CN101315338A - Defect detecting system and method for glass product - Google Patents
Defect detecting system and method for glass product Download PDFInfo
- Publication number
- CN101315338A CN101315338A CNA2008101168729A CN200810116872A CN101315338A CN 101315338 A CN101315338 A CN 101315338A CN A2008101168729 A CNA2008101168729 A CN A2008101168729A CN 200810116872 A CN200810116872 A CN 200810116872A CN 101315338 A CN101315338 A CN 101315338A
- Authority
- CN
- China
- Prior art keywords
- glassware
- detected
- camera
- image
- edge
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 30
- 239000011521 glass Substances 0.000 title claims abstract description 13
- 238000001514 detection method Methods 0.000 claims abstract description 48
- 238000003708 edge detection Methods 0.000 claims abstract description 27
- 230000002950 deficient Effects 0.000 claims description 45
- 239000004744 fabric Substances 0.000 claims description 37
- 238000006243 chemical reaction Methods 0.000 claims description 28
- 238000007689 inspection Methods 0.000 claims description 12
- 235000013405 beer Nutrition 0.000 description 40
- 230000008569 process Effects 0.000 description 10
- 230000008859 change Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 208000037656 Respiratory Sounds Diseases 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 2
- 235000021262 sour milk Nutrition 0.000 description 2
- 206010011376 Crepitations Diseases 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000009514 concussion Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a defect detection system and a defect detection method for glass products. The method comprises the following steps: taking an image of a glass product to be detected by a camera head, carrying out edge detection on the image, and obtaining an edge graph of the glass product to be detected; then carrying out binarization processing to the edge graph obtained by the edge detection; and removing a strong edge representing a typical line in the binary edge graph. Since the glass products are usually transparent or semitransparent, the edge graph is bound to include the strong edge representing a contour line and the typical line with regard to a glass product without defects; if the strong edge representing the typical line in the binary edge graph is removed, and other strong edges are also included besides the strong edge representing the typical line, a control signal representing that defects existing in the detected glass product is generated, thereby finishing the defect detection of the glass product without manual operation, and further improving the efficiency and the reliability in the defect detection of the glass product.
Description
Technical field
The present invention relates to defect detecting technique, particularly the defect inspection method of a kind of defect detecting system of glassware and a kind of glassware.
Background technology
Glassware is increasing to be used in daily life, and glass containers such as beer bottle, Sour milk bottle also have the handicraft of glass material etc.
Because the imperfection of manufacturing process, may there be for example defective such as bubble, calculus in glassware in manufacture process; And glassware is in production, application, transportation, defective such as also may crack, damaged owing to the concussion or the collision in the transmission course of streamline.
The above-mentioned defective of glassware may cause the generation of various accidents, and for example blast of beer bottle, sour milk bottleneck are cut skin etc., therefore, needs usually it is carried out defects detection before glassware comes into operation.
Yet, the defects detection of existing glassware is normally by manually finishing, promptly rely on defects detection personnel's naked-eye observation to judge whether glassware exists defective, this just makes that the defects detection efficient of glassware is lower, and may lacking experience or neglecting and cause the reliability of defects detection lower owing to the defects detection personnel.
Summary of the invention
In view of this, the invention provides a kind of defect detecting system of glassware and a kind of defect inspection method of glassware, can improve the efficient and the reliability of glassware defects detection.
The defect detecting system of a kind of glassware provided by the invention comprises:
Camera is taken the image of detected glassware;
Edge detection unit, the image of the detected glassware that camera is photographed carries out rim detection, obtains the outline map of detected glassware, comprises the appearance profile line of representing detected glassware and the strong edge of characteristic curve in the described outline map at least;
The binary conversion treatment unit, the outline map that described edge detection unit is obtained carries out binary conversion treatment, obtains the binaryzation outline map;
The characteristic curve detecting unit, the strong edge of the described characteristic curve of expression is removed in the binaryzation outline map that described binary conversion treatment unit is obtained;
The defective judging unit, the binaryzation outline map of reception after the characteristic curve detecting unit is handled, when except that the strong edge of the described appearance profile line of expression, also comprising other strong edges in the binaryzation outline map that receives, produce the detected glassware of expression and have defect Control signal and output.
Described camera is at least two, be arranged at the both sides of detected glassware place streamline respectively, and the coverage of described at least two cameras can cover the surrounded surface of detected glassware;
This system further comprises three place's light sources, is arranged at the both sides and the top of detected glassware streamline respectively.
Each camera is taken continuously to detected glassware, obtains comprising the image sets of multiple image;
Described edge detection unit further selects piece image to carry out rim detection respectively from the described image sets that each camera photographs.
In every width of cloth image that described edge detection unit is selected, detected glassware is positioned at the centre position of this width of cloth image.
Described defective judging unit further carries out the defective training of judgement according to detected glassware at the defect shape of diverse location in advance.
Described at least two cameras are divided into some camera groups, and different camera groups are taken the zones of different of detected glassware respectively;
The quantity of edge detection unit, binary conversion treatment unit, characteristic curve detecting unit, defective judging unit is identical with the quantity of described camera group;
And each edge detection unit, binary conversion treatment unit, characteristic curve detecting unit, the corresponding camera group of defective judging unit.
In described at least two camera groups, the downward-sloping predetermined angle of each camera of at least one camera group.
Described detected glassware is a vial;
The characteristic curve of described vial comprises a bottle suture, bottleneck line, bottleneck line at least;
Described at least two camera groups comprise first camera group of taking described vial body zone and the second camera group of taking described glass bottle port area;
Two camera groups are taken the body zone and the bottleneck zone of described vial respectively;
The downward-sloping predetermined angle of each camera in the described second camera group.
The defect inspection method of a kind of glassware provided by the invention is provided with the camera of taking detected glassware image, and this method comprises:
The detected glassware image that camera is photographed carries out rim detection, obtains the outline map of detected glassware; Wherein, comprise the appearance profile line of representing detected glassware and the strong edge of characteristic curve in the described outline map at least;
The described outline map that rim detection is obtained carries out binary conversion treatment, obtains the binaryzation outline map;
The strong edge of representation feature line in the described binaryzation outline map is removed;
In the binaryzation outline map behind the strong edge of having removed the representation feature line, when except that the strong edge of expression appearance profile line, also comprising other strong edges, produce the detected glassware of expression and have the defect Control signal.
Each camera is taken continuously to detected glassware, obtain comprising the image sets of multiple image, the described detected glassware image that camera is photographed carries out before the rim detection, and this method further comprises: select piece image to be used for described edge inspection side from the described image sets that each camera photographs respectively.
As seen from the above technical solution, the present invention carries out rim detection to the image of the detected glassware that camera photographs, and obtains the outline map of detected glassware; The outline map that rim detection is obtained carries out binary conversion treatment then, and the strong edge of representation feature line in the binaryzation outline map is removed.Like this, because glassware is generally transparent or semitransparent, for the glassware that does not have defective, should include only the strong edge of expression appearance profile line and characteristic curve in the outline map, therefore, if removed in the binaryzation outline map at strong edge of representation feature line, except that the strong edge of expression appearance profile line, also comprise other strong edges, represent that then there is defective in this glassware, and there is the defect Control signal in the detected glassware of generation expression, thereby need not the defects detection that manually-operated can be finished glassware, and then improved the efficient and the reliability of glassware defects detection.
And the technical scheme among the present invention can utilize computer program and camera to realize the online detection of the defective of glassware automatically in conjunction with the streamline of producing glassware.
Description of drawings
Fig. 1 is the exemplary block diagram of the defect detecting system of glassware among the present invention;
Fig. 2 is the camera schematic layout pattern of the defect detecting system of glassware in the embodiment of the invention;
Fig. 3 is the structural representation of the defect detecting system of glassware in the embodiment of the invention;
Fig. 4 is the exemplary process diagram of the defect inspection method of glassware among the present invention.
Embodiment
For making purpose of the present invention, technical scheme and advantage clearer, below with reference to the accompanying drawing embodiment that develops simultaneously, the present invention is described in more detail.
The present invention utilizes the defects detection of image processing techniques realization to glassware.
Fig. 1 is the exemplary block diagram of the defect detecting system of glassware among the present invention.As shown in Figure 1, this defect detecting system comprises: camera, edge detection unit, binary conversion treatment unit, characteristic curve detecting unit, defective judging unit, control output unit.
Camera is at least two, is arranged at the both sides of detected glassware streamline respectively, and guarantees that the coverage of all cameras can cover the surrounded surface of detected glassware.The image of the detected glassware different angles that like this, different cameras can photograph.Each camera is taken the detected glassware of this camera of process in the streamline, and the image of each detected glassware that will photograph is exported successively.Certainly, camera also can only be set to one, and the coverage of camera covers the surrounded surface of detected glassware and nonessential.
Edge detection unit receives each width of cloth image of each detected glassware that each camera photographs successively; Utilize existing any one edge detection algorithm, each width of cloth image to each detected glassware of receiving carries out rim detection successively, obtains the outline map of each detected glassware in each width of cloth image.
Wherein, the position that grey scale change is violent in the image has comprised a large amount of information usually, is called strong edge.In the every width of cloth image that contains glassware that camera is taken, owing to wherein represent the target area of detected glassware, the background area outside the detected glassware, grey scale change is very violent, therefore, necessarily comprise the strong edge of representing the target area outline line in the outline map that rim detection obtains, just represent the strong edge of detected glassware appearance profile line; And the grey scale change at glassware its own characteristics line place also can compare acutely, and for example therefore the bottle suture etc. of beer bottle, also can comprise the strong edge of representing detected glassware characteristic curve in the outline map that inspection side in edge obtains.
Because glassware is generally transparent or semitransparent, for the glassware that does not have defective, rim detection to outline map in should include only the strong edge of expression appearance profile line and characteristic curve, certainly, if there is defective in glassware, then the grey scale change of fault location also can compare acutely, at this moment, just might comprise the strong edge of representing detected glassware defect profile line in the outline map that inspection side in edge obtains.
In the practical application, each camera all is to take continuously, and detected glassware can be constantly mobile under the drive of streamline, therefore, for same detected glassware, the continuous shooting of each camera can obtain comprising the image sets of multiple image.Like this, consider the needs of detection speed, the image sets of the corresponding same detected glassware that photographs for each camera, edge detection unit only therefrom selects a width of cloth to carry out rim detection, rather than all images in the image sets is all carried out rim detection.Preferably, in every width of cloth image that edge detection unit is selected, detected glassware is positioned at the centre position of this width of cloth image, and this is that illumination condition and picture quality when it is taken are best because when being positioned at the image in centre position for detected glassware.
The binary conversion treatment unit, receive the outline map of each detected glassware in each width of cloth image that edge detection unit obtains successively, and successively several outline maps of each detected glassware of receiving are carried out binary conversion treatment, obtain several binaryzation outline maps of each detected glassware.
The characteristic curve detecting unit receives each width of cloth binaryzation outline map of each detected glassware of binary conversion treatment unit output successively, and removes the strong edge of representation feature line in each width of cloth binaryzation outline map of each detected glassware successively.
Wherein, in the binaryzation outline map, strong marginal portion is generally white, and correspondingly, other parts then are black, and " removal " described here is meant: with the strong edge transition of white of representation feature line in the binaryzation outline map is black.
The defective judging unit, receive the binaryzation outline map of each detected glassware after the characteristic curve detecting unit is handled successively, in the binaryzation outline map of certain the detected glassware that receives, when except that the strong edge of this detected glassware appearance profile line of expression, also comprising other strong edges, think that then described other strong edges are defective, produce the detected glassware of expression and have defect Control signal and output.
Consider the computational accuracy of precision, edge detection unit and the binary conversion treatment unit of camera, in the binaryzation outline map that the defective judging unit is received, also might comprise picture noise.Therefore, noise can not thought by mistake be the defective of detected glassware for the defective judging unit, in the present invention, can be according to detected glassware at the common defect shape that occurs of diverse location, in advance the defective judging unit is carried out the defective training of judgement, for example, for the zone that the linear crackle often occurs, if the edge pixel of Discrete Distribution point then can be thought picture noise but not the defective of glassware.
In the practical application, the defective judging unit can produce the detected glassware of expression and have the defect Control signal, and exports the pusher that is provided with in the streamline to, is pushed to outside the streamline by the glassware of this pusher with correspondence.
