CN105528782B - A kind of milk carton suction pipe whether there is or not detection method - Google Patents
A kind of milk carton suction pipe whether there is or not detection method Download PDFInfo
- Publication number
- CN105528782B CN105528782B CN201510870509.6A CN201510870509A CN105528782B CN 105528782 B CN105528782 B CN 105528782B CN 201510870509 A CN201510870509 A CN 201510870509A CN 105528782 B CN105528782 B CN 105528782B
- Authority
- CN
- China
- Prior art keywords
- image
- suction pipe
- template
- circular projection
- feature vector
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0008—Industrial image inspection checking presence/absence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30128—Food products
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of milk carton suction pipe whether there is or not detection method, including：The milk carton image that a width is stained with suction pipe is chosen, interception includes the template of suction pipe head zone；The edge of template is acquired by Canny edge detection operators；Circular projection's feature vector of the template after edge detection is acquired by circular projection's method；Acquire a milk carton image to be detected, image referred to as to be matched；By canny edge detection operators acquire image to be matched with 1) in the corresponding region of template edge；Seek circular projection's feature vector by edge detection of image to be matched；Seek the related coefficient of circular projection's feature vector；Judge that suction pipe whether there is.The present invention makes template according to suction edge, overcomes conventional template matching not have resistance problem to rotation image using circular projection's method.
Description
Technical field
The present invention relates to a kind of Mongolia Ox's milk carton suction pipe whether there is or not detection method.
Background technology
It is suction pipe to be pasted toward in milk carton, but there is leakage in paste process to have a procedure in the box-packed milk production process of Mongolia Ox
The problems such as viscous and milk carton is fallen when being driven with production line, leads in vanning that there are part milk cartons not to have the phenomenon that suction pipe.At present
It is main to be detected using the method manually watched, but there are the influences etc. of visual fatigue and artificial subjective factor for artificial detection
It restricts.
Invention content
In view of the above-mentioned problems, the object of the present invention is to provide a kind of on-line quick detection milk carton suction pipe whether there is or not method, take
For artificial detection.
To achieve the above object, the present invention takes following technical scheme：
A kind of milk carton suction pipe whether there is or not detection method, include the following steps：
1) the milk carton image that a width is stained with suction pipe is chosen, interception includes the template of suction pipe head zone, and determines search model
It encloses；
2) edge of template is acquired by Canny edge detection operators；
3) circular projection's feature vector of the template after edge detection is acquired by circular projection's method；
4) a milk carton image to be detected, image referred to as to be matched are acquired；
5) by canny edge detection operators acquire image to be matched with 1) in the corresponding region of template edge；
6) seek image to be matched by edge detection with 1) in the corresponding region of template circular projection feature vector；
7) circular projection's feature vector for being sought in seeking 3) and 6) in the circular projection's feature vector sought related coefficient；
8) when coefficient R (i, j) is more than given threshold Thres, judge that suction pipe exists, conversely, judging no suction pipe.
The invention adopts the above technical scheme, which has the following advantages：
(1) it is directed to suction edge evident characteristic and edge detection is carried out to suction pipe first.
(2) template is made according to suction edge, overcomes conventional template matching not anti-to rotation image using circular projection's method
Sex chromosome mosaicism.
Description of the drawings
Fig. 1 is that have suction pipe figure
Fig. 2 is no suction pipe figure
Fig. 3 is template interception
Fig. 4 is edge treated
Fig. 5 is circular projection's template
Specific implementation mode
The present invention is described in detail with reference to the accompanying drawings and examples.
This system using the method based on template matches come detect suction pipe whether there is or not.Traditional template matches pass through calculation template
The correlation of image and image to be matched searches in image to be detected whether to have template image.There are calculation amounts for this method
Greatly, and to have the image of rotation can not matched well defect.
The present invention is directed to the apparent characteristic of suction pipe template image edge feature, proposes a kind of by template image and figure to be matched
Feature vector is acquired by circular projection's method again as carrying out edge detection first, then carries out the template matching method of correlation calculations.
This method has the advantages that speed is fast and is not influenced by image rotation.
1, it chooses a width and is stained with the milk carton image of suction pipe as template image；
Being stained with suction pipe and nothing has the image of suction pipe as shown in Figure 1, Figure 2.Find out at the top of suction pipe to be bend pipe by Fig. 1, has preferable
Edge feature.Template such as Fig. 3 is intercepted, the equal wide height of template of interception is all W, and small frame is template image, and big frame is search range.
2, edge such as Fig. 4 of template image (suction pipe head zone) is acquired by Canny edge detection operators；
Canny edge detection operators are a multistage edge detection algorithms, and Canny edge detections are according to signal-to-noise ratio and calmly
Position product is estimated, and obtains optimizing Approximation Operator.It can be divided into following 5 steps：
(1) carry out smoothed image using gaussian filtering
(2) intensity gradient of image is looked for
(3) side flase drop is eliminated using non-maximum suppression technology
(4) possible boundary is determined using the method for dual threshold
(5) boundary is tracked using hysteresis techniques
3, the feature vector of edge detection rear pattern plate image is acquired by circular projection's method
Circular projection is defined as：With the gray scale summation for all pixels point at a distance from subgraph center pixel being r, it is denoted as Ce (r)
Such as Fig. 5.Figure sciagraphy has the advantages that calculating speed is fast, rotational invariance.The template image that the present invention intercepts is square, side
A length of W, take image center be circular projection center, seek successively distance center radius be 1 to W pixel gray scale and, ask
Obtaining feature vector is:
{Ce(1), Ce(2),Ce(3)...Ce(w)}
4, a milk carton image to be detected, image referred to as to be matched are acquired；
5, the edge of the suction pipe head zone of image to be matched is acquired by canny edge detection operators；
6, image circular projection feature vector { Te to be matched is sought(1), Te(2),Te(3)...Te(w)}
Seek Prototype drawing feature vector { Ce(1), Ce(2),Ce(3)...Ce(w)And image feature vector { Te to be matched(1),
Te(2),Te(3)...Te(w)Correlation
When template image and just the same image to be matched, coefficient R=1, template image and image phase to be matched
Higher like spending, related coefficient is bigger.When coefficient R is more than given threshold Thres, suction pipe exists, conversely, without suction pipe.
Claims (1)
1. a kind of milk carton suction pipe whether there is or not detection method, include the following steps：
1) the milk carton image that a width is stained with suction pipe is chosen, interception includes the template of suction pipe head zone, and determines search range；
2) edge of template is acquired by Canny edge detection operators；
3) circular projection's feature vector of the template after edge detection is acquired by circular projection's method；
4) a milk carton image to be detected, image referred to as to be matched are acquired；
5) by canny edge detection operators acquire image to be matched with 1) in the corresponding region of template edge；
6) seek image to be matched by edge detection with 1) in the corresponding region of template circular projection feature vector；
7) circular projection's feature vector for being sought in seeking 3) and 6) in the circular projection's feature vector sought related coefficient；
8) when coefficient R (i, j) is more than given threshold Thres, judge that suction pipe exists, conversely, judging no suction pipe.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510870509.6A CN105528782B (en) | 2015-12-01 | 2015-12-01 | A kind of milk carton suction pipe whether there is or not detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510870509.6A CN105528782B (en) | 2015-12-01 | 2015-12-01 | A kind of milk carton suction pipe whether there is or not detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105528782A CN105528782A (en) | 2016-04-27 |
| CN105528782B true CN105528782B (en) | 2018-08-21 |
Family
ID=55770985
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510870509.6A Active CN105528782B (en) | 2015-12-01 | 2015-12-01 | A kind of milk carton suction pipe whether there is or not detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105528782B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106204534B (en) * | 2016-06-28 | 2019-02-26 | 西安理工大学 | A kind of printed matter image characteristic region extraction method without handmarking |
| CN110586500A (en) * | 2019-09-20 | 2019-12-20 | 苏州市兆众自动化有限公司 | Intelligent detection system for defects of paper bottle straws |
| CN112686840A (en) * | 2020-12-16 | 2021-04-20 | 广州大学 | Method, system and device for detecting straw on surface of beverage packaging box and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156978A (en) * | 2010-12-24 | 2011-08-17 | 辽宁科锐科技有限公司 | Industrial device rapid locating method based on machine vision |
| CN102509300A (en) * | 2011-11-18 | 2012-06-20 | 深圳市宝捷信科技有限公司 | Defect detection method and system |
| CN102530327A (en) * | 2011-12-29 | 2012-07-04 | 内蒙古伊利实业集团股份有限公司 | Online detection system for adhesion of straws to packing boxes |
| CN102968785A (en) * | 2012-10-18 | 2013-03-13 | 华中科技大学 | High-speed parallel image matching method based on multi-core digital signal processor (DSP) |
| US8774510B2 (en) * | 2012-09-11 | 2014-07-08 | Sharp Laboratories Of America, Inc. | Template matching with histogram of gradient orientations |
-
2015
- 2015-12-01 CN CN201510870509.6A patent/CN105528782B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156978A (en) * | 2010-12-24 | 2011-08-17 | 辽宁科锐科技有限公司 | Industrial device rapid locating method based on machine vision |
| CN102509300A (en) * | 2011-11-18 | 2012-06-20 | 深圳市宝捷信科技有限公司 | Defect detection method and system |
| CN102530327A (en) * | 2011-12-29 | 2012-07-04 | 内蒙古伊利实业集团股份有限公司 | Online detection system for adhesion of straws to packing boxes |
| US8774510B2 (en) * | 2012-09-11 | 2014-07-08 | Sharp Laboratories Of America, Inc. | Template matching with histogram of gradient orientations |
| CN102968785A (en) * | 2012-10-18 | 2013-03-13 | 华中科技大学 | High-speed parallel image matching method based on multi-core digital signal processor (DSP) |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105528782A (en) | 2016-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104574353B (en) | The surface defect decision method of view-based access control model conspicuousness | |
| CN103871053B (en) | Vision conspicuousness-based cloth flaw detection method | |
| CN106530297B (en) | Grasping body area positioning method based on point cloud registering | |
| CN117173189A (en) | Visual inspection system for polishing effect of aluminum alloy surface | |
| CN104636722B (en) | A kind of overlapping fruit quick Tracking Recognition method of picking robot | |
| CN105528782B (en) | A kind of milk carton suction pipe whether there is or not detection method | |
| CN110687904A (en) | Visual navigation routing inspection and obstacle avoidance method for inspection robot | |
| CN104268598B (en) | Human leg detection method based on two-dimensional scanning lasers | |
| CN103793693A (en) | Method for detecting face turning and facial form optimizing method with method for detecting face turning | |
| CN105023027B (en) | Live soles spoor decorative pattern image search method based on multiple feedback mechanism | |
| Schauwecker et al. | A new feature detector and stereo matching method for accurate high-performance sparse stereo matching | |
| CN103759644B (en) | A kind of separation refinement intelligent detecting method of optical filter blemish | |
| Zhou et al. | A new design of sugarcane seed cutting systems based on machine vision | |
| CN105513037B (en) | Angular-point detection method and device | |
| CN110738631A (en) | prawn shape information detection method and detection system based on images | |
| WO2016034104A1 (en) | Self-moving surface walking robot and image processing method therefor | |
| CN104680538A (en) | SAR image CFAR target detection method on basis of super pixels | |
| CN104282018A (en) | Method for on-line detection of overall diameter of industrial product based on machine vision | |
| CN102663384A (en) | Curve identification method based on Bezier control point searching and apparatus thereof | |
| CN107993219A (en) | A kind of deck of boat detection method of surface flaw based on machine vision | |
| CN109858438A (en) | A kind of method for detecting lane lines based on models fitting | |
| CN106780437B (en) | A kind of quick QFN chip plastic packaging image obtains and amplification method | |
| CN111242888A (en) | Image processing method and system based on machine vision | |
| Bui et al. | Road area detection based on texture orientations estimation and vanishing point detection | |
| CN104537632A (en) | Infrared image histogram enhancing method based on edge extraction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |