CN111199224A - Curved character recognition method and device - Google Patents
Curved character recognition method and device Download PDFInfo
- Publication number
- CN111199224A CN111199224A CN201811379524.0A CN201811379524A CN111199224A CN 111199224 A CN111199224 A CN 111199224A CN 201811379524 A CN201811379524 A CN 201811379524A CN 111199224 A CN111199224 A CN 111199224A
- Authority
- CN
- China
- Prior art keywords
- word
- character
- curved
- text
- recognition
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/245—Aligning, centring, orientation detection or correction of the image by locating a pattern; Special marks for positioning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Abstract
The disclosure provides a method and a device for recognizing bent characters, and relates to the field of character recognition. Acquiring character frames in an image to be detected and words corresponding to the character frames, detecting whether the words are bent characters according to the angle difference between the adjacent character frames in the words, inserting blank spaces between the character frames of the bent characters, and inputting the processed bent characters into a character recognition model for character recognition. Thus, detection and recognition of the bent characters are realized.
Description
Technical Field
The present disclosure relates to the field of text recognition, and in particular, to a method and an apparatus for recognizing curved text.
Background
In the field of artificial intelligence at present, a single shot multi-box detection (SSD) method can only detect horizontal characters, and an extended Seglink method can only detect oblique characters on the same straight line. The related art can not detect the bent characters.
Disclosure of Invention
The present disclosure proposes a scheme capable of detecting and recognizing curved characters.
Some embodiments of the present disclosure provide a curved word recognition method, including:
acquiring character frames in an image to be detected and words corresponding to the character frames;
detecting whether the word is a bent character or not according to the angle difference between adjacent character frames in the word;
inserting a space between character frames of the bent characters;
and inputting the processed bent characters into a character recognition model for character recognition.
In some embodiments, the text box in the image to be detected is obtained by inputting the image to be detected into a convolutional neural network CNN algorithm,
the convolutional neural network algorithm is trained by utilizing character samples in advance.
In some embodiments, the words corresponding to each text box are obtained by entering each text box into a depth first search DFS algorithm.
In some embodiments, a word is determined to be curved text if the difference in angles between adjacent text boxes in the word is between a minimum threshold and a maximum threshold.
In some embodiments, a word is determined to be non-curved text if the difference in angle between adjacent text boxes in the word is less than or equal to a minimum threshold;
if the angular difference between adjacent text boxes in a word is greater than or equal to a maximum threshold, the word is split.
In some embodiments, further comprising:
and inputting the angle mean value information of each character frame in the non-bent characters and the non-bent characters into a character recognition model for character recognition.
In some embodiments, the word recognition model is a join-sense time-classification CTC word recognition model.
In some embodiments, a graph model is built with each text box as a node, and the DFS algorithm is used to find connected components from the graph model, wherein each connected component is a word.
Some embodiments of the present disclosure provide a curved word recognition device, including:
a memory; and
a processor coupled to the memory, the processor configured to perform the curved-word recognition method of any of the preceding embodiments based on instructions stored in the memory.
Some embodiments of the present disclosure propose a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the curved text recognition method of any of the preceding embodiments.
Drawings
The drawings that will be used in the description of the embodiments or the related art will be briefly described below. The present disclosure will be more clearly understood from the following detailed description, which proceeds with reference to the accompanying drawings,
it is to be understood that the drawings in the following description are merely exemplary of the disclosure, and that other drawings may be derived from those drawings by one of ordinary skill in the art without undue inventive faculty.
Fig. 1 is a flow chart of a curved text recognition method according to some embodiments of the present disclosure.
Fig. 2 is a schematic structural diagram of a curved text recognition device according to some embodiments of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure.
Fig. 1 is a flow chart of a curved text recognition method according to some embodiments of the present disclosure.
As shown in fig. 1, the method of this embodiment includes:
s110, a text box (segment, which can be abbreviated as seg) in the image to be detected is obtained.
Here, the text box is also called a "segment", and is a bounding box covering a part of a word.
In some embodiments, the image to be detected is input into a Convolutional Neural Network (CNN) algorithm, and a text box in the image to be detected is output. The convolutional neural network algorithm is trained by utilizing character samples in advance. The description parameters of the text box include, for example, (x, y, w, h, θ), where (x, y) is the position coordinate, (w, h) is the width and height, and the angle is θ.
S120, obtaining the word (set as word) corresponding to each character frame.
In some embodiments, the information for each text box is input into a Depth-First Search (DFS) algorithm to obtain the corresponding word for each text box.
Specifically, a graph model is established by taking each text box as a node, and connected components are found from the graph model by using a DFS algorithm, wherein each connected component is a word.
S130, detecting whether the word is a bent character or not according to the angle difference between the adjacent character frames in the word.
The detection rules are for example:
and comparing the angle difference between the adjacent text frames with a preset minimum threshold value and a preset maximum threshold value, and judging according to the comparison result.
And if the angle difference between the adjacent character frames in the word is between the minimum threshold value and the maximum threshold value, the adjacent character frames are shown to be on the curve, and the word is judged to be the bent character.
If the angle difference between adjacent text boxes in the word is less than or equal to the minimum threshold, it indicates that the adjacent text boxes are on a straight line, and the word is determined as a non-curved text, which may be a horizontal text or an inclined text on the same straight line, for example.
And if the angle difference between the adjacent text boxes in the word is larger than or equal to the maximum threshold value, the two text boxes belong to different words, and the word is split.
And S140, inserting a space between character frames of the bent characters.
And S150, inputting the processed bent characters into a character recognition model for character recognition.
In some embodiments, the word recognition model is a connected semantic temporal Classification (CTC) word recognition model.
And S160, inputting the angle mean value information of each character frame in the non-bent characters and the non-bent characters into a character recognition model for character recognition.
In some embodiments, the word recognition model is a CTC word recognition model.
The embodiment realizes the detection and the identification of the bent characters, and can be applied to the detection of trademark identifications, advertisements, artistic characters and other bent characters.
The above scheme is described algorithmically below.
Firstly, a text box (x, y, w, h, theta) in an image to be detected is obtained through a CNN algorithm.
Then, the word corresponding to each text box is obtained by using the DFS algorithm, and a word consisting of n text boxes can be represented as follows:
word=(seg1,seg2,seg3,……,segi,……,segn)
let each count variable i be 0, j be 1, and k be 0 in the loop.
Let k be k +1, i be i +1, if i < n holds, then the following operations 1-3 are performed in a loop:
1)wordi[k]=segi,curvej=false,sum_θj=θi
wherein, curve represents the mark of the bending character, if false, the description is not the bending character, if true, the description is the bending character, thetaiPresentation text box segiThe angle of (c).
2) Calculating the angle difference between adjacent character frames in the same connected component
diff_θi=|θi+1-θi|
3) Will make the angular difference diff _ thetaiComparison with a minimum threshold diff _ min and a maximum threshold diff _ max:
3-1) if diff _ θi≤diffMin, which shows that the adjacent text boxes are on the straight line, and calculating the angle average value aver of the text boxesj:sum_θj=sum_θj+θi+1,averj=sum_θj/k;
3-2) if diff _ min < diff _ θi< diff _ max, indicating that the adjacent text box is on the curve, curvejTrue, text box segiAnd segi+1Inserting a space between the two plates;
3-3) if diff _ θiAnd the two character boxes belong to different words, the words are split, j is equal to j +1, and k is equal to 0.
By performing operations 1-3) in a loop, each word can be obtainedj。
Finally, wordjInputting the CTC model for character recognition.
Fig. 2 is a schematic structural diagram of a curved text recognition device according to some embodiments of the present disclosure.
As shown in fig. 2, the apparatus 200 of this embodiment includes:
a memory 210 and a processor 220 coupled to the memory 210, the processor 220 being configured to perform the network performance monitoring method in any of the foregoing embodiments based on instructions stored in the memory 210.
Memory 210 may include, for example, system memory, fixed non-volatile storage media, and the like. The system memory stores, for example, an operating system, an application program, a Boot Loader (Boot Loader), and other programs.
Some embodiments of the present disclosure also provide a computer-readable storage medium on which a computer program is stored, which when executed by a processor, implements the curved word recognition method of any of the foregoing embodiments.
As will be appreciated by one skilled in the art, embodiments of the present disclosure may be provided as a method, system, or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only exemplary of the present disclosure and is not intended to limit the present disclosure, so that any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
Claims (10)
1. A curved word recognition method, comprising:
acquiring character frames in an image to be detected and words corresponding to the character frames;
detecting whether the word is a bent character or not according to the angle difference between adjacent character frames in the word;
inserting a space between character frames of the bent characters;
and inputting the processed bent characters into a character recognition model for character recognition.
2. The method of claim 1, wherein,
the text box in the image to be detected is obtained by inputting the image to be detected into a Convolutional Neural Network (CNN) algorithm,
the convolutional neural network algorithm is trained by utilizing character samples in advance.
3. The method of claim 1, wherein,
the words corresponding to each text box are obtained by inputting each text box into a depth-first search DFS algorithm.
4. The method of claim 1, wherein,
if the angle difference between adjacent text boxes in a word is between a minimum threshold and a maximum threshold, the word is determined to be a curved character.
5. The method of claim 4, wherein the first and second light sources are selected from the group consisting of a red light source, a green light source, and a blue light source,
if the angle difference between adjacent text boxes in the word is less than or equal to a minimum threshold, the word is determined to be non-curved text;
if the angular difference between adjacent text boxes in a word is greater than or equal to a maximum threshold, the word is split.
6. The method of claim 5, further comprising:
and inputting the angle mean value information of each character frame in the non-bent characters and the non-bent characters into a character recognition model for character recognition.
7. The method of claim 1 or 6,
the character recognition model is a joint meaning time classification CTC character recognition model.
8. The method of claim 3, wherein,
and establishing a graph model by taking each text box as a node, and finding connected components from the graph model by using a DFS algorithm, wherein each connected component is a word.
9. A curved text recognition device comprising:
a memory; and
a processor coupled to the memory, the processor configured to perform the curved-word recognition method of any one of claims 1-7 based on instructions stored in the memory.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the curved word recognition method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811379524.0A CN111199224B (en) | 2018-11-20 | 2018-11-20 | Method and device for recognizing curved characters |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811379524.0A CN111199224B (en) | 2018-11-20 | 2018-11-20 | Method and device for recognizing curved characters |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111199224A true CN111199224A (en) | 2020-05-26 |
CN111199224B CN111199224B (en) | 2023-06-23 |
Family
ID=70745695
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811379524.0A Active CN111199224B (en) | 2018-11-20 | 2018-11-20 | Method and device for recognizing curved characters |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111199224B (en) |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1131301A (en) * | 1995-03-13 | 1996-09-18 | 财团法人工业技术研究院 | Word cutting method |
EP0905643A2 (en) * | 1997-09-29 | 1999-03-31 | Xerox Corporation | Method and system for recognizing handwritten words |
JPH11353415A (en) * | 1999-05-31 | 1999-12-24 | Fujitsu Ltd | Image extracting device |
US6188790B1 (en) * | 1996-02-29 | 2001-02-13 | Tottori Sanyo Electric Ltd. | Method and apparatus for pre-recognition character processing |
JP2007316754A (en) * | 2006-05-23 | 2007-12-06 | Canon Inc | Handwritten character processing device and method |
US20090016606A1 (en) * | 2005-06-02 | 2009-01-15 | Lumex As | Method, system, digital camera and asic for geometric image transformation based on text line searching |
CN101408937A (en) * | 2008-11-07 | 2009-04-15 | 东莞市微模式软件有限公司 | Method and apparatus for locating character row |
US20090097750A1 (en) * | 2007-10-16 | 2009-04-16 | Canon Kabushiki Kaisha | Image processing apparatus |
CN105809164A (en) * | 2016-03-11 | 2016-07-27 | 北京旷视科技有限公司 | Character identification method and device |
JP2017161969A (en) * | 2016-03-07 | 2017-09-14 | 日本電気株式会社 | Character recognition device, method, and program |
US20170351931A1 (en) * | 2016-06-03 | 2017-12-07 | Adobe Systems Incorporated | System and method for straightening curved page content |
US20180018305A1 (en) * | 2015-02-05 | 2018-01-18 | Hewlett-Packard Development Company, L.P. | Character spacing adjustment of text columns |
US20180018774A1 (en) * | 2016-07-15 | 2018-01-18 | Abbyy Development Llc | Method and system for preparing text images for optical-character recognition |
-
2018
- 2018-11-20 CN CN201811379524.0A patent/CN111199224B/en active Active
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1131301A (en) * | 1995-03-13 | 1996-09-18 | 财团法人工业技术研究院 | Word cutting method |
US6188790B1 (en) * | 1996-02-29 | 2001-02-13 | Tottori Sanyo Electric Ltd. | Method and apparatus for pre-recognition character processing |
EP0905643A2 (en) * | 1997-09-29 | 1999-03-31 | Xerox Corporation | Method and system for recognizing handwritten words |
JPH11353415A (en) * | 1999-05-31 | 1999-12-24 | Fujitsu Ltd | Image extracting device |
US20090016606A1 (en) * | 2005-06-02 | 2009-01-15 | Lumex As | Method, system, digital camera and asic for geometric image transformation based on text line searching |
JP2007316754A (en) * | 2006-05-23 | 2007-12-06 | Canon Inc | Handwritten character processing device and method |
US20090097750A1 (en) * | 2007-10-16 | 2009-04-16 | Canon Kabushiki Kaisha | Image processing apparatus |
CN101408937A (en) * | 2008-11-07 | 2009-04-15 | 东莞市微模式软件有限公司 | Method and apparatus for locating character row |
US20180018305A1 (en) * | 2015-02-05 | 2018-01-18 | Hewlett-Packard Development Company, L.P. | Character spacing adjustment of text columns |
JP2017161969A (en) * | 2016-03-07 | 2017-09-14 | 日本電気株式会社 | Character recognition device, method, and program |
CN105809164A (en) * | 2016-03-11 | 2016-07-27 | 北京旷视科技有限公司 | Character identification method and device |
US20170351931A1 (en) * | 2016-06-03 | 2017-12-07 | Adobe Systems Incorporated | System and method for straightening curved page content |
US20180018774A1 (en) * | 2016-07-15 | 2018-01-18 | Abbyy Development Llc | Method and system for preparing text images for optical-character recognition |
Non-Patent Citations (2)
Title |
---|
LIU YANG: "synthetically supervised feature learning for scene text recognition" * |
LIU,YULIANG: "deep matching prior network:toward tighter multi-oriented text detection" * |
Also Published As
Publication number | Publication date |
---|---|
CN111199224B (en) | 2023-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111327945B (en) | Method and apparatus for segmenting video | |
CN110996169B (en) | Method, device, electronic equipment and computer-readable storage medium for clipping video | |
US9852511B2 (en) | Systems and methods for tracking and detecting a target object | |
US9959466B2 (en) | Object tracking apparatus and method and camera | |
US20150248592A1 (en) | Method and device for identifying target object in image | |
US20140369555A1 (en) | Tracker assisted image capture | |
CN107423306B (en) | Image retrieval method and device | |
CN109919209A (en) | A kind of domain-adaptive deep learning method and readable storage medium storing program for executing | |
CN113850300A (en) | Method and device for training classification model | |
EP3043315B1 (en) | Method and apparatus for generating superpixels for multi-view images | |
WO2022033015A1 (en) | Method and apparatus for processing abnormal region in image, and image segmentation method and apparatus | |
CN107948721B (en) | Method and device for pushing information | |
US9082045B2 (en) | Method and apparatus for matching key pixels, and method and apparatus for matching images | |
CN111199224A (en) | Curved character recognition method and device | |
CN112084855A (en) | Outlier elimination method for video stream based on improved RANSAC method | |
CN109948605B (en) | Picture enhancement method and device for small target | |
CN108154521B (en) | Moving target detection method based on target block fusion | |
CN111667419A (en) | Moving target ghost eliminating method and system based on Vibe algorithm | |
CN113837067A (en) | Organ contour detection method and device, electronic equipment and readable storage medium | |
KR20180082680A (en) | Method for learning classifier and prediction classification apparatus using the same | |
CN108038863B (en) | Image segmentation method and device | |
JP2013037658A (en) | Edge detection device and program thereof | |
CN112884730A (en) | Collaborative significance object detection method and system based on collaborative learning | |
CN106202130A (en) | A kind of method and device of shot segmentation | |
CN110969210A (en) | Small and slow target identification and classification method, device, equipment and 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 |