CN110097616A - A kind of joint drawing method, device, terminal device and readable storage medium storing program for executing - Google Patents
A kind of joint drawing method, device, terminal device and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN110097616A CN110097616A CN201910311177.6A CN201910311177A CN110097616A CN 110097616 A CN110097616 A CN 110097616A CN 201910311177 A CN201910311177 A CN 201910311177A CN 110097616 A CN110097616 A CN 110097616A
- Authority
- CN
- China
- Prior art keywords
- article
- main body
- picture
- manuscript base
- line
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000003860 storage Methods 0.000 title claims abstract description 28
- 239000000203 mixture Substances 0.000 claims abstract description 56
- 239000000463 material Substances 0.000 claims abstract description 42
- 239000002131 composite material Substances 0.000 claims abstract description 19
- 239000003973 paint Substances 0.000 claims abstract description 11
- 238000004590 computer program Methods 0.000 claims description 17
- 238000004519 manufacturing process Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 238000000465 moulding Methods 0.000 claims description 4
- 230000001537 neural effect Effects 0.000 claims 1
- 238000010422 painting Methods 0.000 abstract description 14
- 238000010586 diagram Methods 0.000 abstract description 6
- 239000004744 fabric Substances 0.000 abstract description 4
- 239000013589 supplement Substances 0.000 abstract description 3
- 239000000047 product Substances 0.000 description 14
- 230000006870 function Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 210000000746 body region Anatomy 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007591 painting process Methods 0.000 description 2
- 210000003813 thumb Anatomy 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/60—Editing figures and text; Combining figures or text
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of joint drawing method, device, terminal device and readable storage medium storing program for executing, the described method includes: obtaining the hand-drawing line manuscript base picture that user draws, when judgement is with given drawing subject article tag match, one group of composition distributed data for containing the article tag is chosen from preset composition material database；The article tag of each main body article according to included in the fabric diagram distributed data chooses material line manuscript base picture corresponding with each main body article from preset line original text shape library；In conjunction with the article size data and coordinate position data of each main body article, hand-drawing line manuscript base picture and material line manuscript base picture are successively subjected to uniform zoom and are shown in corresponding position, joint pictorial image is obtained.The present invention can by material database to the single item element image that user draws carry out draw supplement it is perfect so that the people without professional painting skill can also create the full element of composition paint abundant.
Description
Technical field
The present invention relates to technical field of image processing, more particularly, to a kind of joint drawing method, device, terminal device and
Readable storage medium storing program for executing.
Background technique
With the continuous development and progress of society and science and technology, occur many intelligent graphic products or work on the market
Tool, is supplied to people and paints to meet the cultural life demand of people.But existing graphic products and tool are required to
The artistic work that there is people certain professional painting skill and comprehensive composition ability can just draw out better effects.Due to most
People's not professional painting skill is counted, drawing single article element can also be satisfied with reluctantly, therefore existing graphic products can not be helped
Help others create that composition is full, element is abundant is satisfied with works.
Summary of the invention
The technical problem to be solved by the embodiment of the invention is that providing a kind of joint drawing method, device, terminal device
And readable storage medium storing program for executing, the method can be applied in graphic products or tool, so that graphic products or tool can be with
User carries out joint drawing, to help the people without professional painting skill that can also create the full element of composition drawing abundant
Works.
In order to solve the above-mentioned technical problems, the present invention provides a kind of joint drawing methods, comprising:
The hand-drawing line manuscript base picture that user draws is obtained, when the drawing subject article for judging the hand-drawing line manuscript base picture and giving
When tag match, one group of composition distribution for containing the drawing subject article label is randomly selected from preset composition material database
Data；Wherein, the composition distributed data includes article tag, article size data, the coordinate position number of each main body article
According to；
The article tag of each main body article according to included in the composition distributed data, from preset line manuscript base shape
Material line manuscript base picture corresponding with each main body article is chosen in library；
In conjunction with the article size data and the coordinate position data, successively by the hand-drawing line manuscript base picture and described
Material line manuscript base picture carries out uniform zoom and is shown in corresponding position, obtains joint pictorial image.
Further, before the hand-drawing line manuscript base picture that the acquisition user draws, further includes:
Obtain different types of common main body article, several main body article photos under different angle and different light；
According to the space density of various main body articles arrangement feature, the feature vector of the main body article photo is extracted；Its
In, space density arrangement feature includes composed structure, appearance profile, one or more in color material；
Convolution is utilized using the article tag of each classification main body article as output using the feature vector of extraction as input
Neural network carries out recognition training, extracts the standard template of each main body article under every kind of classification and stores, obtains each main body object
The identification model library of product.
Further, the method for building up of the composition material database includes:
It obtains and is thumbed up several high-quality pictures of the number higher than preset threshold on network, and feature is carried out to the high-quality picture
It extracts, obtains high-quality picture feature vector；
By the standard form feature vector of each main body article in the high-quality picture feature vector and the identification model library
It is compared, identifies each main body article for including in the high-quality picture, and frame choosing is carried out to each main body article
Label；
The article tag of the main body article of each frame choosing label is obtained using the identification model library, and calculates each frame choosing
The size of the main body article of label and its corresponding coordinate position in the high-quality picture, obtain each main body article
Article size data and coordinate position data；
The article tag, article size data and coordinate position data of each main body article are associated, corresponded to
The composition distributed data of high-quality picture；
The composition distributed data of several high-quality pictures is obtained one by one and is stored, and the composition material database is obtained.
Further, the method for building up of the line original text shape library includes:
Obtain several similar images of each main body article by frame choosing label in the high-quality picture；
Respectively using several similar images of each human subject article as input, model is drawn using preset line original text and is generated
Several corresponding material line manuscript base pictures；
By the article tag of each main body article and several material line manuscript base pictures corresponding with the human subject article into
Row associated storage obtains the line original text shape library.
Further, the method for building up of the line original text picture model includes:
Obtain the main body images of items of various moulding on network；
Obtain the manual line manuscript base shape for each main body article that user sketches out according to the main body images of items；
Using the main body images of items as input, using the corresponding line manuscript base shape of the main body images of items as export,
And using the manual line manuscript base shape as output learning sample, is learnt and trained using production confrontation network；
When the output of the arbiter in production confrontation network reaches preset stable state threshold value, training is completed, is obtained
Model is drawn to the line original text.
Optionally, judge the hand-drawing line manuscript base picture and the given whether matched mode of drawing subject article label, packet
It includes:
Arbiter using production confrontation network judges goods categories belonging to the hand-drawing line manuscript base picture, obtains probability
Highest n article tags to be matched, judge whether there is at least one described article tag to be matched and the drawing topic object
The success of product tag match；Wherein, n is the integer greater than 0；
If so, determining that the hand-drawing line manuscript base picture matches with the drawing subject article label；
If it is not, then determining that the hand-drawing line manuscript base picture and the drawing subject article label mismatch.
Further, the joint drawing method further include:
When judging that the hand-drawing line manuscript base picture and given drawing subject article label mismatch, the hand-drawing line is removed
Manuscript base picture, and prompt to paint again.
In order to solve identical technical problem, the present invention also provides a kind of joint picture-drawing devices, including composition distribution number
Module and joint drawing module are chosen according to selection module, material line manuscript base picture；
The composition distributed data chooses module, for obtaining the hand-drawing line manuscript base picture of user's drafting, when judging the hand
When line drawing manuscript base picture and given drawing subject article tag match, one group is randomly selected from preset composition material database and is contained
The composition distributed data of the drawing subject article label；Wherein, the composition distributed data includes the object of each main body article
Product label, article size data, coordinate position data；
The material line manuscript base picture chooses module, is used for each main body object according to included in the composition distributed data
The article tag of product chooses material line manuscript base picture corresponding with each main body article from preset line original text shape library；
The joint drawing module, is used in conjunction with the article size data and the coordinate position data, successively by institute
It states hand-drawing line manuscript base picture and the material line manuscript base picture carries out uniform zoom and is shown in corresponding position, obtain combining and draw
Picture picture.
In order to solve identical technical problem, the present invention also provides a kind of joint paint terminal device, including processor,
Memory and storage in the memory and are configured as the computer program executed by the processor, the memory
It is coupled with the processor, and when the processor execution computer program, realizes described in any item joint drawing sides
Method.
In order to solve identical technical problem, the present invention also provides a kind of computer readable storage medium, the calculating
Machine readable storage medium storing program for executing is stored with computer program, wherein controls in computer program operation described computer-readable
Equipment where storage medium executes described in any item joint drawing methods.
It is had the following beneficial effects: by implementing the embodiment of the present invention
The present invention provides a kind of joint drawing method, device, terminal device and readable storage medium storing program for executing, the method is first
First, go out an article tag at random from preset article tag exam pool as drawing topic, user is prompted to correspond to
The drawing of article, when the hand-drawing line manuscript base picture for determining user article tag corresponding with the topic provided matches, from pre-
If composition material database in randomly select one group of composition distributed data comprising the article tag and carry out drawing composition, then basis
The article tag for each main body article for including in the composition distributed data, selected respectively from preset line original text shape library with respectively
The corresponding material line manuscript base picture of a main body article, and according to the article size data and position coordinates in the composition distributed data
Data are painted, and the hand-drawing line manuscript base picture of user is first carried out uniform zoom simultaneously in painting canvas by size and coordinate position
It has been shown that, then by remaining each main body article, by size and coordinate position, completion is shown one by one in painting canvas again, completes joint
Drawing.
Joint drawing method provided by the invention makes painting process simple, easy to operate, and user is very professional without having
Foundation of painting does not need to carry out complicated composition thinking, only needs arbitrarily to draw some single item liked according to drawing prompt
Element, system obtain user painting element after judgement is analyzed and processed to image, determine user drafting object it is consistent after,
The drafting content that corresponding element and user are matched from material database carries out comprehensive composition, on the basis of retaining user's original image,
Draw, it is perfect to supplement, and drawing is combined together with user, to complete the full element of width composition paint abundant, is made
The paint of higher level can also be drawn out by obtaining ordinary people, increase the interest of drawing.
Detailed description of the invention
Fig. 1 is the flow diagram for the joint drawing method that the embodiment of the present invention one provides；
Fig. 2 is the model schematic for the production confrontation network that the embodiment of the present invention one provides；
Fig. 3 is the structural schematic diagram for the joint picture-drawing device that the embodiment of the present invention one provides.
Specific embodiment
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
Whole description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Referring to Figure 1, the embodiment of the invention provides a kind of joint drawing methods, comprising:
Step S1, the hand-drawing line manuscript base picture that user draws is obtained, when the drawing for judging the hand-drawing line manuscript base picture and giving
When subject article tag match, one group is randomly selected from preset composition material database and contains the drawing subject article label
Composition distributed data；Wherein, the composition distributed data includes the article tag of each main body article, article size data, sits
Cursor position data；
In embodiments of the present invention, step S1 is to obtain the hand-drawing line manuscript base picture of user's drafting and choose composition distribution number
According to；Specifically, going out an article tag at random first from preset article tag exam pool as drawing topic, prompt to use
Family carries out the drafting of counterpart product on electronic touch screen, obtains the hand-drawing line manuscript base picture that user draws, and judges what user drew
Taxonomy of goods belonging to figure line manuscript base picture, it is random from preset composition material database if can match with given topic
Choose one group of composition distributed data for containing the drawing subject article label, wherein the composition distributed data contains this
Article tag, the article size data, coordinate position data of each main body article in fabric diagram.It is understood that for
Given drawing subject article label may have multiple groups composition distributed data to contain the article tag, this reality in composition material database
It applies in example and is patterned to randomly choose one group from the composition distributed data that these meet.
Step S2, the article tag of each main body article according to included in the composition distributed data, from preset
Material line manuscript base picture corresponding with each main body article is chosen in line original text shape library；
In embodiments of the present invention, step S2 for the fabric diagram distributed data chosen according to step S1 included it is each
The article tag of main body article selects material line corresponding with each main body article one by one from preset line original text shape library
Manuscript base picture；It is understood that may include in line original text shape library multiple and the main body article for each main body article
The corresponding material line manuscript base picture of article tag, the present embodiment randomly choose one from these material line manuscript base pictures and are drawn
It draws.
Step S3, in conjunction with the article size data and the coordinate position data, successively by the hand-drawing line manuscript base picture
And the material line manuscript base picture carries out uniform zoom and is shown in corresponding position, obtains joint pictorial image.
In embodiments of the present invention, step S3 for the fabric diagram distributed data chosen in conjunction with step S1 included it is each
The size data and coordinate position data of main body article, the hand-drawing line manuscript base picture for first drawing user press size and coordinate bit
It sets and carries out uniform zoom in painting canvas and show, the material line manuscript base picture of remaining each article is then pressed into size and coordinate
Position zooms in and out one by one in painting canvas and completion is shown, completes joint drawing.
In embodiments of the present invention, further, it before the hand-drawing line manuscript base picture that the acquisition user draws, also wraps
It includes:
Obtain different types of common main body article, several main body article photos under different angle and different light；
According to the space density of various main body articles arrangement feature, the feature vector of the main body article photo is extracted；Its
In, space density arrangement feature includes composed structure, appearance profile, one or more in color material；
Convolution is utilized using the article tag of each classification main body article as output using the feature vector of extraction as input
Neural network carries out recognition training, extracts the standard template of each main body article under every kind of classification and stores, obtains each main body object
The identification model library of product.
It is understood that the present invention can be built by the identification model library that above-mentioned steps carry out each main body article in advance
Vertical, establishment process is as follows:
Internet is first passed through to obtain under various personages, all angles of animals and plants and common main body article and various light
The picture of shooting, and picture is pre-processed, by being pre-processed except noise and interference, image cutting, normalization etc., to improve
The performance that server identifies image procossing；
It is arranged feature according to the different space density such as the composed structure of various main body articles, appearance profile, color material,
The image of each article is divided into M*N grid spaces, points in each grid is calculated and the ratio between images of items is always counted, obtain
To M*N dimensional feature vector, the feature vector of extraction is inputted, the reference name of each main body article of classifying is utilized as output
Convolutional neural networks carry out recognition training repeatedly, extract in the standard template deposit file of each main body article under every kind of classification,
Establish the identification model library of various articles.
In a preferred embodiment, the method for building up of the composition material database includes:
It obtains and is thumbed up several high-quality pictures of the number higher than preset threshold on network, and feature is carried out to the high-quality picture
It extracts, obtains high-quality picture feature vector；
By the standard form feature vector of each main body article in the high-quality picture feature vector and the identification model library
It is compared, identifies each main body article for including in the high-quality picture, and frame choosing is carried out to each main body article
Label；
The article tag of the main body article of each frame choosing label is obtained using the identification model library, and calculates each frame choosing
The size of the main body article of label and its corresponding coordinate position in the high-quality picture, obtain each main body article
Article size data and coordinate position data；
The article tag, article size data and coordinate position data of each main body article are associated, corresponded to
The composition distributed data of high-quality picture；
The composition distributed data of several high-quality pictures is obtained one by one and is stored, and the composition material database is obtained.
It should be noted that preferably, preset composition material database described in step S1 can carry out through the above steps
It establishes, specific:
It first passes through major social media website etc. and obtains the high-quality picture that user thumbs up number high (such as thumb up number and be higher than 50),
Pretreatment and feature extraction are carried out to picture, it will be in the feature vector of each picture of extraction and above-mentioned established identification model library
The feature vector of the standard template of each sorting articles of storage compares, and judges each main body article for including in image member
Element, and carry out frame choosing label one by one；Wherein it is determined that the main body article method for including in image can be with are as follows: chosen since center
3/4 (value can according to need any setting) surface area of image is set as body region, each object that judge mark goes out out
Whether product element has more than half image-region to be in the body region of setting, if then determining that it is primary articles element；
Then the size and coordinate in the picture of each main body article of every high-quality picture center choosing label are calculated
Position distribution data, and be associated with corresponding article tag, it obtains composition distributed data and stores；
Finally by composition distributed data (the corresponding article tag, main body article of the main body article identified in every image
Size, coordinate position distributed data) it stores one by one, obtain composition material database.
In a preferred embodiment, the method for building up of the line original text shape library includes:
Obtain several similar images of each main body article by frame choosing label in the high-quality picture；
Respectively using several similar images of each human subject article as input, model is drawn using preset line original text and is generated
Several corresponding material line manuscript base pictures；
By the article tag of each main body article and several material line manuscript base pictures corresponding with the human subject article into
Row associated storage obtains the line original text shape library.
It should be noted that preferably, the line original text shape library in step S2 can be established according to the method described above, specifically
, it is obtained and the main body of several corresponding same types according to above-mentioned high-quality picture by each main body article of frame choosing label first
Images of items, then respectively using these main body images of items as input, if it is corresponding to draw model generation using preset line original text
Dry material line manuscript base picture, finally obtains several corresponding material line manuscript base pictures of each main body images of items, by these
Material line manuscript base picture and article tag associated storage, obtain the line original text shape library.
In a preferred embodiment, the method for building up of the line original text picture model includes:
Obtain the main body images of items of various moulding on network；
Obtain the manual line manuscript base shape for each main body article that user sketches out according to the main body images of items；
Using the main body images of items as input, using the corresponding manual line manuscript base shape of the main body images of items as defeated
Out, learnt and trained using production confrontation network；
When the output of the arbiter in production confrontation network reaches preset stable state threshold value, training is completed, is obtained
Model is drawn to the line original text.
It should be noted that the method for building up that the line original text draws model includes but is not limited to above-mentioned steps, it is specific:
The image for obtaining the various moulding of various main body articles by internet first carries out gray proces to image.By
It is made of in color image multiple pixels, and each pixel is indicated by tri- values of RGB；Image is carried out at gray scale
Reason, can't influence the texture feature information of image, and each pixel only needs a gray value that can indicate, greatly improves
Image processing efficiency, is gray proces weighted mean method formula below:
F (i, j)=0.30R (i, j)+0.59G (i, j)+0.11B (i, j)
Wherein, i, j represent a pixel in the position of two-dimensional space vector, it may be assumed that the i-th row, jth column.
According to above-mentioned formula, the gray value of each pixel of each still image is calculated, value range is 0-255, is made above-mentioned
Black-white-gray state is all presented in image, the corresponding manual line manuscript base of each images of items of then being sketched out by way of manual drawing
Picture；
Then using the main body images of items as input, using the corresponding line manuscript base shape of the main body images of items as defeated
Out, and using the manual line manuscript base shape as output learning sample, learnt and trained using production confrontation network, work as institute
The output for stating the arbiter in production confrontation network reaches preset stable state threshold value (close to 1/2, specific value can be as needed
Setting) when, training is completed, the line original text is obtained and draws model.
Refer to Fig. 2, it is to be understood that production fights network (GAN, Generative Adversarial
Networks) it is a kind of deep learning model, is one of the method for unsupervised learning most prospect in complex distributions in recent years.Mesh
The preceding most often applied technical field of GAN is exactly that image generates.The basic principle of GAN are as follows: assuming that model there are two us, G
(Generator) and D (Discriminator), their function is respectively: G is the model of a generation picture, it receives one
A random noise z generates picture by this noise, is denoted as G (z)；D is a discrimination model, differentiates that a picture is not
It is " true ", its input parameter is x, and x represents a picture, and output D (x) represents x as the probability of true picture, if it is
1, just representing 100% is true picture, and exporting is 0, is unlikely to be true picture with regard to representing.In the training process, raw
It is just to try to generate true picture at the target of model G and removes deception discrimination model D.And the target of D is just to try to G generation
Picture and true picture are distinguished from.Under optimal state, the picture G (z) for being enough " mixing the spurious with the genuine " is can be generated in G；
For D, it is difficult to determine whether true the picture that G is generated is actually, therefore D (G (z))=0.5.When two models reach
When to stable state, i.e., the output of discrimination model D completes training, obtains the model G an of production, it can use next life close to 0.5
At picture.
In embodiments of the present invention, for production confrontation network (GAN) in training, generator G is for capturing data
Distribution, arbiter D be used to estimate one sample from truthful data rather than the probability of generation sample, arbiter D is constantly more
Newly, by will differentiate that result goes to guide how generator G optimizes as the optimization aim of generator G, to play generator G
The effect of objective function.If arbiter D can be easily by itself and real picture region for ambiguous conversion results
It does not come, while instructing generator G that generation is gone to be more clear reliable picture again.On the one hand study generates objective function to GAN, separately
On the one hand again according to study to objective function go to optimize model.When generator G and arbiter D reach stable state, i.e.,
When the output of arbiter D is close to 1/2, training is completed, the line original text is obtained and draws model.
In a preferred embodiment, judge the hand-drawing line manuscript base picture and given drawing subject article label whether
The mode matched, comprising:
Arbiter using production confrontation network judges goods categories belonging to the hand-drawing line manuscript base picture, obtains probability
Highest n article tags to be matched, judge whether there is at least one described article tag to be matched and the drawing topic object
The success of product tag match；Wherein, n is the integer greater than 0；
If so, determining that the hand-drawing line manuscript base picture matches with the drawing subject article label；
If it is not, then determining that the hand-drawing line manuscript base picture and the drawing subject article label mismatch.
In embodiments of the present invention, the hand-drawing line manuscript base picture and given drawing topic can be judged through the above way
Whether article tag matches, specifically, n can be set as 5, it may be assumed that and arbiter judges goods categories belonging to the hand-drawing line manuscript base picture,
Obtain highest five article tags of probability, if having in this five article tags one can with drawing subject article label
Match, that is, illustrates that the hand-drawing line manuscript base picture matches with the drawing subject article label.
Further, the joint drawing method further include:
When judging that the hand-drawing line manuscript base picture and given drawing subject article label mismatch, the hand-drawing line is removed
Manuscript base picture, and prompt to paint again.
Fig. 3 is referred to, in order to solve identical technical problem, the present invention also provides a kind of joint picture-drawing devices, including
Composition distributed data chooses module 1, material line manuscript base picture chooses module 2 and joint drawing module 3；
The composition distributed data chooses module 1, for obtaining the hand-drawing line manuscript base picture of user's drafting, when judging the hand
When line drawing manuscript base picture and given drawing subject article tag match, one group is randomly selected from preset composition material database and is contained
The composition distributed data of the drawing subject article label；Wherein, the composition distributed data includes the object of each main body article
Product label, article size data, coordinate position data；
The material line manuscript base picture chooses module 2, is used for each main body according to included in the composition distributed data
The article tag of article chooses material line manuscript base corresponding with each main body article from preset line original text shape library
Picture；
The joint drawing module 3, is used in conjunction with the article size data and the coordinate position data, successively by institute
It states hand-drawing line manuscript base picture and the material line manuscript base picture carries out uniform zoom and is shown in corresponding position, obtain combining and draw
Picture picture.
It is understandable that above-mentioned apparatus item embodiment, is, present invention reality corresponding with the method for the present invention item embodiment
The joint drawing side that any one method item embodiment of the present invention provides may be implemented in a kind of joint picture-drawing device for applying example offer
Method.
In order to solve identical technical problem, the present invention also provides a kind of joint paint terminal device, including processor,
Memory and storage in the memory and are configured as the computer program executed by the processor, the memory
It is coupled with the processor, and when the processor execution computer program, realizes described in any item joint drawing sides
Method.
The joint drawing terminal device can be desktop PC, notebook, palm PC and cloud server etc.
Calculate equipment.The processor can be central processing unit (Central Processing Unit, CPU), can also be it
His general processor, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
It is the control centre of the joint drawing terminal device Deng, the processor, is entirely combined using various interfaces and connection
Draw the various pieces of terminal device.
The memory can mainly include storing program area and storage data area, wherein storing program area can store operation
Application program needed for system, at least one function etc.；Storage data area, which can be stored, uses created data according to mobile phone
Deng.It can also include nonvolatile memory in addition, memory may include high-speed random access memory, such as hard disk, interior
It deposits, plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card,
Flash card (Flash Card), at least one disk memory, flush memory device or other volatile solid-state parts.
In order to solve identical technical problem, the present invention also provides a kind of computer readable storage medium, the calculating
Machine readable storage medium storing program for executing is stored with computer program, wherein controls in computer program operation described computer-readable
Equipment where storage medium executes described in any item joint drawing methods.
The computer program can be stored in a computer readable storage medium, and the computer program is by processor
When execution, it can be achieved that the step of above-mentioned each embodiment of the method.Wherein, the computer program includes computer program code,
The computer program code can be source code form, object identification code form, executable file or certain intermediate forms etc..Institute
State computer-readable medium may include: can carry the computer program code any entity or device, recording medium,
USB flash disk, mobile hard disk, magnetic disk, CD, computer storage, read-only memory (ROM, Read-Only Memory), arbitrary access
Memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It needs
It is bright, the content that the computer-readable medium includes can according in jurisdiction make laws and patent practice requirement into
Row increase and decrease appropriate, such as do not include electric load according to legislation and patent practice, computer-readable medium in certain jurisdictions
Wave signal and telecommunication signal.
It should be noted that the apparatus embodiments described above are merely exemplary, wherein described be used as separation unit
The unit of explanation may or may not be physically separated, and component shown as a unit can be or can also be with
It is not physical unit, it can it is in one place, or may be distributed over multiple network units.It can be according to actual
It needs that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.In addition, device provided by the invention
In embodiment attached drawing, the connection relationship between module indicate between them have communication connection, specifically can be implemented as one or
A plurality of communication bus or signal wire.Those of ordinary skill in the art are without creative efforts, it can understand
And implement.
To sum up, the implementation of the embodiments of the present invention has the following beneficial effects:
The present invention provides a kind of joint drawing method, device, terminal device and readable storage medium storing program for executing, the method is first
First, go out an article tag at random from preset article tag exam pool as drawing topic, user is prompted to correspond to
The drawing of article, when the hand-drawing line manuscript base picture for determining user article tag corresponding with the topic provided matches, from pre-
If composition material database in randomly select one group of composition distributed data comprising the article tag and carry out drawing composition, then basis
The article tag for each main body article for including in the composition distributed data, selected respectively from preset line original text shape library with respectively
The corresponding material line manuscript base picture of a main body article, and according to the article size data and position coordinates in the composition distributed data
Data are painted, and the hand-drawing line manuscript base picture of user is first carried out uniform zoom simultaneously in painting canvas by size and coordinate position
It has been shown that, then by remaining each main body article, by size and coordinate position, completion is shown one by one in painting canvas again, completes joint
Drawing.
Joint drawing method provided by the invention makes painting process simple, easy to operate, and user is very professional without having
Foundation of painting does not need to carry out complicated composition thinking, only needs arbitrarily to draw some single item liked according to drawing prompt
Element, system obtain user painting element after judgement is analyzed and processed to image, determine user drafting object it is consistent after,
The drafting content that corresponding element and user are matched from material database carries out comprehensive composition, on the basis of retaining user's original image,
Draw, it is perfect to supplement, and drawing is combined together with user, to complete the full element of width composition paint abundant, is made
The paint of higher level can also be drawn out by obtaining ordinary people, increase the interest of drawing.
The above is a preferred embodiment of the present invention, it is noted that for those skilled in the art
For, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also considered as
Protection scope of the present invention.
Claims (10)
1. a kind of joint drawing method characterized by comprising
The hand-drawing line manuscript base picture that user draws is obtained, when the drawing subject article label for judging the hand-drawing line manuscript base picture and giving
When matching, one group of composition distribution number for containing the drawing subject article label is randomly selected from preset composition material database
According to；Wherein, the composition distributed data includes article tag, the article size data, coordinate position data of each main body article；
The article tag of each main body article according to included in the composition distributed data, from preset line original text shape library
Choose material line manuscript base picture corresponding with each main body article；
In conjunction with the article size data and the coordinate position data, successively by the hand-drawing line manuscript base picture and the material
Line manuscript base picture carries out uniform zoom and is shown in corresponding position, obtains joint pictorial image.
2. joint drawing method according to claim 1, which is characterized in that in the hand-drawing line original text that the acquisition user draws
Before image, further includes:
Obtain different types of common main body article, several main body article photos under different angle and different light；
According to the space density of various main body articles arrangement feature, the feature vector of the main body article photo is extracted；Wherein, institute
State space density arrangement feature include composed structure, it is appearance profile, one or more in color material；
Convolutional Neural is utilized using the article tag of each classification main body article as output using the feature vector of extraction as input
Network carries out recognition training, extracts the standard template of each main body article under every kind of classification and stores, obtains each main body article
Identification model library.
3. joint drawing method according to claim 2, which is characterized in that the method for building up packet of the composition material database
It includes:
It obtains and is thumbed up several high-quality pictures of the number higher than preset threshold on network, and feature is carried out to the high-quality picture and is mentioned
It takes, obtains high-quality picture feature vector；
The standard form feature vector of each main body article in the high-quality picture feature vector and the identification model library is carried out
It compares, identifies each main body article for including in the high-quality picture, and frame choosing label is carried out to each main body article；
The article tag of the main body article of each frame choosing label is obtained using the identification model library, and calculates each frame choosing label
Main body article size and its corresponding coordinate position in the high-quality picture, obtain the object of each main body article
Product size data and coordinate position data；
The article tag, article size data and coordinate position data of each main body article are associated, it is high-quality to obtain correspondence
The composition distributed data of picture；
The composition distributed data of several high-quality pictures is obtained one by one and is stored, and the composition material database is obtained.
4. joint drawing method according to claim 3, which is characterized in that the method for building up packet of the line original text shape library
It includes:
Obtain several similar images of each main body article by frame choosing label in the high-quality picture；
Respectively using several similar images of each human subject article as input, model is drawn using preset line original text and generates correspondence
Several material line manuscript base pictures；
The article tag of each main body article and several material line manuscript base pictures corresponding with the human subject article are closed
Connection storage, obtains the line original text shape library.
5. joint drawing method according to claim 4, which is characterized in that the line original text draws the method for building up packet of model
It includes:
Obtain the main body images of items of various moulding on network；
Obtain the manual line manuscript base shape for each main body article that user sketches out according to the main body images of items；
Using the main body images of items as input, using the corresponding line manuscript base shape of the main body images of items as output, and with
The craft line manuscript base shape is learnt and is trained as output learning sample, using production confrontation network；
When the output of the arbiter in production confrontation network reaches preset stable state threshold value, training is completed, institute is obtained
It states line original text and draws model.
6. joint drawing method according to claim 1, which is characterized in that judge the hand-drawing line manuscript base as with it is given
The drawing whether matched mode of subject article label, comprising:
Arbiter using production confrontation network judges goods categories belonging to the hand-drawing line manuscript base picture, obtains probability highest
N article tags to be matched, judge whether there is at least one described article tag to be matched and the drawing subject article mark
Sign successful match；Wherein, n is the integer greater than 0；
If so, determining that the hand-drawing line manuscript base picture matches with the drawing subject article label；
If it is not, then determining that the hand-drawing line manuscript base picture and the drawing subject article label mismatch.
7. joint drawing method according to claim 1, which is characterized in that further include:
When judging that the hand-drawing line manuscript base picture and given drawing subject article label mismatch, the hand-drawing line manuscript base is removed
Picture, and prompt to paint again.
8. a kind of joint picture-drawing device, which is characterized in that choose module including composition distributed data, material line manuscript base picture chooses mould
Block and joint drawing module；
The composition distributed data chooses module, for obtaining the hand-drawing line manuscript base picture of user's drafting, when judging the hand-drawing line
When manuscript base picture and given drawing subject article tag match, randomly selected from preset composition material database one group contain it is described
The composition distributed data of drawing subject article label；Wherein, the composition distributed data includes the article mark of each main body article
Label, article size data, coordinate position data；
The material line manuscript base picture chooses module, for each main body article according to included in the composition distributed data
Article tag chooses material line manuscript base picture corresponding with each main body article from preset line original text shape library；
The joint drawing module, is used in conjunction with the article size data and the coordinate position data, successively by the hand
Line drawing manuscript base picture and the material line manuscript base picture carry out uniform zoom and are shown in corresponding position, obtain joint drawing design
Picture.
The terminal device 9. a kind of joint is painted, which is characterized in that in the memory including processor, memory and storage
And it is configured as the computer program executed by the processor, the memory and the processor couple, and the processing
When device executes the computer program, joint drawing method as described in any one of claim 1 to 7 is realized.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, wherein the equipment where controlling the computer readable storage medium in computer program operation executes such as right
It is required that 1 to 7 described in any item joint drawing methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910311177.6A CN110097616B (en) | 2019-04-17 | 2019-04-17 | Combined drawing method and device, terminal equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910311177.6A CN110097616B (en) | 2019-04-17 | 2019-04-17 | Combined drawing method and device, terminal equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110097616A true CN110097616A (en) | 2019-08-06 |
| CN110097616B CN110097616B (en) | 2023-11-28 |
Family
ID=67445151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910311177.6A Active CN110097616B (en) | 2019-04-17 | 2019-04-17 | Combined drawing method and device, terminal equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110097616B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111008573A (en) * | 2019-11-15 | 2020-04-14 | 广东智媒云图科技股份有限公司 | Limb structure generation method and device, terminal equipment and readable storage medium |
| CN111061902A (en) * | 2019-12-12 | 2020-04-24 | 广东智媒云图科技股份有限公司 | Drawing method and device based on text semantic analysis and terminal equipment |
| CN111612864A (en) * | 2020-04-27 | 2020-09-01 | 厦门盈趣科技股份有限公司 | Drawing method and system based on photo and image recognition |
| CN112148944A (en) * | 2020-09-22 | 2020-12-29 | 西京学院 | Manufacturing method, manufacturing system, terminal and storage medium of manual fiber picture |
| CN112837396A (en) * | 2021-01-29 | 2021-05-25 | 深圳市天耀创想网络科技有限公司 | Line draft generation method and device based on machine learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011002875A (en) * | 2009-06-16 | 2011-01-06 | Nippon Telegr & Teleph Corp <Ntt> | Plotting support device, plotting support method, and plotting support program |
| US20140267393A1 (en) * | 2013-03-15 | 2014-09-18 | Disney Enterprises, Inc. | Virtual scene generation based on imagery |
| US20180260984A1 (en) * | 2017-02-28 | 2018-09-13 | Corel Corporation | Vector graphics based live sketching methods and systems |
| CN109389660A (en) * | 2018-09-28 | 2019-02-26 | 百度在线网络技术（北京）有限公司 | Image generating method and device |
| CN109448079A (en) * | 2018-10-25 | 2019-03-08 | 广东智媒云图科技股份有限公司 | A kind of drawing bootstrap technique and equipment |
-
2019
- 2019-04-17 CN CN201910311177.6A patent/CN110097616B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011002875A (en) * | 2009-06-16 | 2011-01-06 | Nippon Telegr & Teleph Corp <Ntt> | Plotting support device, plotting support method, and plotting support program |
| US20140267393A1 (en) * | 2013-03-15 | 2014-09-18 | Disney Enterprises, Inc. | Virtual scene generation based on imagery |
| US20180260984A1 (en) * | 2017-02-28 | 2018-09-13 | Corel Corporation | Vector graphics based live sketching methods and systems |
| CN109389660A (en) * | 2018-09-28 | 2019-02-26 | 百度在线网络技术（北京）有限公司 | Image generating method and device |
| CN109448079A (en) * | 2018-10-25 | 2019-03-08 | 广东智媒云图科技股份有限公司 | A kind of drawing bootstrap technique and equipment |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111008573A (en) * | 2019-11-15 | 2020-04-14 | 广东智媒云图科技股份有限公司 | Limb structure generation method and device, terminal equipment and readable storage medium |
| CN111008573B (en) * | 2019-11-15 | 2024-04-26 | 广东智媒云图科技股份有限公司 | Limb structure generation method and device, terminal equipment and readable storage medium |
| CN111061902A (en) * | 2019-12-12 | 2020-04-24 | 广东智媒云图科技股份有限公司 | Drawing method and device based on text semantic analysis and terminal equipment |
| CN111061902B (en) * | 2019-12-12 | 2023-12-19 | 广东智媒云图科技股份有限公司 | Drawing method and device based on text semantic analysis and terminal equipment |
| CN111612864A (en) * | 2020-04-27 | 2020-09-01 | 厦门盈趣科技股份有限公司 | Drawing method and system based on photo and image recognition |
| CN111612864B (en) * | 2020-04-27 | 2023-05-09 | 厦门盈趣科技股份有限公司 | Drawing method and system based on photo and image recognition |
| CN112148944A (en) * | 2020-09-22 | 2020-12-29 | 西京学院 | Manufacturing method, manufacturing system, terminal and storage medium of manual fiber picture |
| CN112837396A (en) * | 2021-01-29 | 2021-05-25 | 深圳市天耀创想网络科技有限公司 | Line draft generation method and device based on machine learning |
| CN112837396B (en) * | 2021-01-29 | 2024-05-07 | 深圳市天耀创想网络科技有限公司 | Line manuscript generation method and device based on machine learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110097616B (en) | 2023-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110097616A (en) | A kind of joint drawing method, device, terminal device and readable storage medium storing program for executing | |
| Baldassarre et al. | Deep koalarization: Image colorization using cnns and inception-resnet-v2 | |
| CN110378985B (en) | Animation drawing auxiliary creation method based on GAN | |
| Li et al. | SHREC’13 track: large scale sketch-based 3D shape retrieval | |
| CN107844795B (en) | Convolutional neural network feature extraction method based on principal component analysis | |
| US9449253B2 (en) | Learning painting styles for painterly rendering | |
| CN110222722A (en) | Interactive image stylization processing method, calculates equipment and storage medium at system | |
| CN106096542B (en) | Image video scene recognition method based on distance prediction information | |
| CN103942571B (en) | Graphic image sorting method based on genetic programming algorithm | |
| CN109785400B (en) | Silhouette image manufacturing method and device, electronic equipment and storage medium | |
| Carlucci et al. | $^ 2$ CO: Deep depth colorization | |
| CN103745201B (en) | A kind of program identification method and device | |
| CN108734520B (en) | Jade price evaluation method and device based on machine learning | |
| Zhou et al. | An interactive and generative approach for chinese shanshui painting document | |
| Zhang et al. | A comprehensive survey on computational aesthetic evaluation of visual art images: Metrics and challenges | |
| Sapp et al. | A Fast Data Collection and Augmentation Procedure for Object Recognition. | |
| Lu et al. | Sketch simplification based on conditional random field and least squares generative adversarial networks | |
| CN113838158B (en) | Image and video reconstruction method and device, terminal equipment and storage medium | |
| CN104978583B (en) | The recognition methods of figure action and device | |
| Wang et al. | CLAST: Contrastive learning for arbitrary style transfer | |
| He | Application of local color simulation method of landscape painting based on deep learning generative adversarial networks | |
| CN109859284B (en) | Dot-based drawing implementation method and system | |
| CN112529978B (en) | Man-machine interactive abstract picture generation method | |
| CN114638743A (en) | Image artistic style migration method, system, electronic device and storage medium | |
| CN109993816A (en) | Joint drawing method, device, terminal setting and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |