CN109508628A - Font type detection method and system in font image based on convolutional neural networks - Google Patents
Font type detection method and system in font image based on convolutional neural networks Download PDFInfo
- Publication number
- CN109508628A CN109508628A CN201811107052.3A CN201811107052A CN109508628A CN 109508628 A CN109508628 A CN 109508628A CN 201811107052 A CN201811107052 A CN 201811107052A CN 109508628 A CN109508628 A CN 109508628A
- Authority
- CN
- China
- Prior art keywords
- font
- font type
- image
- background image
- type detection
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 81
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 40
- 238000000034 method Methods 0.000 claims description 14
- 238000012360 testing method Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 6
- 238000007689 inspection Methods 0.000 claims 1
- 238000013461 design Methods 0.000 description 6
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/333—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/32—Digital ink
- G06V30/36—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/28—Character recognition specially adapted to the type of the alphabet, e.g. Latin alphabet
- G06V30/287—Character recognition specially adapted to the type of the alphabet, e.g. Latin alphabet of Kanji, Hiragana or Katakana characters
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
Font type detection method in font image provided by the invention based on convolutional neural networks, comprising: several background images are grabbed by background image material website；The several font type of copyright has been obtained by the crawl of intellectual property website；Every kind of font type organic assembling in every background image and font type database is obtained into font image, by font image, font type and default font coordinate dimension information input into training pattern, training pattern is trained using convolutional neural networks algorithm and obtains font type detection model；Font image to be measured is input to font type detection model, font type detection model exports corresponding font type.Font type detection method in font image provided by the invention based on convolutional neural networks, survey font image progress font type identification is treated by establishing font type detection model, a large amount of font image to be measured can be quickly handled, saves cost of labor, working efficiency is significant.
Description
Technical field
The present invention relates to font information searching fields, more particularly to font kind in the font image based on convolutional neural networks
Class detection method and system.
Background technique
Chinese character style is according to a set of font made of certain rule, Style Design.The font is not mysterious, with a variety of loads
The form of body appears in human lives ubiquitously, such as: copybook, newspaper, books and periodicals etc..Not with society and technology
Disconnected development, since the esbablished corporations such as apple, Google release the customization font of oneself one after another, more and more famous brand names start
Obtain the exclusive font for authorizing or be customized development oneself.And font industry is also unprecedentedly paid close attention in design circle, either
Webpage design, e-book design, the always mostly important a part of text composition, font are the portion of core the most in typesetting again
Point, a kind of its software and plane design element as basis, is a part that designer does not walk around anyway.
As font is widely used, font abuse happens occasionally.Due to font there are many multiplicity application scenarios,
And different application scenarios determine different form of authorisation, resulting in user more or less is occurred passively
Abuse.For example, the Chinese font of Microsoft can at will use personal use；Business is used, in font
Having 9 kinds of fonts to need to obtain license could use, and otherwise be considered as infringement.Once infringement, user is not only condemned by law, and
And it needs to be compensated according to the expense of regulation.Dying young for huge reparation and product design achievement is faced, numerous businessmans are enabled
It suffers untold misery.Therefore important for the font type identification in font image, but the side of current font type identification
Method is substantially through artificial matching identification, and such identification method efficiency is slower, and the accuracy rate identified is relatively low.
Summary of the invention
For overcome the deficiencies in the prior art, one of the objects of the present invention is to provide the fonts based on convolutional neural networks
Font type detection method in image can solve current font type knowledge method for distinguishing and be substantially by manually comparing knowledge
Not, such identification method efficiency is slower, and the problem that the accuracy rate identified is relatively low.
The second object of the present invention is to provide font type detection system in the font image based on convolutional neural networks,
It, which can solve current font type, knows method for distinguishing and is substantially through artificial matching identification, such identification method efficiency compared with
And the relatively low problem of accuracy rate of identification slowly,.
The present invention provides the first purpose and is implemented with the following technical solutions:
Font type detection method in font image based on convolutional neural networks, comprising:
Background image data library is established, several background images are grabbed by background image material website, store the back
Scape image obtains background image data library；
Font type database is established, the several font type of copyright has been obtained by the crawl of intellectual property website, has been deposited
The font type is stored up, font type database is obtained；
Establish font sample database, by the background image data library every background image and the font
Every kind of font type in type database carries out organic assembling, according to default font coordinate dimension information in the font type
The corresponding font of the font type is edited in the corresponding background image, obtains font image；By the font image, institute
It states font type and the default font coordinate dimension information is stored into default sample database, obtain font image sample
Database, wherein every background image corresponds to a kind of font type；
Font type detection model is established, by the font image in the font image sample database, the word
Body type and the default font coordinate dimension information input are into training pattern, using convolutional neural networks algorithm to described
Training pattern is trained and obtains font type detection model；
Font image to be measured is obtained, the font image to be measured with font that user uploads is obtained；
Font type detection, is input to the font type detection model, the font kind for the font image to be measured
Class detection model exports the corresponding font type.
Further, the background image data library of establishing further includes carrying out gray proces to the background image.
Further, described is specially to use weighted mean method to the background to background image progress gray proces
Image carries out gray proces.
Further, described to establish font type detection model specifically: will be in the font image sample database
The font image, the font type and the default font coordinate dimension information input are into training pattern, using volume
Product neural network algorithm is trained the training pattern, described when the number of the training reaches preset times threshold value
Training stops, and obtains training pattern, will test, obtains in the test font image prestored the input training pattern
Font type is exported, the output font type is compared with corresponding test font type and has trained mould described in obtaining
The recognition accuracy of type, when the recognition accuracy reaches default discrimination threshold value, the training pattern is used as font kind
Class detection model.
Further, further include before font type detection using weighted mean method to the font image to be measured into
Row gray proces.
Further, the font type detection model the corresponding font type is exported to be sent to the user terminal.
The present invention provides the second purpose and is implemented with the following technical solutions:
Font type detection system in font image based on convolutional neural networks, comprising:
Establish background image data library module, the background image data library module of establishing is for by background image material
Website grabs several background images, stores the background image, obtains background image data library；
Establish font type database module, the font type database module of establishing is for by intellectual property website
Crawl has obtained the several font type of copyright, stores the font type, obtains font type database；
Font sample data library module is established, the font sample data library module of establishing is used for the background image number
Organic assembling is carried out according to every kind of font type in the every background image and the font type database in library, according to
It is corresponding that default font coordinate dimension information edits the font type in the corresponding background image of the font type
Font obtains font image；By the font image, the font type and the default font coordinate dimension information storage
Into default sample database, font image sample database is obtained, wherein every background image corresponds to a kind of word
Body type；
Font type detection model module is established, the font type detection model module of establishing is used for the fontmap
The font image, the font type and the default font coordinate dimension information input in decent database are extremely instructed
Practice in model, the training pattern is trained using convolutional neural networks algorithm and obtains font type detection model；
Obtain module, the font image to be measured with font for obtaining module and being used to obtain user's upload；
Font type detection module, the font type detection module are used to for the font image to be measured being input to described
Font type detection model, the font type detection model export the corresponding font type.
Further, the background image data library module of establishing is also used to carry out gray proces to the background image；
The background image data library module of establishing includes the first picking unit, the first storage unit and the first gray scale processing unit,
First picking unit is used to grab several background images by background image material website；The first gray proces list
Member for carrying out gray proces to the background image, carried on the back for storing the background image by first storage unit
Scape image data base.
Further, the font type database module of establishing includes the second picking unit and the second storage unit,
Second picking unit is used to obtain the several font type of copyright by the crawl of intellectual property website, and described second deposits
Storage unit obtains font type database for storing the font type.
Compared with prior art, the beneficial effects of the present invention are the font images of the invention based on convolutional neural networks
Middle font type detection method, comprising: several background images are grabbed by background image material website, store background image,
Obtain background image data library；The several font type that copyright has been obtained by the crawl of intellectual property website, stores font kind
Class obtains font type database；By in background image data library every background image with it is every in font type database
Kind font type carries out organic assembling, is edited in the corresponding background image of font type according to default font coordinate dimension information
The corresponding font of font type, obtains font image；Font image, font type and default font coordinate dimension information are deposited
Storage obtains font image sample database, wherein every background image corresponds to a kind of font kind into default sample database
Class；By font image, font type and the default font coordinate dimension information input in font image sample database to instruction
Practice in model, training pattern is trained using convolutional neural networks algorithm and obtains font type detection model；It obtains and uses
The font image to be measured with font that family uploads；Font image to be measured is input to font type detection model, font type
Detection model exports corresponding font type.Survey font image progress font type is treated by establishing font type detection model
Identification can quickly handle a large amount of font image to be measured, save cost of labor, be not necessarily to manual intervention, and working efficiency is significant.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And can be implemented in accordance with the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention and the accompanying drawings.
A specific embodiment of the invention is shown in detail by following embodiment and its attached drawing.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is the process signal of font type detection method in the font image of the invention based on convolutional neural networks
Figure；
Fig. 2 is the module architectures of font type detection system in the font image of the invention based on convolutional neural networks
Figure.
Specific embodiment
In the following, being described further in conjunction with attached drawing and specific embodiment to the present invention, it should be noted that not
Under the premise of conflicting, new implementation can be formed between various embodiments described below or between each technical characteristic in any combination
Example.
As shown in Figure 1, font type detection method in the font image of the invention based on convolutional neural networks, including with
Lower step:
Background image data library is established, several background images are grabbed by background image material website, store Background
Picture obtains background image data library；And gray proces are carried out to background image；The background image grabbed in the present embodiment is equal
For color image, since color image is made of multiple pixels, and each pixel is indicated by tri- values of RGB；According to
Weighted mean method carries out gray proces to background image, makes background image that black-white-gray state be presented, will not influence background image
Texture feature information, and each pixel only needs a gray value that can indicate, substantially increases background image processing effect
Rate；Background image after gray proces is stored in local server and obtains background image data library.
Font type database is established, the several font type of copyright has been obtained by the crawl of intellectual property website, has been deposited
Font type is stored up, font type database is obtained；Such as: the refined black equal font types of Microsoft.
Font sample database is established, it will be in the every background image and font type database in background image data library
Every kind of font type carry out organic assembling, according to default font coordinate dimension information in the corresponding background image of font type
The corresponding font of editing fonts type, obtains font image；Font image, font type and default font coordinate dimension are believed
Breath is stored into default sample database, obtains font image sample database, wherein every background image corresponds to a kind of font
Type.Such as editing fonts type is the refined black font of Microsoft on a kind of background image according to font type.
Font type detection model is established, by the font image in font image sample database, font type and pre-
If font coordinate dimension information input is trained and obtains to training pattern into training pattern, using convolutional neural networks algorithm
To font type detection model.In the present embodiment, specifically: by font image, the font in font image sample database
Type and default font coordinate dimension information input into training pattern, using convolutional neural networks algorithm to training pattern into
Row training, when trained number reaches preset times threshold value, training stops, and obtains training pattern, pre- in the present embodiment
If frequency threshold value takes 200,000 times；It will be tested in the test font image prestored input training pattern, obtain output font
Type will test in the test font image prestored input training pattern, obtain output font type, will export font
Type and corresponding test font type are compared and obtain the recognition accuracy of training pattern, in the present embodiment, identification
Accuracy rate is 80%；When recognition accuracy reaches default discrimination threshold value, training pattern is used as font type detection model.
When recognition accuracy not up to default discrimination threshold value, the font type information for training need to be readjusted.
Font image to be measured is obtained, the font image to be measured with font that user uploads is obtained；User in the present embodiment
Passing the image file with text on the subscriber terminal is font image to be measured；User terminal can be mobile phone, computer, touch-control
The hardware devices such as screen, notebook in the present embodiment, are also surveyed font image and carry out gray proces to being treated using weighted mean method.
Font type detection, is input to font type detection model for font image to be measured, font type detection model is defeated
Corresponding font type out.
Font type detection model is exported corresponding font type to be sent to the user terminal, user terminal is for showing
State font type.
As shown in Fig. 2, the present embodiment also provides font type detection system in the font image based on convolutional neural networks,
Include:
Background image data library module is established, establishes background image data library module for by background image material website
Several background images are grabbed, background image is stored, obtains background image data library；
Font type database module is established, establishes font type database module for grabbing by intellectual property website
The several font type of copyright has been obtained, font type has been stored, obtains font type database；
Font sample data library module is established, establishes font sample data library module for will be in background image data library
Every kind of font type in every background image and font type database carries out organic assembling, according to default font coordinate dimension
Information corresponding font of editing fonts type in the corresponding background image of font type, obtains font image；By font image,
Font type and default font coordinate dimension information are stored into default sample database, obtain font image sample data
Library, wherein every background image corresponds to a kind of font type；
Font type detection model module is established, font type detection model module is established and is used for font image sample number
According to font image, font type and the default font coordinate dimension information input in library into training pattern, using convolution mind
Training pattern is trained through network algorithm and obtains font type detection model；
Module is obtained, the font image to be measured with font that module is used to obtain user's upload is obtained；
Font type detection module, font type detection module are used to for font image to be measured to be input to font type detection
Model, font type detection model export corresponding font type.
Background image data library module is established in the present embodiment to be also used to carry out gray proces to background image；Establish background
Image database module includes the first picking unit, the first storage unit and the first gray scale processing unit, the first picking unit
For grabbing several background images by background image material website；First gray scale processing unit is used to carry out background image
Gray proces, the first storage unit obtain background image data library for storing background image.Establish font type database mould
Block includes the second picking unit and the second storage unit, and the second picking unit is used to obtain by the crawl of intellectual property website
The several font type of copyright, the second storage unit obtain font type database for storing font type.
Font type detection method in font image based on convolutional neural networks of the invention, comprising: pass through Background
Pixel material website grabs several background images, stores background image, obtains background image data library；Pass through intellectual property website
Crawl has obtained the several font type of copyright, stores font type, obtains font type database；By background image data
Every kind of font type in every background image and font type database in library carries out organic assembling, is sat according to default font
Dimensioning information corresponding font of editing fonts type in the corresponding background image of font type, obtains font image；By word
Body image, font type and default font coordinate dimension information are stored into default sample database, and it is decent to obtain fontmap
Database, wherein every background image corresponds to a kind of font type；By in font image sample database font image,
Font type and default font coordinate dimension information input are into training pattern, using convolutional neural networks algorithm to training mould
Type is trained and obtains font type detection model；Obtain the font image to be measured with font that user uploads；It will be to be measured
Font image is input to font type detection model, and font type detection model exports corresponding font type.By establishing word
Body species detection model, which is treated, surveys font image progress font type identification, can quickly handle a large amount of font image to be measured,
Cost of labor is saved, manual intervention is not necessarily to, working efficiency is significant.
More than, only presently preferred embodiments of the present invention is not intended to limit the present invention in any form；All current rows
The those of ordinary skill of industry can be shown in by specification attached drawing and above and swimmingly implement the present invention；But all to be familiar with sheet special
The technical staff of industry without departing from the scope of the present invention, is made a little using disclosed above technology contents
The equivalent variations of variation, modification and evolution is equivalent embodiment of the invention；Meanwhile all substantial technologicals according to the present invention
The variation, modification and evolution etc. of any equivalent variations to the above embodiments, still fall within technical solution of the present invention
Within protection scope.
Claims (9)
1. font type detection method in the font image based on convolutional neural networks, characterized by comprising:
Background image data library is established, several background images are grabbed by background image material website, store the Background
Picture obtains background image data library；
Font type database is established, the several font type of copyright has been obtained by the crawl of intellectual property website, has stored institute
Font type is stated, font type database is obtained；
Font sample database is established, by the every background image and the font type in the background image data library
Every kind of font type in database carries out organic assembling, corresponding in the font type according to default font coordinate dimension information
The background image in edit the corresponding font of the font type, obtain font image；By the font image, the word
Body type and the default font coordinate dimension information are stored into default sample database, obtain font image sample data
Library, wherein every background image corresponds to a kind of font type；
Font type detection model is established, by the font image, the font kind in the font image sample database
Class and the default font coordinate dimension information input are into training pattern, using convolutional neural networks algorithm to the training
Model is trained and obtains font type detection model；
Font image to be measured is obtained, the font image to be measured with font that user uploads is obtained；
The font image to be measured is input to the font type detection model, the font type inspection by font type detection
It surveys model and exports the corresponding font type.
2. font type detection method in the font image based on convolutional neural networks, feature exist as described in claim 1
In: the background image data library of establishing further includes carrying out gray proces to the background image.
3. font type detection method in the font image based on convolutional neural networks, feature exist as claimed in claim 2
In: described is specially that weighted mean method is used to carry out at gray scale the background image to background image progress gray proces
Reason.
4. font type detection method in the font image based on convolutional neural networks, feature exist as described in claim 1
In: it is described to establish font type detection model specifically: by the font image in the font image sample database, institute
Font type and the default font coordinate dimension information input are stated into training pattern, using convolutional neural networks algorithm pair
The training pattern is trained, and when the number of the training reaches preset times threshold value, the training stops, and has been instructed
Practice model, will be tested in the test font image prestored the input training pattern, output font type is obtained, by institute
It states output font type and the recognition accuracy of the training pattern is compared and obtained with corresponding test font type, when
When the recognition accuracy reaches default discrimination threshold value, the training pattern is used as font type detection model.
5. font type detection method in the font image based on convolutional neural networks, feature exist as described in claim 1
In: it further include that gray proces are carried out to the font image to be measured using weighted mean method before the font type detection.
6. font type detection method in the font image based on convolutional neural networks, feature exist as described in claim 1
In: the font type detection model is exported into the corresponding font type and is sent to the user terminal.
7. font type detection system in the font image based on convolutional neural networks, characterized by comprising:
Establish background image data library module, the background image data library module of establishing is for by background image material website
Several background images are grabbed, the background image is stored, obtain background image data library；
Font type database module is established, the font type database module of establishing by intellectual property website for grabbing
The several font type for having obtained copyright, stores the font type, obtains font type database；
Font sample data library module is established, the font sample data library module of establishing is used for the background image data library
In every background image and the font type database in every kind of font type carry out organic assembling, according to default
Font coordinate dimension information edits the corresponding font of the font type in the corresponding background image of the font type,
Obtain font image；The font image, the font type and the default font coordinate dimension information are stored to pre-
If in sample database, obtaining font image sample database, wherein every background image corresponds to a kind of font kind
Class；
Font type detection model module is established, the font type detection model module of establishing is for decent by the fontmap
The font image, the font type and the default font coordinate dimension information input in database are to training mould
In type, the training pattern is trained using convolutional neural networks algorithm and obtains font type detection model；
Obtain module, the font image to be measured with font for obtaining module and being used to obtain user's upload；
Font type detection module, the font type detection module are used to the font image to be measured being input to the font
Species detection model, the font type detection model export the corresponding font type.
8. font type detection system in the font image based on convolutional neural networks, feature exist as claimed in claim 7
In: the background image data library module of establishing is also used to carry out gray proces to the background image；It is described to establish Background
As database module includes the first picking unit, the first storage unit and the first gray scale processing unit, the first crawl list
Member is for grabbing several background images by background image material website；First gray scale processing unit is used for the back
Scape image carries out gray proces, and first storage unit obtains background image data library for storing the background image.
9. font type detection system in the font image based on convolutional neural networks, feature exist as claimed in claim 7
In: the font type database module of establishing includes the second picking unit and the second storage unit, and second crawl is single
Member is for having obtained the several font type of copyright by the crawl of intellectual property website, and second storage unit is for storing
The font type obtains font type database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811107052.3A CN109508628A (en) | 2018-09-21 | 2018-09-21 | Font type detection method and system in font image based on convolutional neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811107052.3A CN109508628A (en) | 2018-09-21 | 2018-09-21 | Font type detection method and system in font image based on convolutional neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109508628A true CN109508628A (en) | 2019-03-22 |
Family
ID=65746127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811107052.3A Pending CN109508628A (en) | 2018-09-21 | 2018-09-21 | Font type detection method and system in font image based on convolutional neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109508628A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978078A (en) * | 2019-04-10 | 2019-07-05 | 厦门元印信息科技有限公司 | Font copyright detection method, medium, computer equipment and device |
| CN110110982A (en) * | 2019-04-26 | 2019-08-09 | 特赞（上海）信息科技有限公司 | The checking method and device of intention material |
| CN111401362A (en) * | 2020-03-06 | 2020-07-10 | 上海眼控科技股份有限公司 | Tampering detection method, device, equipment and storage medium for vehicle VIN code |
| CN111814428A (en) * | 2020-06-29 | 2020-10-23 | 远光软件股份有限公司 | Method, device, terminal and storage medium for detecting font copyright information |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117740A (en) * | 2015-08-21 | 2015-12-02 | 北京旷视科技有限公司 | Font identification method and device |
| US9501724B1 (en) * | 2015-06-09 | 2016-11-22 | Adobe Systems Incorporated | Font recognition and font similarity learning using a deep neural network |
| US20170098138A1 (en) * | 2015-10-06 | 2017-04-06 | Adobe Systems Incorporated | Font Attributes for Font Recognition and Similarity |
| CN107784316A (en) * | 2016-08-26 | 2018-03-09 | 阿里巴巴集团控股有限公司 | A kind of image-recognizing method, device, system and computing device |
| CN108376244A (en) * | 2018-02-02 | 2018-08-07 | 北京大学 | A kind of recognition methods of text font in natural scene picture |
-
2018
- 2018-09-21 CN CN201811107052.3A patent/CN109508628A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9501724B1 (en) * | 2015-06-09 | 2016-11-22 | Adobe Systems Incorporated | Font recognition and font similarity learning using a deep neural network |
| CN105117740A (en) * | 2015-08-21 | 2015-12-02 | 北京旷视科技有限公司 | Font identification method and device |
| US20170098138A1 (en) * | 2015-10-06 | 2017-04-06 | Adobe Systems Incorporated | Font Attributes for Font Recognition and Similarity |
| CN107784316A (en) * | 2016-08-26 | 2018-03-09 | 阿里巴巴集团控股有限公司 | A kind of image-recognizing method, device, system and computing device |
| CN108376244A (en) * | 2018-02-02 | 2018-08-07 | 北京大学 | A kind of recognition methods of text font in natural scene picture |
Non-Patent Citations (1)
| Title |
|---|
| 高志强: "《深度学习：从入门到实战》", 30 June 2018, 中国铁道出版社 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978078A (en) * | 2019-04-10 | 2019-07-05 | 厦门元印信息科技有限公司 | Font copyright detection method, medium, computer equipment and device |
| CN109978078B (en) * | 2019-04-10 | 2022-03-18 | 厦门元印信息科技有限公司 | Font copyright detection method, medium, computer equipment and device |
| CN110110982A (en) * | 2019-04-26 | 2019-08-09 | 特赞（上海）信息科技有限公司 | The checking method and device of intention material |
| CN111401362A (en) * | 2020-03-06 | 2020-07-10 | 上海眼控科技股份有限公司 | Tampering detection method, device, equipment and storage medium for vehicle VIN code |
| CN111814428A (en) * | 2020-06-29 | 2020-10-23 | 远光软件股份有限公司 | Method, device, terminal and storage medium for detecting font copyright information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109508628A (en) | Font type detection method and system in font image based on convolutional neural networks | |
| CN105740402B (en) | The acquisition methods and device of the semantic label of digital picture | |
| CN110363084A (en) | A kind of class state detection method, device, storage medium and electronics | |
| CN110991171B (en) | Sensitive word detection method and device | |
| CN111061933A (en) | Picture sample library construction method and device, readable storage medium and terminal equipment | |
| CN110111902B (en) | Acute infectious disease attack period prediction method, device and storage medium | |
| WO2020228283A1 (en) | Feature extraction method and apparatus, and computer readable storage medium | |
| CN109117857B (en) | Biological attribute identification method, device and equipment | |
| CN112131881B (en) | Information extraction method and device, electronic equipment and storage medium | |
| CN107169002A (en) | A kind of personalized interface method for pushing and device recognized based on face | |
| CN110163204A (en) | Businessman's monitoring and managing method, device and storage medium based on image recognition | |
| CN110781925B (en) | Software page classification method and device, electronic equipment and storage medium | |
| Shete et al. | TasselGAN: An application of the generative adversarial model for creating field-based maize tassel data | |
| CN107798622A (en) | A kind of method and apparatus for identifying user view | |
| CN110516251B (en) | Method, device, equipment and medium for constructing electronic commerce entity identification model | |
| CN114818710A (en) | Form information extraction method, device, equipment and medium | |
| US20230004595A1 (en) | Image retrieval method, image retrieval devices, image retrieval system and image display system | |
| CN115357699A (en) | Text extraction method, device, equipment and storage medium | |
| CN107168635A (en) | Information demonstrating method and device | |
| CN113239063B (en) | Graph database updating method and device | |
| CN116663495B (en) | Text standardization processing method, device, equipment and medium | |
| CN112487165A (en) | Question and answer method, device and medium based on keywords | |
| CN112839185A (en) | Method, apparatus, device and medium for processing image | |
| CN107908749A (en) | A kind of personage's searching system and method based on search engine | |
| Wham et al. | Measuring perceptual distance of organismal color pattern using the features of deep neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190322 |
|
| RJ01 | Rejection of invention patent application after publication |