letter of entity in picture material is extracted using corpus
Breath;
According to the essential information, high frequency vocabulary relevant to the essential information is obtained using textRank algorithm;
Obtain the poem information comprising the high frequency vocabulary.
5. the generation method of poem content according to claim 4, which is characterized in that described and according to the poem information
The step of generating poem content, comprising:
According to the high frequency vocabulary of the poem information, each sentence is generated based on the high frequency vocabulary using sequence to series model
The poem content of topic.
6. the generation method of poem content according to claim 1, which is characterized in that described and according to the poem information
After generation poem content, further includes:
Assessment models are constructed, are scored using poem content of the assessment models to generation;
When the verse score value in poem content is lower than preset value, the verse is re-generated, until in poem content
The score values of all verses stop when being all higher than preset value.
7. the generation method of poem content according to claim 6, which is characterized in that described to utilize the assessment models pair
The step of poem content of generation is scored, comprising:
According to the collocation frequency in poem content between words, commented using continuity of the assessment models to poem content
Point;
According to the correspondence that sound between words is restrained, two verses in front and back in poem content are commented using the assessment models
Point;
According to rhymed rule, scored using the assessment models the rhymed of poem content.
8. a kind of generating means of poem content characterized by comprising
Module is obtained, for obtaining the picture of client upload;
Analysis module analyzes described image content for extracting the picture material of the picture, obtains in described image
Hold included context information;
Generation module generates poem for obtaining the poem information to match with the context information, and according to the poem information
Word content is back to the client.
9. a kind of terminal, which is characterized in that including memory and processor, computer-readable finger is stored in the memory
It enables, when the computer-readable instruction is executed by the processor, so that the processor is executed as appointed in claim 1 to 7
The step of generation method of poem content described in one.
10. a kind of storage medium, is stored thereon with computer program, which is characterized in that the computer program is held by processor
When row, the generation method of the poem content as described in any one of claims 1 to 7 is realized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811520555.3A CN109784165A (en) | 2018-12-12 | 2018-12-12 | Generation method, device, terminal and the storage medium of poem content |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811520555.3A CN109784165A (en) | 2018-12-12 | 2018-12-12 | Generation method, device, terminal and the storage medium of poem content |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109784165A true CN109784165A (en) | 2019-05-21 |
Family
ID=66495897
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811520555.3A Pending CN109784165A (en) | 2018-12-12 | 2018-12-12 | Generation method, device, terminal and the storage medium of poem content |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109784165A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110232180A (en) * | 2019-06-06 | 2019-09-13 | 北京师范大学 | A kind of automatic proposition method and system towards classic poetry assessment |
CN110245339A (en) * | 2019-06-20 | 2019-09-17 | 北京百度网讯科技有限公司 | Article generation method, device, equipment and storage medium |
CN111797262A (en) * | 2020-06-24 | 2020-10-20 | 北京小米松果电子有限公司 | Poetry generation method and device, electronic equipment and storage medium |
CN111814488A (en) * | 2020-07-22 | 2020-10-23 | 网易(杭州)网络有限公司 | Poetry generation method and device, electronic equipment and readable storage medium |
CN113794915A (en) * | 2021-09-13 | 2021-12-14 | 海信电子科技(武汉)有限公司 | Server, display equipment, poetry and song endowing generation method and media asset playing method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107480132A (en) * | 2017-07-25 | 2017-12-15 | 浙江工业大学 | A kind of classic poetry generation method of image content-based |
CN107832292A (en) * | 2017-11-02 | 2018-03-23 | 合肥工业大学 | A kind of conversion method based on the image of neural network model to Chinese ancient poetry |
US20180150444A1 (en) * | 2016-11-28 | 2018-05-31 | Microsoft Technology Licensing, Llc | Constructing a Narrative Based on a Collection of Images |
CN108874779A (en) * | 2018-06-21 | 2018-11-23 | 东北大学 | The control method that system is write the poem according to figure established based on K8s cluster |
-
2018
- 2018-12-12 CN CN201811520555.3A patent/CN109784165A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180150444A1 (en) * | 2016-11-28 | 2018-05-31 | Microsoft Technology Licensing, Llc | Constructing a Narrative Based on a Collection of Images |
CN107480132A (en) * | 2017-07-25 | 2017-12-15 | 浙江工业大学 | A kind of classic poetry generation method of image content-based |
CN107832292A (en) * | 2017-11-02 | 2018-03-23 | 合肥工业大学 | A kind of conversion method based on the image of neural network model to Chinese ancient poetry |
CN108874779A (en) * | 2018-06-21 | 2018-11-23 | 东北大学 | The control method that system is write the poem according to figure established based on K8s cluster |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110232180A (en) * | 2019-06-06 | 2019-09-13 | 北京师范大学 | A kind of automatic proposition method and system towards classic poetry assessment |
CN110232180B (en) * | 2019-06-06 | 2020-11-03 | 北京师范大学 | Automatic proposition method and system for ancient poetry evaluation |
CN110245339A (en) * | 2019-06-20 | 2019-09-17 | 北京百度网讯科技有限公司 | Article generation method, device, equipment and storage medium |
CN111797262A (en) * | 2020-06-24 | 2020-10-20 | 北京小米松果电子有限公司 | Poetry generation method and device, electronic equipment and storage medium |
CN111814488A (en) * | 2020-07-22 | 2020-10-23 | 网易(杭州)网络有限公司 | Poetry generation method and device, electronic equipment and readable storage medium |
CN113794915A (en) * | 2021-09-13 | 2021-12-14 | 海信电子科技(武汉)有限公司 | Server, display equipment, poetry and song endowing generation method and media asset playing method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109784165A (en) | Generation method, device, terminal and the storage medium of poem content | |
CN109783798A (en) | Method, apparatus, terminal and the storage medium of text information addition picture | |
CN109145303A (en) | Name entity recognition method, device, medium and equipment | |
CN106845390B (en) | Video title generation method and device | |
CN110276075A (en) | Model training method, name entity recognition method, device, equipment and medium | |
CN104239535B (en) | A kind of method, server, terminal and system for word figure | |
CN107943860A (en) | The recognition methods and device that the training method of model, text are intended to | |
CN110232696A (en) | A kind of method of image region segmentation, the method and device of model training | |
CN104951432B (en) | The method and device that a kind of pair of information is handled | |
CN105005563B (en) | A kind of abstraction generating method and device | |
CN107784114A (en) | Recommendation method, apparatus, terminal and the storage medium of facial expression image | |
CN105303149B (en) | The methods of exhibiting and device of character image | |
CN109599079A (en) | A kind of generation method and device of music | |
CN107959883A (en) | Video editing method for pushing, system and intelligent mobile terminal | |
CN109063583A (en) | A kind of learning method and electronic equipment based on read operation | |
CN110334344A (en) | A kind of semanteme intension recognizing method, device, equipment and storage medium | |
CN110166828A (en) | A kind of method for processing video frequency and device | |
CN108228720B (en) | Identify method, system, device, terminal and the storage medium of target text content and original image correlation | |
CN104123937A (en) | Method, device and system for reminding setting | |
CN108062404A (en) | Processing method, device, readable storage medium storing program for executing and the terminal of facial image | |
CN109815363A (en) | Generation method, device, terminal and the storage medium of lyrics content | |
CN108959274A (en) | A kind of interpretation method and server of application program | |
CN110704661A (en) | Image classification method and device | |
CN110443769A (en) | Image processing method, image processing apparatus and terminal device | |
CN109947650A (en) | Script step process methods, devices and systems |
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 |