CN106874903A - The device and method that a kind of article profile is obtained and recognized - Google Patents
The device and method that a kind of article profile is obtained and recognized Download PDFInfo
- Publication number
- CN106874903A CN106874903A CN201710089340.XA CN201710089340A CN106874903A CN 106874903 A CN106874903 A CN 106874903A CN 201710089340 A CN201710089340 A CN 201710089340A CN 106874903 A CN106874903 A CN 106874903A
- Authority
- CN
- China
- Prior art keywords
- camera
- profile
- under test
- light
- object under
- 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
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/145—Illumination specially adapted for pattern recognition, e.g. using gratings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/147—Details of sensors, e.g. sensor lenses
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Vascular Medicine (AREA)
- Artificial Intelligence (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Sorting Of Articles (AREA)
Abstract
A kind of device and method for being obtained the invention discloses article profile and being recognized, described device includes light source, Fresnel Lenses, reflector, light source, camera and microprocessor；The described method comprises the following steps：1) light for being arranged in the injection of the light source near camera is shaped to directional light by Fresnel Lenses；2) directional light leaves profile shade corresponding with object under test by object under test on reflector；3) light that reflector is reflected back is converged by Fresnel Lenses, is captured to form image by camera；4) binaryzation, contour extraction processing are carried out to the image that camera is captured by microprocessor, obtains the profile of object under test；(5) profile is contrasted with profile Sample Storehouse, is recognized object under test.The present invention is especially suitable for the images of items identification in Packaging Bottle the like waste intelligence removal process, with simple structure, the characteristics of cheap.
Description
Technical field
The invention belongs to subject image identification technology field and discarded object intelligence recovery technology field, more particularly to one kind
The device and method that the small article such as Packaging Bottle profile is obtained and recognized.
Background technology
The intelligent retracting device for developing and promoting the rubbish such as waste plastic bottle is considered as the recovery system of specification Packaging Bottle, is subtracted
Few " white garbage " pollution it is maximally effective by way of one of.In the intelligent retracting device of Packaging Bottle, the identification technology of article is
Core component.Efficiently, the identification for preparing effectively preventing error can reclaim and prevent from reclaiming fraud, so as to greatly carry
The reliability of machine high.
One kind is disclosed in the utility model patent of Application No. 201220364458.1 and 200620092237.8 to be based on
The scheme being identified to the bar code scan of Packaging Bottle；But, it is also easily damaged because bar code is easily copied.Therefore, base is caused
The popularization of intelligent way of recycling is hindered in the RM of bar code.It is recoverable intelligence using integrated article means of identification
The image recognition technology of development trend, wherein article is one of technology of most critical.The hair of Application No. 201410670766.0
Bright patent application discloses a kind of not including by the use of camera capture and recovery capsule image comprising article and as base map and ratio
Relatively scheme, thus recognize article whether be recyclable plastics bottle scheme, but the accuracy of identification of the technology depends on the quality of image.
The application for a patent for invention of Application No. 200810200993.1 discloses a kind of tank body pattern recognition device, is taken the photograph by CCD camera
Take image and it is contrasted with image in database, but do not provide the ins and outs of realization.Application No.
200620092237.8 utility model patent disclose it is a kind of beverage bottle both sides place respectively light emitting source and light-sensitive element with
Judge the scheme of beverage bottle size, it is done so that the diameter of object can only be judged, it is difficult to the beverage bottle profile to putting into is entered
Row is accurately detected and judged.In addition, the research in terms of the profile image recognition of plastic bottle also has other some progress, such as
Machine vision method, RGB methods, image comparison method etc., above-mentioned technical sophistication, high cost, and also it is not applied to recovery really
In machine product.
As described above, in order to solve the above problems, and accelerate the popularization of intelligent recycling machine, it is accordingly required in particular to invention is a kind of high-precision
Degree, low cost, it is adaptable to the image of the small article such as Packaging Bottle, outline identification apparatus and method.
The content of the invention
It is an object of the invention in solving the problems, such as prior art and product, there is provided a kind of article profile is obtained and recognized
Device and method, high precision, low cost.
Technical scheme solved by the invention is：
The device for being obtained it is an aspect of the present invention to provide a kind of article profile and being recognized, including light source, Fresnel are saturating
Mirror, reflector, camera and microprocessor；
The light source is arranged near camera, and both are located at object under test the same side；The Fresnel lens is located at light
Between source and camera and object under test；The reflector is arranged on object under test opposite side, relative with light source and camera.
The camera and light source are arranged in the focal position of Fresnel lens.
The Fresnel Lenses is used to for the light that light source is projected to be shaped to directional light, and the light that reflector is reflected back
(containing object under test profile shade) is converged；
Described reflector is used to reflect the directional light for coming from Fresnel lens, can be by specular reflective mirrors or lens type
Reflective membrane (such as 3M engineering grades EGP reflective membranes, reflective paster) is constituted；
Described camera is used to capturing the light that reflector is reflected back and the image to be formed is converged via Fresnel lens；
The image that described microprocessor is used to export camera carries out gray processing and binary conversion treatment, and carries out side
Edge detection, contours extract and outline etc..
Method for distinguishing is obtained and known it is an aspect of the present invention to provide a kind of article profile, is specifically entered by the following method
OK：
1) light for being arranged in the injection of the light source near camera is shaped to directional light by Fresnel Lenses.
2) directional light leaves profile shade corresponding with object under test by object under test on reflector.
3) light (containing object under test profile shade) that reflector is reflected back is converged by Fresnel Lenses, by camera
Capture form image, the image of capture contains the profile shade of object under test, on image the profile shadow region of object under test it
Outer is uniform, light ground, has very strong contrast with the profile shade of object under test；
4) binaryzation and contour extraction processing are carried out to the image that camera is captured by microprocessor, obtains object under test
Profile；
5) profile is contrasted with profile Sample Storehouse, is recognized object under test.
Described rim detection be to the view data after binary conversion treatment, using Image Edge-Detection Laplace or
The operators such as Prewitt, extract data boundary.
Beneficial effect：
The device that a kind of article profile that the present invention is provided is obtained and recognized, is particularly well-suited to Packaging Bottle the like waste intelligence
Images of items identification in removal process, with simple structure, the characteristics of cheap, can be used as domestic and international similar image recognition
There are the potentiality for being applied to more identification fields in the substitute products of sensor, future.Particular technique feature show it is following some：
1) shortcoming that currently marketed Packaging Bottle intelligence retracting device means of identification depends on bar code scan is solved.It is logical
The contours extract to article and contrast are crossed, the recognition accuracy of article is favorably improved, the availability of intelligent retracting device is improved.
2) the utilization optical system of innovation solves the problems, such as contours extract and the matching of article, the cost of the technology
It is low.Other image recognition technologys are different from, equipment and mounting condition requirement of the present invention to image recognition is relatively low, such as only needs
Achieved the goal by taking common camera and light source.
3) present invention is also applied to the other field outside Packaging Bottle is intelligently reclaimed, such as product sorting, refuse classification
Deng field.
Brief description of the drawings
Fig. 1 is the device basic principle schematic that article profile of the invention is obtained and recognized；
Fig. 2 is that the article profile of one embodiment of the invention is obtained and recognition methods design sketch；Fig. 2 (a), (b), (c) point
Image, the image after binary conversion treatment and object under test profile that Wei be after gray processing treatment；
Fig. 3 is that the article profile of another embodiment of the present invention is obtained and recognition methods design sketch；Fig. 3 (a), (b), (c)
Image, the image after binary conversion treatment and object under test profile respectively after gray processing treatment.
Description of reference numerals:
In figure, 1 is microprocessor, 2 is camera, 3 is light source, 4 is Fresnel lens, 5 is object under test, 6 is reflective
Plate.
Specific embodiment
Fig. 1 is the device basic principle schematic that article profile of the invention is obtained and recognized.As shown in figure 1, described
Article profile obtain and identification device, it is including microprocessor 1, camera 2, light source 3, Fresnel lens 4, object under test 5, anti-
Tabula rasa 6.
Described microprocessor 1 is used for image gray processing, the binary conversion treatment for exporting camera 2, and carries out edge inspection
Survey, contours extract, outline etc..
Described camera 2 is used for the figure for capturing the reflector light being reflected back and formation being converged via Fresnel lens 4
Picture.The image of capture contains the profile shade of object under test, is uniform, light ground outside image shadow region, with article wheel
Wide shade has very strong contrast.The image of the capture of camera 2 is via data line transfer to microprocessor 1.
The light source 3 similar to spot light, for projecting light.
The Fresnel lens 4 is used to for the light that light source is projected to be shaped to directional light, and directional light passes through object under test,
Profile shade corresponding with object under test is left on reflector；In addition, Fresnel lens 4 is used for the light that reflector is reflected back again
Line (containing object under test profile shade) is converged.
The reflector 6 can be by specular reflective mirrors or lens type reflective membrane (such as 3M engineering grades EGP reflective membranes, reflection plaster
Paper etc.) composition, the directional light of Fresnel lens is come from for reflecting.
With reference to embodiment, the present invention is described in further detail.
Embodiment 1：
Obtained using article profile of the invention and recognition methods, in one embodiment, using specular reflective mirrors conduct
Reflector, with beverage bottle as object under test.Fig. 2 is the recognition effect in the present embodiment, is specifically carried out by the following method：
1) light for being arranged in the injection of the light source near camera is shaped to directional light by Fresnel Lenses；
2) directional light leaves profile shade corresponding with object under test by object under test on reflector；
3) light (containing object under test profile shade) that reflector (specular reflective mirrors) is reflected back is by Fresnel Lenses
Converge, captured to form image by camera, the image of capture contains the profile shade of object under test, the wheel of object under test on image
It is uniform, light ground outside wide shadow region, there is very strong contrast with the profile shade of object under test；
4) binaryzation and contour extraction processing are carried out to the image that camera is captured by microprocessor, obtains determinand
The profile of body；
5) profile is contrasted with profile Sample Storehouse, is recognized object under test.
In the present embodiment, reflector uses specular reflective mirrors.Now, background light intensity is higher, more uniform, background with treat
The image contrast for surveying object is big, it is easy to process.But the relative position requirement to each equipment in device is higher.Shown by Fig. 2, this reality
The Outside contour extraction effect for applying beverage bottle in example fully meets identification requirement.
Embodiment 2：
Obtained using article profile of the invention and recognition methods, in one embodiment, made using lens type reflective membrane
It is reflector, with beverage bottle as object under test.Fig. 3 is the recognition effect in the present embodiment, is specifically carried out by the following method：
1) light for being arranged in the injection of the light source near camera is shaped to directional light by Fresnel Lenses；
2) directional light leaves profile shade corresponding with object under test by object under test on reflector；
3) light (containing object under test profile shade) that reflector (lens type reflective membrane) is reflected back is saturating by Fresnel
Mirror is converged, and is captured to form image by camera, and the image of capture contains the profile shade of object under test, object under test on image
It is uniform, light ground outside profile shadow region, there is very strong contrast with the profile shade of object under test；
4) binaryzation and contour extraction processing are carried out to the image that camera is captured by microprocessor, obtains determinand
The profile of body；
5) profile is contrasted with profile Sample Storehouse, is recognized object under test.
In the present embodiment, reflector uses lens type reflective membrane.Now, the profile shade contrast of background and object under test
Relatively lower slightly, image procossing is increasingly complex, and carefully regulation threshold value is needed during image binaryzation treatment, can use maximum between-cluster variance
Method chooses optimal threshold；But the position accuracy demand to device all parts is relatively low.Shown by Fig. 3, beverage bottle in the present embodiment
Outside contour extraction effect fully meet identification require.
Above example is merely to thought of the invention described in more detail and operation principle, it should be understood that above-mentioned
Mentioning embodiment should not limit this.Every all changes and replacement within the spirit and scope of the present invention, all should be in this hair
The row of bright protection.
Claims (4)
1. the device that a kind of article profile is obtained and recognized, it is characterised in that：Including light source, Fresnel Lenses, reflector, light
Source, camera and microprocessor；
The light source is arranged near camera, and both are located at object under test the same side；The Fresnel lens be located at light source and
Between camera and object under test, camera and light source are arranged in the focal position of Fresnel lens；The reflector is arranged on
Object under test opposite side is relative with light source and camera；
Described Fresnel Lenses is used to for the light that light source is projected to be shaped to directional light, and the light that reflector is reflected back enters
Row is converged；
Described reflector is used to reflect the directional light for coming from Fresnel lens；
Described camera is used to capturing the light that reflector is reflected back and the image to be formed is converged via Fresnel lens；
Described microprocessor is used to carry out gray processing and binary conversion treatment to the image of camera output, and carries out edge inspection
Survey, contours extract and outline.
2. the device that a kind of article profile according to claim 1 is obtained and recognized, it is characterised in that：The reflector is adopted
With specular reflective mirrors or lens type reflective membrane.
3. a kind of installation method that article profile is obtained and recognized, it is characterised in that the one kind described in usage right requirement 1 or 2
The device that article profile is obtained and recognized, specifically step is carried out by the following method：
1) light for being arranged in the injection of the light source near camera is shaped to directional light by Fresnel Lenses；
2) directional light leaves profile shade corresponding with object under test by object under test on reflector；
3) light that reflector is reflected back is converged by Fresnel Lenses, is captured to form image by camera；
4) gray processing and binary conversion treatment are carried out to the image that camera is captured by microprocessor, and carries out rim detection and wheel
Exterior feature is extracted, and obtains the profile of object under test；
5) profile is contrasted with profile Sample Storehouse, is recognized object under test.
4. article profile according to claim 3 is obtained and recognition methods, it is characterised in that to the figure after binary conversion treatment
As data, rim detection is carried out using Laplace operators or Prewitt operators.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710089340.XA CN106874903A (en) | 2017-02-20 | 2017-02-20 | The device and method that a kind of article profile is obtained and recognized |
| CN201820003382.7U CN207690117U (en) | 2017-02-20 | 2018-01-02 | A kind of device that article profile is obtained and identified |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710089340.XA CN106874903A (en) | 2017-02-20 | 2017-02-20 | The device and method that a kind of article profile is obtained and recognized |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106874903A true CN106874903A (en) | 2017-06-20 |
Family
ID=59167115
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710089340.XA Pending CN106874903A (en) | 2017-02-20 | 2017-02-20 | The device and method that a kind of article profile is obtained and recognized |
| CN201820003382.7U Expired - Fee Related CN207690117U (en) | 2017-02-20 | 2018-01-02 | A kind of device that article profile is obtained and identified |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201820003382.7U Expired - Fee Related CN207690117U (en) | 2017-02-20 | 2018-01-02 | A kind of device that article profile is obtained and identified |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN106874903A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111623961A (en) * | 2020-06-10 | 2020-09-04 | 太仓考斯茂石英有限公司 | Glass lens detector and control method |
| CN115200480A (en) * | 2022-09-17 | 2022-10-18 | 深圳市巨力方视觉技术有限公司 | Alignment and lamination visual detection system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5986230A (en) * | 1996-09-13 | 1999-11-16 | Uncle Ben's, Inc. | Method and apparatus for sorting product |
| CN201340622Y (en) * | 2008-10-10 | 2009-11-04 | 徐季敏 | Can detection and identification system |
| CN101825582A (en) * | 2010-05-19 | 2010-09-08 | 山东明佳包装检测科技有限公司 | Method and device for detecting wall of cylindrical transparent bottle |
| CN102944563A (en) * | 2012-09-28 | 2013-02-27 | 肇庆中导光电设备有限公司 | Lighting device with transmission and reflection light source, detection system and detection method thereof |
| CN103886311A (en) * | 2014-03-31 | 2014-06-25 | 北京大恒图像视觉有限公司 | Identification system for identifying codes on side of bottle |
| CN203773560U (en) * | 2014-03-31 | 2014-08-13 | 北京大恒图像视觉有限公司 | Mold coding imaging apparatus |
| CN104550051A (en) * | 2014-12-23 | 2015-04-29 | 山东明佳科技有限公司 | Empty bottle sorting system for glass bottle |
| CN204544803U (en) * | 2014-12-23 | 2015-08-12 | 山东明佳科技有限公司 | A kind of vial empty bottle sorting system |
-
2017
- 2017-02-20 CN CN201710089340.XA patent/CN106874903A/en active Pending
-
2018
- 2018-01-02 CN CN201820003382.7U patent/CN207690117U/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5986230A (en) * | 1996-09-13 | 1999-11-16 | Uncle Ben's, Inc. | Method and apparatus for sorting product |
| CN201340622Y (en) * | 2008-10-10 | 2009-11-04 | 徐季敏 | Can detection and identification system |
| CN101825582A (en) * | 2010-05-19 | 2010-09-08 | 山东明佳包装检测科技有限公司 | Method and device for detecting wall of cylindrical transparent bottle |
| CN102944563A (en) * | 2012-09-28 | 2013-02-27 | 肇庆中导光电设备有限公司 | Lighting device with transmission and reflection light source, detection system and detection method thereof |
| CN103886311A (en) * | 2014-03-31 | 2014-06-25 | 北京大恒图像视觉有限公司 | Identification system for identifying codes on side of bottle |
| CN203773560U (en) * | 2014-03-31 | 2014-08-13 | 北京大恒图像视觉有限公司 | Mold coding imaging apparatus |
| CN104550051A (en) * | 2014-12-23 | 2015-04-29 | 山东明佳科技有限公司 | Empty bottle sorting system for glass bottle |
| CN204544803U (en) * | 2014-12-23 | 2015-08-12 | 山东明佳科技有限公司 | A kind of vial empty bottle sorting system |
Non-Patent Citations (2)
| Title |
|---|
| 王刚: "图像二值化方法研究及其在监控识别系统中的应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 高海涛: "激光智能识别分拣系统关键技术研究", 《激光智能识别分拣系统关键技术研究》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111623961A (en) * | 2020-06-10 | 2020-09-04 | 太仓考斯茂石英有限公司 | Glass lens detector and control method |
| CN115200480A (en) * | 2022-09-17 | 2022-10-18 | 深圳市巨力方视觉技术有限公司 | Alignment and lamination visual detection system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN207690117U (en) | 2018-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110220917B (en) | Crown cap surface defect online detection method based on image processing | |
| CN104297255B (en) | A kind of dixie cup defective vision detection method and system and device | |
| CN108918543B (en) | Dynamic detection device and method for mirror surface scratch | |
| CN105865570A (en) | Machine vision-based glass bottle liquid level detection method | |
| Liu et al. | Automatic detection technology of surface defects on plastic products based on machine vision | |
| CN103063572A (en) | Automatic on-line vision detecting and sorting system and sorting method | |
| CN107742119B (en) | Object contour extraction and matching device and method based on back-image imaging | |
| CN108760747A (en) | A kind of 3D printing model surface defect visible detection method | |
| CN102842034B (en) | A kind of laser scanning and the device and the recognition methods that automatically identify engraving character | |
| CN104463067B (en) | Method for extracting macro blocks of Grid Matrix two-dimensional bar code | |
| CN104034638A (en) | Diamond wire particle online quality inspection method based on machine vision | |
| CN108120353B (en) | Fire transmission hole detection device and method based on intelligent camera | |
| CN104034637B (en) | Based on the online quality inspection device of diamond wire particle of machine vision | |
| CN207690117U (en) | A kind of device that article profile is obtained and identified | |
| CN110108712A (en) | Multifunctional visual sense defect detecting system | |
| CN201340622Y (en) | Can detection and identification system | |
| CN102374997A (en) | High-precision detection device of coin surface quality based on vision system | |
| CN113903024A (en) | Handwritten bill numerical value information identification method, system, medium and device | |
| CN204758899U (en) | Lens intelligent recognition assembles equipment | |
| CN103093229A (en) | Positioning method and device of vehicle logo | |
| CN207397296U (en) | A kind of object identification device based on figure viewed from behind imaging | |
| CN203965287U (en) | The online quality inspection device of diamond wire particle based on machine vision | |
| CN113189005B (en) | Portable surface defect integrated detection device and surface defect automatic detection method | |
| CN113066087B (en) | Intelligent detection and labeling system and method for parts | |
| CN105510339A (en) | Beverage bottle fullness detection method based on machine vision |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170620 |
|
| WD01 | Invention patent application deemed withdrawn after publication |