In order to make the grey scale change of image of fault location of glassware more obvious, the present invention can also further be provided with three place's light sources in said system, lay respectively at the both sides and the top of detected glassware streamline.
As seen, said system is carried out rim detection to the image of the detected glassware that camera photographs, and obtains the outline map of detected glassware; The outline map that rim detection is obtained carries out binary conversion treatment then, and the strong edge of representation feature line in the binaryzation outline map is removed.Like this, because glassware is generally transparent or semitransparent, for the glassware that does not have defective, should include only the strong edge of expression appearance profile line and characteristic curve in the outline map, therefore, if removed in the binaryzation outline map at strong edge of representation feature line, except that the strong edge of expression appearance profile line, also comprise other strong edges, represent that then there is defective in this glassware, and there is the defect Control signal in the detected glassware of generation expression, thereby need not the defects detection that manually-operated can be finished glassware, and then improved the efficient and the reliability of glassware defects detection.
Below, be beer bottle with detected glassware, characteristic curve comprises that a bottle suture, bottleneck line, bottleneck line are example, and said system is illustrated.
The position that beer bottle exists defective relatively to concentrate is a body, bottle mouth position, wherein, there is bubble easily in body, calculus etc., there is breakage easily in bottleneck, crackle etc., therefore, present embodiment provides at least two cameras, and at least two cameras are divided into two groups, take the body of beer bottle and the image in bottleneck zone respectively, and the edge detection unit of setting and camera group equal number, the binary conversion treatment unit, the characteristic curve detecting unit, defective judging unit, and each edge detection unit, the binary conversion treatment unit, the characteristic curve detecting unit, the corresponding camera group of defective judging unit.
Fig. 2 is the camera schematic layout pattern of the defect detecting system of glassware in the embodiment of the invention.In Fig. 2, the streamline of beer bottle moves from left to right, and has 8 cameras in Fig. 2, and described 8 cameras are equally divided into two camera groups.
One of them camera group is called body camera group, and 4 cameras in this group are arranged at the both sides of beer bottle streamline left end in twos respectively, is used to take the body zone of beer bottle; Another camera group is called bottleneck camera group, and 4 cameras in this group are arranged at the both sides of beer bottle streamline right-hand member in twos respectively, is used to take the bottleneck zone of beer bottle.
Preferably, in the bottleneck camera group in corresponding bottleneck zone, each camera all adopts the mode of downward-sloping certain angle to tilt to take the bottleneck zone of beer bottle, to obtain the bottleneck area image of better effects if.
In Fig. 2, the right side of each group camera all is provided with bottle device that picks of a correspondence, as foregoing pusher.
Fig. 3 is the structural representation of the defect detecting system of glassware in the embodiment of the invention.As shown in Figure 3, except two camera groups as shown in Figure 2, the glassware defect detecting system in the present embodiment also comprises:
Edge detection unit A receives 4 width of cloth body images of each beer bottle that body camera group photographs successively; Utilize existing any one edge detection algorithm, 4 width of cloth body images to each beer bottle of receiving carry out rim detection successively, obtain each beer bottle 4 width of cloth body edge of image figure.
Binary conversion treatment unit A, receive each beer bottle successively at 4 width of cloth body edge of image figure, and successively 4 width of cloth body edge of image figure of each beer bottle of receiving are carried out binary conversion treatment, obtain the binaryzation outline map of 4 width of cloth body images of each beer bottle.
Characteristic curve detecting unit A receives the binaryzation outline map of 4 width of cloth body images of each beer bottle successively, and removes each beer bottle is represented the bottle suture in the binaryzation outline map of 4 width of cloth body images strong edge successively.
Defective judging unit A, receive the binaryzation outline map of 4 width of cloth body images of each beer bottle after characteristic curve detecting unit A handles successively, when except that the strong edge of expression beer bottle body appearance profile line, also comprising other strong edges in the binaryzation outline map that receives, then produce bottle device A that picks that there is the defect Control signal in the expression beer bottle and exports as shown in Figure 2 body camera group right side to, from streamline, pick out so that will have the beer bottle of defective.
Edge detection unit B receives 4 width of cloth bottleneck images of each beer bottle that bottleneck camera group photographs successively; Utilize existing any one edge detection algorithm, 4 width of cloth bottleneck images to each beer bottle of receiving carry out rim detection successively, obtain 4 width of cloth bottleneck edge of image figure of each beer bottle.
The binary conversion treatment unit B, receive 4 width of cloth bottleneck edge of image figure of each beer bottle successively, and successively the outline map in 4 width of cloth bottleneck images of each beer bottle of receiving is carried out binary conversion treatment, obtain the binaryzation outline map of 4 width of cloth bottleneck images of each beer bottle.
Characteristic curve detecting unit B receives the binaryzation outline map of 4 width of cloth bottleneck images of each beer bottle successively, and removes the strong edge of expression bottleneck line, bottleneck line and bottle suture in the binaryzation outline map of 4 width of cloth bottleneck images of each beer bottle successively.
Need to prove that therefore the bottle suture, in the characteristic curve in the bottleneck zone of beer bottle, except bottleneck line and bottleneck line, still comprises a bottle suture at the bottom of being extended to bottle by bottleneck.
Defective judging unit B, receive the binaryzation outline map of 4 width of cloth bottleneck images of each beer bottle after characteristic curve detecting unit B handles successively, when except that the strong edge of expression beer bottle bottleneck appearance profile line, also comprising other strong edges in the binaryzation outline map that receives, then produce bottle device B that picks that there is the defect Control signal in the expression beer bottle and exports as shown in Figure 2 bottleneck camera group right side to, from streamline, pick out so that will have the beer bottle of defective.
In present embodiment system as shown in Figure 2, at defect shapes such as the common bubble that occurs in body zone, calculus, in advance defective judging unit A is carried out the defective training of judgement according to beer bottle; At defect shapes such as the common breakage that occurs in bottleneck zone, crackles, in advance defective judging unit B is carried out the defective training of judgement according to beer bottle.
Certainly, for other vials except that beer bottle, present embodiment system as shown in Figure 3 also is suitable for.And, also can camera group and other functional units be set respectively at more and more tiny zone in order to detect the defective of vial or other glasswares more accurately.
Correspondingly, if be provided with several camera groups, then can be according to actual conditions, the downward-sloping predetermined angle of each camera in any one or a plurality of camera group is set.
More than be detailed description to the defect detecting system of glassware among the present invention, below, the defect inspection method to glassware among the present invention describes again.
Fig. 4 is the exemplary process diagram of the defect inspection method of glassware among the present invention.As shown in Figure 4, at least two cameras are set in the both sides of detected glassware streamline, and the coverage of at least two cameras can cover after the surrounded surface of detected glassware, this method comprises:
Step 401, each width of cloth image of the detected glassware that each camera is photographed carries out rim detection, obtains the outline map of detected glassware in each width of cloth image.
Wherein, the strong edge that comprises detected glassware appearance profile line of expression and characteristic curve in the outline map that this step obtains at least.
In the practical application, each camera all is to carry out shooting continuously, and detected glassware can be constantly mobile under the drive of streamline, and therefore, the image of the corresponding same detected glassware that each camera can photograph is several.Like this, consider the needs of detection speed, the multiple image of the corresponding same detected glassware that photographs for each camera, only selecting wherein in this step, a width of cloth carries out rim detection.Preferably, the piece image of selecting detected glassware to be positioned at the centre position carries out rim detection, and this is because the illumination condition of target area when being positioned at the centre position is good, picture quality is the highest.
Step 404 in the binaryzation outline map at the strong edge of having removed the representation feature line, when also comprising other strong edges except that the strong edge of expression appearance profile line, produces the detected glassware of expression and has the defect Control signal.
So far, this flow process finishes.
In above-mentioned flow process, camera also can only be set to one, and the coverage of camera covers the surrounded surface of detected glassware and nonessential.
Above-mentioned flow process only is the processing procedure for a detected glassware, in the practical application, for each detected glassware, can arrive order in the camera coverage, the above-mentioned flow process of executed in parallel according to glassware.
And, for same detected glassware, at least two cameras can be divided into different camera groups, different camera groups are taken the zones of different of this detected glassware respectively, and simultaneously or successively carry out above-mentioned flow process at the zones of different of same detected glassware.
By above-mentioned flow process as seen, the image of the detected glassware that the defect inspection method of glassware photographs camera earlier among the present invention carries out rim detection, obtains the outline map of detected glassware; The outline map that rim detection is obtained carries out binary conversion treatment then, and the strong edge of representation feature line in the binaryzation outline map is removed.Like this, because glassware is generally transparent or semitransparent, for the glassware that does not have defective, should include only the strong edge of expression appearance profile line and characteristic curve in the outline map, therefore, if removed in the binaryzation outline map at strong edge of representation feature line, except that the strong edge of expression appearance profile line, also comprise other strong edges, represent that then there is defective in this glassware, and there is the defect Control signal in the detected glassware of generation expression, thereby need not the defects detection that manually-operated can be finished glassware, and then improved the efficient and the reliability of glassware defects detection.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to and replace and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1, a kind of defect detecting system of glassware is characterized in that, this system comprises:
Camera is taken the image of detected glassware;
Edge detection unit, the image of the detected glassware that camera is photographed carries out rim detection, obtains the outline map of detected glassware, comprises the appearance profile line of representing detected glassware and the strong edge of characteristic curve in the described outline map at least;
The binary conversion treatment unit, the outline map that described edge detection unit is obtained carries out binary conversion treatment, obtains the binaryzation outline map;
The characteristic curve detecting unit, the strong edge of the described characteristic curve of expression is removed in the binaryzation outline map that described binary conversion treatment unit is obtained;
The defective judging unit, the binaryzation outline map of reception after the characteristic curve detecting unit is handled, when except that the strong edge of the described appearance profile line of expression, also comprising other strong edges in the binaryzation outline map that receives, produce the detected glassware of expression and have defect Control signal and output.
2, the system as claimed in claim 1, it is characterized in that, described camera is at least two, be arranged at the streamline both sides at detected glassware place respectively, and the coverage of described at least two cameras can cover the surrounded surface of detected glassware;
This system further comprises three place's light sources, is arranged at the both sides and the top of detected glassware streamline respectively.
3, system as claimed in claim 2 is characterized in that,
Each camera is taken continuously to detected glassware, obtains comprising the image sets of multiple image;
Described edge detection unit further selects piece image to carry out rim detection respectively from the described image sets that each camera photographs.
4, system as claimed in claim 3 is characterized in that, in every width of cloth image that described edge detection unit is selected, detected glassware is positioned at the centre position of this width of cloth image.
5, the system as claimed in claim 1 is characterized in that, described defective judging unit further carries out the defective training of judgement according to detected glassware at the defect shape of diverse location in advance.
6, as any described system in the claim 1 to 5, it is characterized in that described at least two cameras are divided into some camera groups, different camera groups are taken the zones of different of detected glassware respectively;
The quantity of edge detection unit, binary conversion treatment unit, characteristic curve detecting unit, defective judging unit is identical with the quantity of described camera group;
And each edge detection unit, binary conversion treatment unit, characteristic curve detecting unit, the corresponding camera group of defective judging unit.
7, system as claimed in claim 6 is characterized in that, in described at least two camera groups, and the downward-sloping predetermined angle of each camera of at least one camera group.
8, system as claimed in claim 7 is characterized in that,
Described detected glassware is a vial;
The characteristic curve of described vial comprises a bottle suture, bottleneck line, bottleneck line at least;
Described at least two camera groups comprise first camera group of taking described vial body zone and the second camera group of taking described glass bottle port area;
Two camera groups are taken the body zone and the bottleneck zone of described vial respectively;
The downward-sloping predetermined angle of each camera in the described second camera group.
9, a kind of defect inspection method of glassware is characterized in that, the camera of taking detected glassware image is set, and this method comprises:
The detected glassware image that camera is photographed carries out rim detection, obtains the outline map of detected glassware; Wherein, comprise the appearance profile line of representing detected glassware and the strong edge of characteristic curve in the described outline map at least;
The described outline map that rim detection is obtained carries out binary conversion treatment, obtains the binaryzation outline map;
The strong edge of representation feature line in the described binaryzation outline map is removed;
In the binaryzation outline map behind the strong edge of having removed the representation feature line, when except that the strong edge of expression appearance profile line, also comprising other strong edges, produce the detected glassware of expression and have the defect Control signal.
10, method as claimed in claim 9, it is characterized in that, each camera is taken continuously to detected glassware, obtain comprising the image sets of multiple image, the described detected glassware image that camera is photographed carries out before the rim detection, and this method further comprises: select piece image to be used for described edge inspection side from the described image sets that each camera photographs respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101168729A CN101315338A (en) | 2008-07-18 | 2008-07-18 | Defect detecting system and method for glass product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101168729A CN101315338A (en) | 2008-07-18 | 2008-07-18 | Defect detecting system and method for glass product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101315338A true CN101315338A (en) | 2008-12-03 |
Family
ID=40106417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008101168729A Pending CN101315338A (en) | 2008-07-18 | 2008-07-18 | Defect detecting system and method for glass product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101315338A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996405A (en) * | 2010-08-30 | 2011-03-30 | 中国科学院计算技术研究所 | Method and device for rapidly detecting and classifying defects of glass image |
| CN102213681A (en) * | 2011-04-01 | 2011-10-12 | 哈尔滨工业大学（威海） | Novel method for detecting sewage in anti-skidding region at bottom of glass bottle |
| CN103759644A (en) * | 2014-01-23 | 2014-04-30 | 广州市光机电技术研究院 | Separating and refining type intelligent optical filter surface defect detecting method |
| CN104502359A (en) * | 2014-12-10 | 2015-04-08 | 天津普达软件技术有限公司 | Method for accurately detecting bottle blank and bottle opening defects |
| CN105447854A (en) * | 2015-11-12 | 2016-03-30 | 程涛 | Small-size glass panel surface defect detection method and small-size glass panel surface defect detection system |
| CN105866137A (en) * | 2016-06-15 | 2016-08-17 | 秦皇岛可视自动化设备有限公司 | Device and method for online detection of calendered tempered glass edge defects |
| CN106198567A (en) * | 2015-05-06 | 2016-12-07 | 法国圣戈班玻璃公司 | Optical detecting method, optical detection apparatus and Systems for optical inspection |
| CN106770372A (en) * | 2016-11-29 | 2017-05-31 | 中国科学院自动化研究所 | White glass detection method of surface flaw based on micro-vision |
| CN108665458A (en) * | 2018-05-17 | 2018-10-16 | 杭州智谷精工有限公司 | Transparent body surface defect is extracted and recognition methods |
| CN110161054A (en) * | 2019-06-19 | 2019-08-23 | 长沙世通纸塑箸业有限公司 | A kind of dixie cup detection device of removing waste cup |
| CN112750113A (en) * | 2021-01-14 | 2021-05-04 | 深圳信息职业技术学院 | Glass bottle defect detection method and device based on deep learning and linear detection |
| CN113240626A (en) * | 2021-04-08 | 2021-08-10 | 西安电子科技大学 | Neural network-based method for detecting and classifying concave-convex flaws of glass cover plate |
-
2008
- 2008-07-18 CN CNA2008101168729A patent/CN101315338A/en active Pending
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101996405A (en) * | 2010-08-30 | 2011-03-30 | 中国科学院计算技术研究所 | Method and device for rapidly detecting and classifying defects of glass image |
| CN102213681A (en) * | 2011-04-01 | 2011-10-12 | 哈尔滨工业大学（威海） | Novel method for detecting sewage in anti-skidding region at bottom of glass bottle |
| CN102213681B (en) * | 2011-04-01 | 2013-01-02 | 哈尔滨工业大学（威海） | Novel method for detecting sewage in anti-skidding region at bottom of glass bottle |
| CN103759644B (en) * | 2014-01-23 | 2016-05-18 | 广州市光机电技术研究院 | A kind of separation refinement intelligent detecting method of optical filter blemish |
| CN103759644A (en) * | 2014-01-23 | 2014-04-30 | 广州市光机电技术研究院 | Separating and refining type intelligent optical filter surface defect detecting method |
| CN104502359A (en) * | 2014-12-10 | 2015-04-08 | 天津普达软件技术有限公司 | Method for accurately detecting bottle blank and bottle opening defects |
| CN106198567A (en) * | 2015-05-06 | 2016-12-07 | 法国圣戈班玻璃公司 | Optical detecting method, optical detection apparatus and Systems for optical inspection |
| CN105447854A (en) * | 2015-11-12 | 2016-03-30 | 程涛 | Small-size glass panel surface defect detection method and small-size glass panel surface defect detection system |
| CN105866137A (en) * | 2016-06-15 | 2016-08-17 | 秦皇岛可视自动化设备有限公司 | Device and method for online detection of calendered tempered glass edge defects |
| CN105866137B (en) * | 2016-06-15 | 2018-11-30 | 秦皇岛可视自动化设备有限公司 | A kind of calendering tempered glass edge defect on-line measuring device and method |
| CN106770372A (en) * | 2016-11-29 | 2017-05-31 | 中国科学院自动化研究所 | White glass detection method of surface flaw based on micro-vision |
| CN106770372B (en) * | 2016-11-29 | 2019-07-26 | 中国科学院自动化研究所 | Detection method of white glass surface defects based on microscopic vision |
| CN108665458A (en) * | 2018-05-17 | 2018-10-16 | 杭州智谷精工有限公司 | Transparent body surface defect is extracted and recognition methods |
| CN108665458B (en) * | 2018-05-17 | 2022-02-01 | 杭州智谷精工有限公司 | Method for extracting and identifying surface defects of transparent body |
| CN110161054A (en) * | 2019-06-19 | 2019-08-23 | 长沙世通纸塑箸业有限公司 | A kind of dixie cup detection device of removing waste cup |
| CN112750113A (en) * | 2021-01-14 | 2021-05-04 | 深圳信息职业技术学院 | Glass bottle defect detection method and device based on deep learning and linear detection |
| CN113240626A (en) * | 2021-04-08 | 2021-08-10 | 西安电子科技大学 | Neural network-based method for detecting and classifying concave-convex flaws of glass cover plate |
| CN113240626B (en) * | 2021-04-08 | 2023-07-11 | 西安电子科技大学 | Glass cover plate concave-convex type flaw detection and classification method based on neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101315338A (en) | Defect detecting system and method for glass product | |
| CN101858734B (en) | Method for detecting PET bottleneck quality | |
| CN102854192B (en) | Method for detecting apple surface defect based on apple surface defect detecting system | |
| CN101063662A (en) | Method for detecting empty bottle bottom defect and device for detecting empty bottle bottom defect based on DSP | |
| CN107052086A (en) | Stamping parts surface defect detection apparatus and detection method based on 3D vision | |
| CN101105459A (en) | Empty bottle mouth defect inspection method and device | |
| CN105073380B (en) | For checking the method and system of the quality of colour of unordered bottle base | |
| CN103163144B (en) | For monitoring the method for the quality of the prime coat be applied on motorcar body before spraying paint | |
| CN104237252A (en) | Machine-vision-based method and device for intelligently detecting surface micro-defects of product | |
| CN109001212A (en) | A kind of stainless steel soup ladle defect inspection method based on machine vision | |
| CN105784722B (en) | On a kind of assembly line in medicinal liquid bottle visible foreign matters detection method and system | |
| KR20080009720A (en) | Methods, devices and programs for exploration of knots in wood | |
| CN104634791A (en) | Omnidirectional and visual on-line detection system of surface defects of strip-shaped workpiece | |
| KR101505738B1 (en) | Appartus and method for detecting defect | |
| CN206981462U (en) | Stamping parts surface defect detection apparatus based on 3D vision | |
| CN100547394C (en) | Fruit quality detection system based on image information fusion technology | |
| CN108120353B (en) | Fire transmission hole detection device and method based on intelligent camera | |
| CN103698337A (en) | Method for automatically detecting adhesion defect of label of flat wine bottle | |
| CN103499590A (en) | Method and system for detecting and screening end defects in circular parts | |
| CN110434090A (en) | Carrot surface flaw detection method | |
| CN112837287B (en) | Method and device for extracting defect area on board surface | |
| CN104992446B (en) | The image split-joint method of non-linear illumination adaptive and its realize system | |
| CN103759644A (en) | Separating and refining type intelligent optical filter surface defect detecting method | |
| CN203245133U (en) | Lens defect detecting system | |
| CN201689062U (en) | PET bottle mouth quality detection device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20081203 |