CN109241805A - A kind of image module recognizer - Google Patents
A kind of image module recognizer Download PDFInfo
- Publication number
- CN109241805A CN109241805A CN201810898214.3A CN201810898214A CN109241805A CN 109241805 A CN109241805 A CN 109241805A CN 201810898214 A CN201810898214 A CN 201810898214A CN 109241805 A CN109241805 A CN 109241805A
- Authority
- CN
- China
- Prior art keywords
- image
- module
- ranking
- connected component
- profile
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/1417—2D bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1452—Methods for optical code recognition including a method step for retrieval of the optical code detecting bar code edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Electromagnetism (AREA)
- General Health & Medical Sciences (AREA)
- Toxicology (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of image module recognizers, which comprises the following steps: the image preprocessing for first identifying needs；The image outline of connection is found in image after the pre-treatment；Calculate area shared by contoured interior and rating indication；Estimate the ranking of module outer profile occupied area in image connectivity body, and the characteristics of image of the intensity profile of extraction module, shape, the special graph for including；Occupied area ranking, the characteristics of image whether connected component meets module is verified in image connectivity body since the module outer profile estimated.The present invention by finding outer connected component and interior connected component in image, can quick determining module position, image module is identified, method is easy, applied widely.
Description
Technical field
The present invention relates to image identification technical fields, more particularly, to a kind of image module recognizer.
Background technique
With the fast development of internet, data increase at increment type, and traditional plaintext transmission mode becomes no longer suitable,
So people come up with other media modes to express data, such as entry, chip, bar code, two dimensional code etc..Institute of the present invention
The algorithm being related to is a kind of algorithm that can identify image module, wherein the positioning that typical application is exactly two dimensional code module is grabbed
It takes.Two dimensional code is big with its capacity, anti-interference strong, identifies the features such as convenient and efficient, has obtained extensively as a suitable carrier
Use.It is predicted according to the Ministry of Industry and Information Technology, with the development of related industry chain, road the year two thousand twenty, China two dimensional code market rule
Mould is up to 200,000,000,000 yuan.The main application of two dimensional code has at present: electronic business card, anti-fake looks into source, mobile branch at product promotion publicity
It pays, electronic bill etc..
What existing two-dimensional code identification method relied primarily on is that three mark frames are positioned on two dimensional code, navigates to mark
Other image algorithms are reused after frame to separate two dimensional code with background, are come out to extract entire two dimensional code, finally
The monochrome pixels point on two dimensional code calculate according to relevant coding rule and derives former data.This method is only applicable to
Certain types of two dimensional code can not effectively extract two dimensional code if replacing other kinds of two dimensional code.
Summary of the invention
The purpose of the present invention is to provide a kind of image module recognizers, it is intended to solve to replace different figures in a picture
As module, quickly the problem of identification positioning image module.
In order to solve the above technical problems, technical scheme is as follows:
The image module periphery of a kind of image module recognizer, required identification can constitute connected component.The method includes such as
Lower step:
S1: the image identified will be needed to pre-process；
S2: the image that pretreatment is obtained finds edge and profile using operator, determines profile relationship；
S3: to the profiles found out all in image, area shared by contoured interior is calculated separately, and carry out rating indication；
S4: estimating the ranking of module outer profile occupied area in image connectivity body, and the intensity profile of extraction module, shape,
The characteristics of image for the special graph for including；
S5: occupied area ranking, verify whether connected component meets mould in image connectivity body since the module outer profile estimated
The characteristics of image of block.
Wherein, the pretreatment of step S1 image includes:
S1.1: the image elder generation gray processing for needing to identify is handled first；
S1.2: grayscale image is filtered denoising；
S1.3: the image after filtering processing is subjected to binarization operation, obtains black white image；
S1.4: by black white image, etching operation is first carried out, then carry out expansive working.
The image for obtaining pretreatment in step S2 finds edge and profile using operator, determines profile relationship.
The module outer profile estimated in step S5 occupied area ranking front three in image connectivity body.
Whether step S5 occupied area ranking, verifies connected component in image connectivity body since the module outer profile estimated
Meet the characteristics of image of module, process includes:
S5.1: the connected component for estimating ranking meets the characteristics of image of module, then finds module, carries out subsequent processing, and algorithm finishes；
S5.2: the connected component for estimating ranking does not meet the characteristics of image of module, and continuation successively carries out connected component near the ranking
Verifying, meets modular character until finding some connected component, algorithm finishes；
S5.3: it is not found still after the identification of all connected components, which does not contain the module of required identification.
Preferably,
Compared with prior art, the beneficial effect of technical solution of the present invention is:
The present invention provides a kind of image module recognizers, remove identification module relative to conventional detection, this algorithm is using quickly
The mode of profile is found, method is easier, and profile processing means are selectively more, and the scope of application is wider.
Detailed description of the invention
Fig. 1 is picture module recognizer flow chart.
Specific embodiment
The following further describes the technical solution of the present invention with reference to the accompanying drawings and examples.
Referring to Fig. 1, the present invention provides a kind of image module recognizer, the image module periphery of required identification being capable of structure
At connected component the following steps are included:
Step 1 description will need the image identified to pre-process in Fig. 1, and pretreated process includes:
The image for needing to identify is subjected to gray processing processing, obtains the grayscale image of image.
Grayscale image is filtered denoising；Wherein the filtering method of gaussian filtering can be used in filtering mode, removes
Noise jamming.
Image after filtering processing is subjected to binarization operation, obtains black white image；Wherein big rule can be used in binaryzation
Two value-based algorithm such as method two-value.
By black white image, etching operation is first carried out, then carry out expansive working, which can be by some non-interconnected body regions
Become a section, that is, does not constitute connected component.Wherein difference can be set into corrosion and the coefficient of expansion, and parameter can optionally change.
Such as: corrosion uses the structural element of 3*3, and expands the structural element for choosing 5*5, convenient for interrupting the dry of image-type
It disturbs.
The image that step 2 obtains pretreatment in Fig. 1 finds edge and profile using operator, determines profile relationship；It is required
The modular peripheral of searching can constitute connected component, and have characteristic feature, find convenient for calculating, wherein operator can be used
The contour identifications operator such as canny.
Step 3 calculates separately area shared by contoured interior, and arranged to the profiles found out all in image in Fig. 1
Name label；Such as: profile C1 is found out by calculating, C2, C3, C4 ... area are respectively S1, and S2, S3, S4 ... are according to S
Size ranking, give C1, C2, C3, C4 ... mark corresponding ranking respectively.
Step 4 estimates the ranking of module outer profile occupied area in image connectivity body, and the gray scale of extraction module in Fig. 1
Distribution, shape, the characteristics of image such as special graph for including.Such as: when intelligent mobile terminal scanning recognition two dimensional code, two dimensional code mould
Block occupied area ranked first position, and two dimensional code module more presentation rectangles after scanned in the connection bulk area ranking of image
Or irregular quadrilateral, within the scope of black and white ratio meets some after two-value.
Step 5 occupied area ranking, verifies connected component in image connectivity body since the module outer profile estimated in Fig. 1
Whether the characteristics of image of module is met, wherein module outer profile occupied area ranking front three in image connectivity body, step 5 mistake
Journey includes:
The connected component that S5.1 estimates ranking meets the characteristics of image of module, then finds module, carries out subsequent processing, and algorithm finishes.
The connected component that S5.2 estimates ranking does not meet the characteristics of image of module, continue successively to connected component near the ranking into
Row verifying, meets modular character until finding some connected component, algorithm finishes.
S5.3 is not found still after all connected components identify, which does not contain the module of required identification.
Claims (5)
1. a kind of image module recognizer, which is characterized in that the image module periphery of required identification can constitute connected component, packet
Include following steps:
S1: the image identified will be needed to pre-process；
S2: the image that pretreatment is obtained finds edge and profile using operator, determines profile relationship；
S3: to the profiles found out all in image, area shared by contoured interior is calculated separately, and carry out rating indication；
S4: estimating the ranking of module outer profile occupied area in image connectivity body, and the intensity profile of extraction module, shape,
The characteristics of image for the special graph for including；
S5: occupied area ranking, verify whether connected component meets mould in image connectivity body since the module outer profile estimated
The characteristics of image of block.
2. a kind of image module recognizer according to claim 1, which is characterized in that step S1 includes:
S1.1: the image elder generation gray processing for needing to identify is handled first；
S1.2: grayscale image is filtered denoising；
S1.3: the image after filtering processing is subjected to binarization operation, obtains black white image；
S1.4: by black white image, etching operation is first carried out, then carry out expansive working.
3. a kind of image module recognizer according to claim 1, which is characterized in that in step S1.2, using Gauss
Filtering is filtered denoising.
4. a kind of image module recognizer according to claim 1, which is characterized in that module foreign steamer in the step S5
Exterior feature occupied area ranking front three in image connectivity body.
5. a kind of image module recognizer according to claim 1, which is characterized in that the step S5 includes:
S5.1: the connected component for estimating ranking meets the characteristics of image of module, then finds module, carries out subsequent processing, and algorithm finishes；
S5.2: the connected component for estimating ranking does not meet the characteristics of image of module, and continuation successively carries out connected component near the ranking
Verifying, meets modular character until finding some connected component, algorithm finishes；
S5.3: it is not found still after the identification of all connected components, which does not contain the module of required identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810898214.3A CN109241805A (en) | 2018-08-08 | 2018-08-08 | A kind of image module recognizer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810898214.3A CN109241805A (en) | 2018-08-08 | 2018-08-08 | A kind of image module recognizer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109241805A true CN109241805A (en) | 2019-01-18 |
Family
ID=65070838
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810898214.3A Pending CN109241805A (en) | 2018-08-08 | 2018-08-08 | A kind of image module recognizer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109241805A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103870790A (en) * | 2014-04-02 | 2014-06-18 | 胡建国 | Recognition method and device of two-dimensional bar code |
| US20150090796A1 (en) * | 2013-09-29 | 2015-04-02 | Founder Mobile Media Technology (Beijing) Co., Ltd. | Method and system for detecting a correction pattern in a qr code |
| CN107545207A (en) * | 2017-09-28 | 2018-01-05 | 云南电网有限责任公司电力科学研究院 | DM two-dimensional code identification methods and device based on image procossing |
-
2018
- 2018-08-08 CN CN201810898214.3A patent/CN109241805A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150090796A1 (en) * | 2013-09-29 | 2015-04-02 | Founder Mobile Media Technology (Beijing) Co., Ltd. | Method and system for detecting a correction pattern in a qr code |
| CN103870790A (en) * | 2014-04-02 | 2014-06-18 | 胡建国 | Recognition method and device of two-dimensional bar code |
| CN107545207A (en) * | 2017-09-28 | 2018-01-05 | 云南电网有限责任公司电力科学研究院 | DM two-dimensional code identification methods and device based on image procossing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shi et al. | Automatic license plate recognition system based on color image processing | |
| CN109002824B (en) | OpenCV-based building drawing label information detection method | |
| Sedighi et al. | A new and robust method for character segmentation and recognition in license plate images | |
| Paunwala et al. | A novel multiple license plate extraction technique for complex background in Indian traffic conditions | |
| CN104361336A (en) | Character recognition method for underwater video images | |
| CN110119741A (en) | A kind of card card image information recognition methods having powerful connections | |
| CN103593695A (en) | Method for positioning DPM two-dimension code area | |
| Su et al. | A learning framework for degraded document image binarization using Markov random field | |
| CN105844337A (en) | Intelligent garbage classification device | |
| CN105740883A (en) | Feather piece classification method in computer based on contour and texture | |
| CN111027544A (en) | MSER license plate positioning method and system based on visual saliency detection | |
| CN110276759A (en) | A kind of bad line defect diagnostic method of Mobile phone screen based on machine vision | |
| CN104616019A (en) | Identification method for electronic equipment signboard image | |
| CN106446920B (en) | A kind of stroke width transform method based on gradient amplitude constraint | |
| Gaceb et al. | Adaptative smart-binarization method: For images of business documents | |
| Grover et al. | Text extraction from document images using edge information | |
| Abderaouf et al. | License plate character segmentation based on horizontal projection and connected component analysis | |
| Saddami et al. | Improvement of binarization performance using local otsu thresholding. | |
| CN109241805A (en) | A kind of image module recognizer | |
| Chowdhury et al. | Extracting and segmenting container name from container images | |
| CN105844260A (en) | Multifunctional smart cleaning robot apparatus | |
| CN105844651A (en) | Image analyzing apparatus | |
| CN109584212A (en) | A kind of SLM powder bed powder laying image scratch defect identification method based on MATLAB | |
| CN111753842B (en) | Method and device for detecting text region of bill | |
| CN112418123B (en) | Hough transformation-based engineering drawing line and line type identification method |
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: 20190118 |
|
| RJ01 | Rejection of invention patent application after publication |