CN106845485A - A kind of cloudy quarter rubber label icon identifying system - Google Patents
A kind of cloudy quarter rubber label icon identifying system Download PDFInfo
- Publication number
- CN106845485A CN106845485A CN201611192024.7A CN201611192024A CN106845485A CN 106845485 A CN106845485 A CN 106845485A CN 201611192024 A CN201611192024 A CN 201611192024A CN 106845485 A CN106845485 A CN 106845485A
- Authority
- CN
- China
- Prior art keywords
- image
- icon
- region
- label
- boundary
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of cloudy quarter rubber label icon recognition methods, by image acquisition step Real-time Collection label gray level image;Gamma values set-up procedure carries out the adjustment of Gamma values to the gray level image for gathering first;Setting contrast step is used for weakening grey gamut pixels in image, the black gamut pixels of protrusion;Bianry image inversion step is above-mentioned gray level image to be converted into bianry image and is inverted;Image boundary object removal step is by the label edge removal in image boundary;And using removing of small regions step, minimum enclosed rectangle framing steps, confine envelope discriminating step, area filling step and midline extraction step.The method can realize the cloudy visual identity for carving rubber label icon, describe for further icon and provide technical support.
Description
Technical field
The present invention relates to industrial coating target identification technology field, the more particularly to cloudy identification for carving rubber label icon.
Background technology
The moon carves the description substantially still handwork at present of rubber label icon, and job order one, workload is big, work is strong
Degree is high, wastes time and energy, it is impossible to be adapted with fast-developing Production requirement, and workman contacts different coating for a long time, right
Healthy unfavorable, human cost also more and more higher, influences product competitiveness in the market in addition.Therefore, in order to improve cloudy quarter rubber
The production efficiency of label product, increases the performance of enterprises, and carving rubber label icon plotting unit correlation technique to the moon carries out research tool
There is important realistic meaning.For the moon carves rubber label icon plotting unit, its vision system is that guiding machine is marked
Board icon describe key, and vision system work top priority be label icon identification, label icon identification it is correct with
It is no to directly affect the success rate that follow-up label icon is described.
The content of the invention
The purpose of the present invention is:There is provided a kind of cloudy quarter rubber label icon recognition methods so that the moon carves rubber label icon
Plotting unit realizes the identification to label icon, describes for further icon and provides technical support.
The technical scheme is that:
A kind of cloudy quarter rubber label icon recognition methods, specifically includes following steps:
(1)Image acquisition step:Based on monocular vision Real-time Collection label gray level image, its camera depends on the distance of label
The edge of label close to image boundary, and should carry out light filling directly over label by disposing linear light sorurce in image is gathered.
(2)Gamma set-up procedures:Gray level image to gathering carries out the adjustment of Gamma values, and Gamma values are adjusted into maximum,
To strengthen the foreground part in image.
(3)Setting contrast step:Using linear contrast's method by setting contrast to maximum, to weaken ash in image
Gamut pixels, the black gamut pixels of protrusion.
(4)Bianry image inversion step:Above-mentioned gray level image is converted into bianry image and is inverted, in order to follow-up
Treatment.
(5)Image boundary object removal step:Edge coordinate is obtained by each connected region Contour extraction, then to being obtained
Each connected region edge coordinate for taking is differentiated to determine whether to be connected with image boundary that finally removal is connected with image boundary
Connected region.
(6)Removing of small regions step:By broken region removal of noise in image, to retain effective icon region.
(7)Minimum enclosed rectangle framing steps:Icon region in image is confined.
(8)Confine envelope discriminating step:Differentiate to confining the relation between rectangle angular coordinate, to distinguish those
Confine region and also confine region comprising less.
(9)Area filling step:According to above-mentioned differentiation result, for not comprising the smaller figure confined in region for confining frame
Symbol is directly filled;In for comprising the smaller filling for confining the icon in region for confining frame, being first filled with confining greatly frame
The region that icon external boundary is included, refill it is small confine the region that icon inner boundary in frame is included, then two regions subtract each other so as to
Realize the area filling between icon inner and outer boundary.
(10)Midline extraction step:Center line is extracted to the icon after filling, in order to subsequently track carries out machinery and retouches accordingly
Paint.
Technique effect of the invention is:For the moon carves rubber label icon plotting unit, the inventive method can be real
Existing identification of the equipment to label icon.
Brief description of the drawings
Fig. 1 carves rubber label icon identification process figure for cloudy.
Specific embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
A kind of cloudy flow for carving rubber label icon recognition methods of the present invention is as shown in figure 1, specifically include following steps:
(1)Image acquisition step
Monocular cam and linear light sorurce, its camera is disposed to be depended on the distance of label in collection image directly over label
The edge of label should gather label gray level image close to image boundary with this.
(2)Gamma set-up procedures
The amendment of Gamma values, thus it is possible to vary picture brightness, increases contrast.Gamma is carried out here by the gray level image to gathering
Value adjustment, is adjusted to maximum, to strengthen the foreground part in image by Gamma values.
(3)Setting contrast step
Using linear contrast's method by setting contrast to maximum, to weaken grey gamut pixels, prominent black domain picture in image
Element.
(4)Bianry image inversion step
Above-mentioned gray level image is converted into bianry image, now label is black picture element in image, and background is white pixel, is just
Inverted in subsequent treatment, white pixel will be converted to by label in image, background transitions are black picture element.
(5)Image boundary object removal step
Label edge belongs to interference region, it should be removed for icon therein identification in image.Schemed according to collection
The characteristics of as label edge close to image boundary, edge coordinate is obtained by each connected region Contour extraction first, then to institute
Each connected region edge coordinate for obtaining is differentiated to determine whether to be connected with image boundary, is finally removed and image boundary phase
Connected region even is label fringe region in image.
(6)Removing of small regions step
Because label itself is to reasons such as light catadioptrics, cause collection image that the broken point of noise can be produced in above-mentioned processing procedure
Region.For the icon identification in image, interference region is still fallen within, should be removed.By calculating each region picture
Vegetarian refreshments number, tests given threshold to be removed according to big data.
(7)Minimum enclosed rectangle framing steps
The icon region in image is marked by neighbourhood signatures' method, and is different values by different icon zone markers;
Calculate the maximum of number of pixels on the different directions of marked region horizontal vertical two;It is most with number of pixels in different marked regions
Ranks for length and width draw rectangle, this rectangle be minimum enclosed rectangle, icon region in image is confined with this,
And the coordinate of zoning rectangle each angle point.
(8)Confine envelope discriminating step
Differentiate to confining the relation between rectangle angular coordinate, confine region distinguishing those and also confined comprising less
Region.
(9)Area filling step
According to above-mentioned differentiation result, for not comprising the smaller icon confined in region for confining frame(Such as Z icons)Directly enter
Row filling;For comprising the smaller filling for confining icon in region for confining frame, such as O icons, in being first filled with confining greatly frame
The region that icon external boundary is included, i.e. icon O outer rings inclusion region, refill and small confine the area that icon inner boundary in frame is included
Domain, i.e. icon O inner rings inclusion region, then two regions subtract each other so as to realize the area filling between icon inner and outer boundary.
(10)Midline extraction step
Center line is extracted to the icon after filling, in order to subsequently track carries out machine description accordingly.
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention.It is all in essence of the invention
Any modification, equivalent and improvement made within god and principle etc., should be included within the scope of the present invention.
Claims (7)
1. a kind of cloudy quarter rubber label icon recognition methods, it is characterised in that comprise the following steps:
(1)Image acquisition step:Based on monocular vision Real-time Collection label gray level image;
(2)Gamma set-up procedures:Gray level image to gathering carries out the adjustment of Gamma values, the foreground part in enhancing image;
(3)Setting contrast step:For weakening grey gamut pixels in image, the black gamut pixels of protrusion;
(4)Bianry image inversion step:Above-mentioned gray level image is converted into bianry image and is inverted, in order to follow-up place
Reason;
(5)Image boundary object removal step:By the label edge removal in image boundary;
(6)Removing of small regions step:By broken region removal of noise in image, to retain effective icon region;
(7)Minimum enclosed rectangle framing steps:Icon region in image is confined;
(8)Confine envelope discriminating step:Determine whether also to confine region comprising smaller in confining region to each;
(9)Area filling step:It is straight for not comprising the smaller icon confined in region for confining frame according to above-mentioned differentiation result
Connect and be filled, for the region being filled only with comprising the smaller icon confined in region for confining frame between icon inner and outer boundary;
(10)Midline extraction step:Center line is extracted to the icon after filling, in order to subsequently track carries out mechanical description accordingly.
2. a kind of cloudy quarter rubber label icon recognition methods according to claim 1, it is characterised in that image acquisition step
Close to image boundary, and should just be gone up in label at the edge that middle monocular cam depends on label in collection image with the distance of label
Side carries out light filling by disposing linear light sorurce.
3. a kind of cloudy quarter rubber label icon recognition methods according to claim 1, it is characterised in that Gamma adjustment steps
Gamma values are adjusted to maximum in rapid.
4. a kind of cloudy quarter rubber label icon recognition methods according to claim 1, it is characterised in that setting contrast is walked
Adjusted to maximum using linear contrast's method in rapid.
5. a kind of cloudy quarter rubber label icon recognition methods according to claim 1, it is characterised in that image boundary object
Removal step obtains edge coordinate by each connected region Contour extraction, then acquired each connected region edge coordinate is entered
Go differentiation to determine whether to be connected with image boundary, finally remove the connected region being connected with image boundary.
6. a kind of cloudy quarter rubber label icon recognition methods according to claim 1, it is characterised in that confine envelope differentiation
Step differentiates by confining the relation between rectangle angular coordinate, and region is confined also comprising less to distinguish those
Confine region.
7. a kind of cloudy quarter rubber label icon recognition methods according to claim 1, it is characterised in that area filling step
For comprising the smaller filling for confining the icon in region for confining frame, being first filled with confining greatly what icon external boundary in frame was included
Region, refill it is small confine the region that icon inner boundary in frame is included, then two regions are subtracted each other so as to realize icon inner and outer boundary
Between area filling.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611192024.7A CN106845485A (en) | 2016-12-21 | 2016-12-21 | A kind of cloudy quarter rubber label icon identifying system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611192024.7A CN106845485A (en) | 2016-12-21 | 2016-12-21 | A kind of cloudy quarter rubber label icon identifying system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106845485A true CN106845485A (en) | 2017-06-13 |
Family
ID=59135187
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611192024.7A Pending CN106845485A (en) | 2016-12-21 | 2016-12-21 | A kind of cloudy quarter rubber label icon identifying system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106845485A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112614443A (en) * | 2020-12-24 | 2021-04-06 | 玺美车众数字传媒科技(上海)有限公司 | Intelligent lamp box capable of carrying out vehicle flow statistics and capturing and analyzing images of vehicle owner groups |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105196297A (en) * | 2015-09-25 | 2015-12-30 | 河南工业职业技术学院 | Intelligent icon painting system and working procedure thereof |
-
2016
- 2016-12-21 CN CN201611192024.7A patent/CN106845485A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105196297A (en) * | 2015-09-25 | 2015-12-30 | 河南工业职业技术学院 | Intelligent icon painting system and working procedure thereof |
Non-Patent Citations (2)
Title |
---|
T. ARRIGHI等: "Recognition and Classification of Numerical Labels Using Digital Image Processing Techniques", 《2012 XVII SYMPOSIUM OF IMAGE, SIGNAL PROCESSING, AND ARTIFICIAL VISION (STSIVA)》 * |
李建美: "标牌压印字符图像获取与处理中的关键技术研究", 《中国博士学位论文全文数据库(信息科技辑)》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112614443A (en) * | 2020-12-24 | 2021-04-06 | 玺美车众数字传媒科技(上海)有限公司 | Intelligent lamp box capable of carrying out vehicle flow statistics and capturing and analyzing images of vehicle owner groups |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108986094B (en) | Automatic data updating method for training image library for face recognition | |
CN112614062B (en) | Colony counting method, colony counting device and computer storage medium | |
CN104636701A (en) | Laser two-dimension code identification method based on image restoration | |
CN107909571A (en) | A kind of weld beam shape method, system, equipment and computer-readable storage medium | |
CN104331876A (en) | Line detection and image processing method and related devices thereof | |
CN108073932A (en) | A kind of workpiece image Feature extraction and recognition method based on Gabor filtering | |
CN103699900A (en) | Automatic batch extraction method for horizontal vector contour of building in satellite image | |
CN103218605A (en) | Quick eye locating method based on integral projection and edge detection | |
CN109426814A (en) | A kind of positioning of the specific plate of invoice picture, recognition methods, system, equipment | |
CN109886935A (en) | A kind of road face foreign matter detecting method based on deep learning | |
CN106127193B (en) | A kind of facial image recognition method | |
CN108009538A (en) | A kind of automobile engine cylinder-body sequence number intelligent identification Method | |
CN105447489B (en) | A kind of character of picture OCR identifying system and background adhesion noise cancellation method | |
CN105389574A (en) | Method and system for detecting human eye irises in pictures | |
CN108198417A (en) | A kind of road cruising inspection system based on unmanned plane | |
CN112991374B (en) | Canny algorithm-based edge enhancement method, canny algorithm-based edge enhancement device, canny algorithm-based edge enhancement equipment and storage medium | |
CN108038482A (en) | A kind of automobile engine cylinder-body sequence number Visual intelligent identifying system | |
CN104778707A (en) | Electrolytic capacitor detecting method for improving general Hough transform | |
CN1240021C (en) | Bill image processing equipment | |
CN107886096A (en) | A kind of insulator of transmission line of electricity comes off defect inspection method | |
CN103295009A (en) | License plate character recognition method based on stroke decomposition | |
CN110687122A (en) | Method and system for detecting surface cracks of ceramic tile | |
CN104715250A (en) | Cross laser detection method and device | |
CN107516315A (en) | A kind of development machine based on machine vision is slagged tap monitoring method | |
CN110443166A (en) | A kind of licence plate recognition method of haze weather |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170613 |