CN103279763B - A kind of wheel hub type automatic identifying method of structure based feature - Google Patents
A kind of wheel hub type automatic identifying method of structure based feature Download PDFInfo
- Publication number
- CN103279763B CN103279763B CN201310221370.3A CN201310221370A CN103279763B CN 103279763 B CN103279763 B CN 103279763B CN 201310221370 A CN201310221370 A CN 201310221370A CN 103279763 B CN103279763 B CN 103279763B
- Authority
- CN
- China
- Prior art keywords
- wheel hub
- wheel
- hub
- database
- type
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of wheel hub type automatic identifying method of structure based feature, (1), wheel hub for certain unknown model, need first through study obtain this wheel hub whole characteristic parameter and stored in database; By splitting the wheel hub visible images collected, extracting the architectural feature of the type wheel hub, utilizing these parameters to realize the automatic identification of different model wheel hub.(2) when, identifying in the scope that error allows with in database deposit wheel hub type parameter make comparisons, the hub type identifying object is obtained when characteristic parameter is all consistent, if there is certain parameter inconsistent, show that this wheel hub fails to identify not exist in database.The feature that the present invention adopts wheel hub the most stable, as identification parameter, effectively can avoid the various erroneous judgements during identification of blank wheel hub, improves the automaticity that wheel hub is produced.
Description
Technical field
The present invention relates to the Iamge Segmentation belonged in field of machine vision, feature extraction and mode identification technology, particularly a kind of wheel hub type automatic identifying method of structure based feature.
Background technology
Machine vision obtains external image by optical imagery and sensor technology, carries out treatment and analysis to image information, replaces human eye and brain realization to the description of objective world object and understanding with machine.Machine vision technique is a comprehensive very strong new and high technology, relates to the field such as image procossing, pattern-recognition, artificial intelligence, automatically control, has the features such as speed is fast, stability is strong, function is many, be widely used in the automated production of modern industry.
Wheel hub is also wheel rim, is important traveling and the stressed member of the motor vehicles such as automobile, train, and the mode at present mainly through casting is produced.First wheel hub is raw-material fusing and refining when producing, then in the mould designed, blank product is obtained by low pressure casting, heat-treat again after X ray non-destructive detecting device Detection job is qualified, high-accuracy mechanical is processed and air tight test, is finally coating process.On the automation production flow line of wheel hub, mixes line production wanted by multi items wheel hub, the problem that first will solve is exactly the identification of wheel hub type, because early stage blank wheel hub X ray Non-Destructive Testing, mid-term semi-manufacture wheel hub machining and last application process, the wheel hub of different model needs different operating process.Address this problem in the past and rely on artificial always, on wheel hub, finish writing model with pen, to automatic production line, select corresponding processing route by staff according to write model again, inefficiency and with high costs, far can not meet the demand of modern production.Visible, the automated production of hub type automatic identification technology to wheel hub based on machine vision is significant.
The domestic research about wheel hub automatic identification technology just launched in recent years.Application publication date be on November 28th, 2007 publication number be that the patent of invention " automatic identification method for hub type " of CN101079107 discloses a kind of hub type recognition technology scheme automatically detected for nondestructive examination checkout equipment, only describe the roughly thinking of wheel hub identification, do not relate to concrete identification step and method, the same trade enters also has the automatic identification to hub type to be studied, propose the construction system of wheel hub identification, give the image processing algorithm and recognition result that identify and use, do not consider the production run of wheel hub on actual flow waterline, the identification parameter adopted is unstable, the recognition methods calculated amount used is large, computing is complicated, poor robustness, do not possess real-time and versatility.
Summary of the invention
The object of this invention is to provide a kind of wheel hub type automatic identifying method of structure based feature, the method algorithm calculated amount is little, strong robustness, execution speed are fast, and can accurately can identify from initial blank product to the semi-manufacture of centre, last finished product the wheel hub of same model, be widely used in the links on wheel hub automatic production line, meet the same intellectuality entering to contribute to realize wheel hub production and administration of the requirement of system real time.
The technical solution adopted in the present invention is:
A wheel hub type automatic identifying method for structure based feature, performing step is:
(1), the Database of the structural characteristic parameter that wheel hub is corresponding;
Step 1, using these characteristic parameters of area of the face shaping of the number of the number of the diameter of wheel hub, spoke, bolt hole, bolt hole region and center pit stored in the basis of characterization of database as the type wheel hub;
(2), the wheel hub identified on production line, obtain the conclusion that there is not this kind of wheel-type in the model of this wheel hub or database:
Step 2, gather the visible images of wheel hub to be identified, be separated the appearance images of wheel hub;
Step 3, find the circumscribed circle and the center of circle thereof of isolating wheel hub, obtain the straight through dimensional data of wheel hub, with the whole wheel hubs deposited in database directly compared with the dimensional data, if there is identical wheel-type, continue to identify, if not, terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Step 4, with the center of circle of wheel hub circumscribed circle for the center of circle, the half of hub size is that diameter draws roundlet, hub area is divided into inner circle and annulus two parts, circle ring area between circumscribed circle and roundlet calculates the number of spoke, the spoke number wheel-type identical with current spoke number is found within the scope of the database limited in step 3, if have, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 5, segmentation roundlet, obtain the number of hub stud holes, within the scope of the database limited in step 4, find bolt hole number wheel-type identical therewith, if having, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 6, segmentation roundlet, obtain the face shaping data of hub stud holes region, bolt hole face shaping wheel-type identical is therewith found within the scope of the database limited in steps of 5, if have, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 7, segmentation roundlet, obtain the area of hub centre bore region, central bore region area wheel-type identical is therewith found within the scope of the database limited in step 6, if have, draw the final model of this wheel hub, if not, draw the conclusion of not this kind of wheel-type in database, terminate identifying.
Further, the Database of the structural characteristic parameter that described wheel hub is corresponding, by study wheel hub to be identified, obtains corresponding structural characteristic parameter and stored in database:
The visible images of step 1, collection wheel hub, therefrom accurately isolates the appearance images of wheel hub;
Step 2, isolated wheel hub image to be processed, find out circumscribed circle and the center of circle thereof of wheel hub, obtain the diameter dimension data of wheel hub;
Step 3, with the center of circle of wheel hub circumscribed circle for the center of circle, the half of hub size is that diameter draws roundlet, hub area is divided into inner circle and annulus two parts;
Step 4, on above-mentioned circle ring area, add up the number of hole on wheel hub image, obtain spoke number destination data;
Wheel hub image in step 5, segmentation roundlet, obtain the area of the number of hub stud holes, the face shaping of bolt hole area and central bore region, and the total data above step obtained is stored in database, as the final identification parameter of the type wheel hub.
Compared with prior art, tool of the present invention has the following advantages:
1, select the geometry appearance feature of wheel hub as whole identification parameters.Wheel hub in a mold Foundry Production time, its geometric configuration is determined substantially, even if through the processing of follow-up precision optical machinery and coating process, these parameters also remain unchanged substantially, select them as basis of characterization, can guarantee wheel hub links in process of production can be accurately identified, effectively prevent other technology and use the gray feature of wheel hub image as drawback during identification parameter, as there being particular requirement to illumination condition, wheel hub can only be applicable to produce before certain specific link.
2, carry out in order during five characteristic parameter identification, first the approximate range of wheel hub to be identified is determined by hub size, secondly the number by means of spoke reduces identified region further, the basis that the two determines is investigated the number of bolt hole, the face shaping of bolt hole region and the area of central bore region again, as long as above parameter has one not mate in a database namely terminate identifying, draw the conclusion that this wheel hub fails to identify, only have when they whole identical time finally could determine the model of wheel hub.Above-mentioned identification step has effectively used for reference thinking during artificial cognition hub type, first the most obviously, the most stable feature, successively reduce matching range with going forward one by one, both reduced the operand of identifying, in turn ensure that the accuracy of recognition result, be obviously better than conventional pattern classification rcognition method.
When 3, extracting the characteristic parameter of wheel hub by image Segmentation Technology, take into full account the geometrical feature of wheel hub structure, this priori is incorporated in the step of Iamge Segmentation, effectively eliminate the erroneous judgement that blank wheel hub and semi-manufacture, finished product wheel hub difference in appearance may cause, improve recognition accuracy of the present invention, expand the scope of application of the present invention.
Accompanying drawing explanation
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in further detail.
Fig. 1 is the system flowchart of learning training wheel hub of the present invention;
Fig. 2 is the system flowchart of identification hub type of the present invention.
Embodiment
Below in conjunction with accompanying drawing and example, the present invention is further described.
The wheel hub type automatic identifying method of a kind of structure based feature disclosed by the invention, concrete enforcement comprises the learning training to wheel hub and automatic identification two large steps, as shown in Figure 1.Before a kind of new wheel hub is identified, first to be obtained whole characteristic parameters of the type wheel hub by learning training, comprise the following steps:
The visible images of step 1, collection wheel hub, therefrom accurately isolate the appearance images of wheel hub, concrete mode, the accessory whorl hub image obtained by CCD camera, first will be extracted wheel hub from entire image.Under the irradiation of visible light source, the gradation of image of whole wheel hub is apparently higher than background, and this is highly stable feature when gathering wheel hub image.For this feature, the method that fixed threshold is set is adopted to realize when extracting wheel hub: what be greater than this threshold value is 1, and corresponding to the wheel hub extracted, what be less than this threshold value is 0, corresponding to background.If the segmentation result of setting threshold value when being 60, whole wheel hub is fully extracted out, and the threshold value in embody rule can be determined in the combining image visible light source brightness that gathers environment;
Step 2, process of fitting treatment is justified to the binaryzation result of isolated wheel hub image, obtain circumscribed circle and the center of circle thereof of wheel hub, the diameter of this external diameter of a circle and wheel hub, as first parameter obtained in learning process stored in database;
Step 3, with the above-mentioned center of circle for the center of circle, the half of circumradius is that radius draws roundlet, obtain second circle on wheel hub image, hub area is divided into inner circle and annulus two parts by roundlet, according to the geometry of wheel hub, must contain bolt hole and the center pit of wheel hub in this inner circle completely, the annular region between cylindrical and inner circle is the spoke position of wheel hub;
The hole of image in step 4, directly this circle ring area of statistics, equal by means of each hole area myopia during statistics, position is the feature angularly distributed, remove the interference that when wheel hub is cast, burr produce, the hole number finally obtained and the spoke number of wheel hub, obtain the second parameter of learning process thus;
Step 5, obtain hub diameter, these two characteristic parameters of spoke number after, start to analyze determined image-region in roundlet (comprise roundlet minimum image region), wherein bolt hole position gray scale is brighter than surrounding, central bore region is in the geometric center of image, cast gate position when casting corresponding to wheel hub, gray scale is darker than surrounding.Carry out first time segmentation to image, object finds bolt hole area, be characterized in that gray scale is higher, spatially angularly distribute, and regional is separate, area approximation is equal; Based on this to Image Segmentation Using, split the number of target area and the number of bolt hole, the girth area ratio of target area is as the criterion of identification of bolt hole area shape.The central bore region of wheel hub is found in the second time segmentation of image, is positioned at geometric center and the gray-scale value feature lower than surrounding, records the area of this segmentation result, i.e. last parameter of learning training, the area of central bore region by it.
By will these characteristic parameters of area of the diameter of wheel hub, the number of spoke, the number of bolt hole, the face shaping of bolt hole region and center pit be comprised stored in the basis of characterization of database as the type wheel hub; The wheel hub of same type only needs learning training once, the parameter in concrete identification in calling data storehouse.
Fig. 2 identifies the wheel hub on production line, wheel hub is in concrete identifying, still follow same order, first be from the image of CCD camera shooting, extract wheel hub and obtain its diameter data, with in database deposit whole wheel-type diameter compared with, if have continuation identification identical therewith in the scope that error allows, then do not terminate identifying, draw the conclusion that wheel hub fails to identify.On the basis extracting wheel hub image, at this region inner analysis spoke number, compared with the spoke data in the database reduced in previous step, if having identical, continuation identification, does not then terminate identifying, draws the conclusion that wheel hub fails to identify.Under the prerequisite that hub diameter, spoke number are consistent, image in the roundlet continued, obtain the number of bolt hole, the shape of bolt hole area, compare with the corresponding data in the database of reduced scope, if bolt hole number is consistent, the shape of bolt hole area is also consistent with some wheel-type in database in the scope that error allows, then continue analytic centre's bore region, otherwise end identifying, draws the conclusion that wheel hub fails to identify.The area of last central bore region is consistent with certain wheel-type in database in the scope that error allows, then obtain the final recognition result of this wheel hub, if it cannot mate in the above-mentioned database progressively reduced, then draw and the conclusion that this wheel hub fails to identify terminate identifying.
Identification concrete steps are as follows:
Step 1, gather the visible images of wheel hub to be identified, be separated the appearance images of wheel hub;
Step 2, find the circumscribed circle and the center of circle thereof of isolating wheel hub, obtain the straight through dimensional data of wheel hub, with the whole wheel hubs deposited in database directly compared with the dimensional data, if there is identical wheel-type, continue to identify, if not, terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Step 3, with the center of circle of wheel hub circumscribed circle for the center of circle, the half of hub size is that diameter draws roundlet, hub area is divided into inner circle and annulus two parts, circle ring area between circumscribed circle and roundlet calculates the number of spoke, the spoke number wheel-type identical with current spoke number is found within the scope of the database limited in step 2, if have, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 4, segmentation roundlet, obtain the number of hub stud holes, within the scope of the database limited in step 3, find bolt hole number wheel-type identical therewith, if having, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 5, segmentation roundlet, obtain the face shaping data of hub stud holes region, bolt hole face shaping wheel-type identical is therewith found within the scope of the database limited in step 4, if have, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 6, segmentation roundlet, obtain the area of hub centre bore region, central bore region area wheel-type identical is therewith found within the scope of the database limited in steps of 5, if have, draw the final model of this wheel hub, if not, draw the conclusion of not this kind of wheel-type in database, terminate identifying.
Above-mentioned identifying can according to concrete cutback in demand.Wheel hub type as to be identified on production line is less, or certain several feature is just enough to be distinguished, and can skip some parameter in identification, and the area as central bore region can not compare, or breast wheel spoke number does not compare, and accelerates recognition speed.
Claims (2)
1. a wheel hub type automatic identifying method for structure based feature, is characterized in that: performing step is:
(1), the Database of the structural characteristic parameter that wheel hub is corresponding;
Step 1, using these characteristic parameters of area of the face shaping of the number of the number of the diameter of wheel hub, spoke, bolt hole, bolt hole region and center pit stored in the basis of characterization of database as the type wheel hub;
(2), the wheel hub identified on production line, obtain the conclusion that there is not this kind of wheel-type in the model of this wheel hub or database:
Step 2, gather the visible images of wheel hub to be identified, be separated the appearance images of wheel hub;
Step 3, find the circumscribed circle and the center of circle thereof of isolating wheel hub, obtain the dimensional data of hub diameter, compared with the dimensional data of the whole hub diameters deposited in database, if there is identical wheel-type, continue to identify, if not, terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Step 4, with the center of circle of wheel hub circumscribed circle for the center of circle, the half of hub size is that diameter draws roundlet, hub area is divided into inner circle and annulus two parts, circle ring area between circumscribed circle and roundlet calculates the number of spoke, the spoke number wheel-type identical with current spoke number is found within the scope of the database limited in step 3, if have, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 5, segmentation roundlet, obtain the number of hub stud holes, within the scope of the database limited in step 4, find bolt hole number wheel-type identical therewith, if having, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 6, segmentation roundlet, obtain the face shaping data of hub stud holes region, bolt hole face shaping wheel-type identical is therewith found within the scope of the database limited in steps of 5, if have, continue to identify, then do not terminate identifying, draw the conclusion that there is not this kind of wheel-type in database;
Wheel hub image in step 7, segmentation roundlet, obtain the area of hub centre bore region, central bore region area wheel-type identical is therewith found within the scope of the database limited in step 6, if have, draw the final model of this wheel hub, if not, draw the conclusion of not this kind of wheel-type in database, terminate identifying.
2. the wheel hub type automatic identifying method of a kind of structure based feature as claimed in claim 1, it is characterized in that: the Database of the structural characteristic parameter that described wheel hub is corresponding is by learning wheel hub to be identified, obtain corresponding structural characteristic parameter and stored in database, concrete steps are:
The visible images of step 1, collection wheel hub, therefrom accurately isolates the appearance images of wheel hub;
Step 2, isolated wheel hub image to be processed, find out circumscribed circle and the center of circle thereof of wheel hub, obtain the diameter dimension data of wheel hub;
Step 3, with the center of circle of wheel hub circumscribed circle for the center of circle, the half of hub size is that diameter draws roundlet, hub area is divided into inner circle and annulus two parts;
Step 4, on above-mentioned circle ring area, add up the number of hole on wheel hub image, obtain spoke number destination data;
Wheel hub image in step 5, segmentation roundlet, obtain the area of the number of hub stud holes, the face shaping of bolt hole area and central bore region, and the total data above step obtained is stored in database, as the final identification parameter of the type wheel hub.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310221370.3A CN103279763B (en) | 2013-05-25 | 2013-05-25 | A kind of wheel hub type automatic identifying method of structure based feature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310221370.3A CN103279763B (en) | 2013-05-25 | 2013-05-25 | A kind of wheel hub type automatic identifying method of structure based feature |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103279763A CN103279763A (en) | 2013-09-04 |
| CN103279763B true CN103279763B (en) | 2016-03-02 |
Family
ID=49062278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310221370.3A Expired - Fee Related CN103279763B (en) | 2013-05-25 | 2013-05-25 | A kind of wheel hub type automatic identifying method of structure based feature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103279763B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105548185B (en) * | 2016-01-08 | 2018-12-11 | 浙江科技学院 | The recognition methods of automotive hub screw hole, covering method and system based on machine vision |
| CN105844047A (en) * | 2016-04-08 | 2016-08-10 | 刘守乾 | Calculation formulas, inspection method and inspection platform for hub refitting comparison |
| CN105867305B (en) * | 2016-05-13 | 2018-06-12 | 南京航空航天大学 | Complex structural member digital control processing realtime monitoring method based on machining feature |
| CN108491841A (en) * | 2018-03-21 | 2018-09-04 | 东南大学 | A kind of automotive hub type identification monitoring management system and method |
| CN109108976A (en) * | 2018-09-10 | 2019-01-01 | 滨州戴森车轮科技有限公司 | A kind of full-automatic identification device of wheel hub and wheel hub automatic identifying method |
| CN109614977A (en) * | 2018-11-14 | 2019-04-12 | 华南理工大学 | A kind of hub type recognition methods |
| CN112288662A (en) * | 2019-08-12 | 2021-01-29 | 中北大学 | Hub X-ray image enhancement method and system |
| CN111390107B (en) * | 2020-04-16 | 2021-10-29 | 杭州喜马拉雅信息科技有限公司 | Sand mold printing method for rotary different-aperture nozzle |
| CN111507404A (en) * | 2020-04-17 | 2020-08-07 | 无锡雪浪数制科技有限公司 | Hub model identification method based on deep vision |
| CN112184600A (en) * | 2020-08-18 | 2021-01-05 | 洛阳中信成像智能科技有限公司 | Hub model identification method in mixed line production line |
| CN113160162B (en) * | 2021-04-14 | 2022-02-18 | 深圳远荣智能制造股份有限公司 | Hole recognition method and device applied to workpiece and hole processing equipment |
| CN113392913B (en) * | 2021-06-21 | 2023-09-29 | 常州大学 | Planar graph matching degree evaluation method, device and system based on boundary feature point set |
| CN115082425A (en) * | 2022-07-19 | 2022-09-20 | 深圳市信润富联数字科技有限公司 | Hub dimension measuring method and device, electronic device and readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1123940A (en) * | 1994-04-29 | 1996-06-05 | 国际商业机器公司 | Produce recognition system |
| US5901247A (en) * | 1993-12-28 | 1999-05-04 | Sandia Corporation | Visual cluster analysis and pattern recognition template and methods |
| US6026189A (en) * | 1997-11-13 | 2000-02-15 | National Research Council Of Canada | Method of recognizing objects within two-dimensional and three-dimensional images |
| CN101079107A (en) * | 2007-06-29 | 2007-11-28 | 丹东华日理学电气有限公司 | Automatic identification method for hub type |
-
2013
- 2013-05-25 CN CN201310221370.3A patent/CN103279763B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5901247A (en) * | 1993-12-28 | 1999-05-04 | Sandia Corporation | Visual cluster analysis and pattern recognition template and methods |
| CN1123940A (en) * | 1994-04-29 | 1996-06-05 | 国际商业机器公司 | Produce recognition system |
| US6026189A (en) * | 1997-11-13 | 2000-02-15 | National Research Council Of Canada | Method of recognizing objects within two-dimensional and three-dimensional images |
| CN101079107A (en) * | 2007-06-29 | 2007-11-28 | 丹东华日理学电气有限公司 | Automatic identification method for hub type |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103279763A (en) | 2013-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103279763B (en) | A kind of wheel hub type automatic identifying method of structure based feature | |
| CN108446678B (en) | Dangerous driving behavior identification method based on skeletal features | |
| CN109490316B (en) | Surface defect detection algorithm based on machine vision | |
| CN104680519B (en) | Seven-piece puzzle recognition methods based on profile and color | |
| CN103499585B (en) | Based on noncontinuity lithium battery film defect inspection method and the device thereof of machine vision | |
| CN109822398B (en) | Numerical control machine tool broken cutter detection system and method based on deep learning | |
| CN107437243B (en) | Tire impurity detection method and device based on X-ray image | |
| CN111189837B (en) | Cigarette appearance online detection method and device | |
| WO2016091016A1 (en) | Nucleus marker watershed transformation-based method for splitting adhered white blood cells | |
| CN102831378B (en) | The detection and tracking method and system of people | |
| CN110942107B (en) | Automatic composite grinding processing characteristic identification method based on part engineering image | |
| CN107016396A (en) | A kind of assembling connecting piece characteristics of image deep learning and recognition methods | |
| CN112907561A (en) | Notebook appearance flaw detection method based on deep learning | |
| CN110866547B (en) | Automatic classification system and method for traditional Chinese medicine decoction pieces based on multiple features and random forests | |
| CN113033620A (en) | Multi-information fusion rotary kiln product quality classification and identification method based on random forest | |
| CN105844282A (en) | Method for detecting defects of fuel injection nozzle O-Ring through line scanning camera | |
| CN111062437A (en) | Bridge structure disease automatic target detection model based on deep learning | |
| CN111260609A (en) | Cigarette appearance defect detection method based on deep learning | |
| CN104259110A (en) | Detection method of ceramic valve element | |
| Zhang et al. | Fabric defect detection based on visual saliency map and SVM | |
| CN105404682A (en) | Digital image content based book retrieval method | |
| CN108021868A (en) | A kind of quick highly reliable circular target detection recognition method | |
| CN111415348A (en) | Method for extracting characteristics of bubbles in automobile brake pipeline | |
| CN104751487B (en) | A kind of moving target detecting method based on the plane discoloration frames of colored RGB tri- difference | |
| CN111178244A (en) | Method for identifying abnormal production scene |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160302 Termination date: 20180525 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |