CN109034146A - Automobile manufacture production line derives from workpiece visual identity method - Google Patents
Automobile manufacture production line derives from workpiece visual identity method Download PDFInfo
- Publication number
- CN109034146A CN109034146A CN201810914249.1A CN201810914249A CN109034146A CN 109034146 A CN109034146 A CN 109034146A CN 201810914249 A CN201810914249 A CN 201810914249A CN 109034146 A CN109034146 A CN 109034146A
- Authority
- CN
- China
- Prior art keywords
- workpiece
- derives
- production line
- visual
- automobile manufacture
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses automobile manufacture production lines to derive from workpiece visual identity method, quickly identifies for being derived from before workpiece feeds intake in automobile manufacture production line, including derives from workpiece image acquisition;The analysis of feature point;It takes pictures identification;Workpiece model is derived from judgement.Derivation workpiece visual identity method of the invention completes vehicle judgement by machine vision, and visual identity method is simply direct, extracts and derives from the highest design feature of workpiece visual recognition degree, can complete identification by simple image recognition, recognition efficiency is high;It is versatile, it is easy to spread.
Description
Technical field
The present invention relates to automobile manufacturing fields more particularly to a kind of automobile manufacture production line to derive from workpiece visual identity side
Method.
Background technique
Growing with personalized consumption demand, auto manufacturing also gradually starts high-volume, less flexible to small quantities of
The Model Transformation of amount, high flexibility.Consumer promotes each automaker in basic vehicle the diversified demand of automobile product
On the basis of, constantly remodeling innovation, often a basic vehicle will derive it is high, medium and low match three kinds of different editions, in different versions
It is possible to will appear more kinds of derivative versions in this.These derivative version foundation structures are mostly consistent, but have in respective location
Difference, or different model engine is carried, or installation different model seat, or install other auxiliary function of dominant accessories additional.
In order to meet derivative version otherness design or size change different and to the progress of certain structural members, or installation
Reinforcer or reserving installation hole slot, the similar workpiece of these basic structures are referred to as deriving from workpiece.Workpiece is derived from identical
Or similar foundation structure, often only part is variant.Therefore, when deriving from the delivery material of workpiece, it is easy to appear wrong material
The problem of.
In actual production, high due to deriving from the similarity between workpiece, the link that feeds intake is error-prone.It is common to derive from work
Part differentiating method is manual identified first, time-consuming and laborious, restricts production line automation transition.Second is by specific position
Close to switch, difference is different from the contact point of workpiece close to switching, and derives from workpiece to approach the unlike signal of switch feedback and distinguish
Model.Due to needing to cooperate the fine positioning tool of workpiece close to the use of switch, and the versatility of fine positioning tool generally compared with
Difference, to realize that deriving from workpiece identification needs to design a variety of toolings, positions, equipment cost investment is larger, and recognition efficiency is low, sternly piece by piece
Ghost image rings subsequent beat.
Summary of the invention
The object of the present invention is to provide one kind to derive from workpiece visual identity method for automobile manufacture production line.
For achieving the above object, the technical scheme is that automobile manufacture production line derives from workpiece visual identity
Method quickly identifies for being derived from before workpiece feeds intake in automobile manufacture production line, includes the following steps:
Step 1 derives from workpiece image acquisition
It determines and derives from piece count, acquire all derivation workpiece when feeding intake state, visual angle is in the flat image of top;
Step 2, the analysis of feature point
Based on derive from workpiece planarization image, analyze every kind derive from workpiece can visual discrimination difference, determine derive from workpiece spy
Different point;
Step 3, identification of taking pictures
Camera is fixed on above the station that feeds intake, and is carried out entirety to the workpiece delivered in place and is taken pictures, visual processing unit identifies photo
The feature point of middle appearance;
Step 4 judges to derive from workpiece model
The feature point and the feature point of each derivation workpiece determined in step 2 are checked one by one, determination is thrown at this time
Expect to derive from workpiece model on station.
Further, in step 1 acquisition plane image visual angle to derive from workpiece distance be robot feeding working distance
From.
Further, in step 2 can visual discrimination difference include no add-on parts, whether there is or not reinforcer, have it is non-porous
Slot, size difference, hole shape difference.
Further, camera is more in step 3, and every camera respectively takes pictures to the regional area of workpiece.
Further, the derivation workpiece of every kind of vehicle determines a special point in step 2.
Further, the area-of-interest in step 3 near visual processing unit identification feature point, does not need again complete
At the identification of whole picture photo.
Further, visual processing unit identifies the multiple photos that more cameras are shot by width, recognizes any feature point
Position can stop, and no longer execute the identification of other photos.
The beneficial effects of the present invention are:
Derivation workpiece visual identity method of the invention completes vehicle judgement by machine vision, and visual identity method is simply straight
It connects, extracts and derive from the highest design feature of workpiece visual recognition degree, identification, identification effect can be completed by simple image recognition
Rate is high;It is versatile, it is easy to spread.
Specific embodiment
Technical solution is clearly and completely described below in conjunction with the embodiment of the present invention.
The vehicle cab recognition for deriving from workpiece is first converted to detecting the presence of for feature point, then benefit by mentality of designing of the invention
It completes to detect with vision technique, finally the identification information further according to feature point retrodicts the vehicle model information for deriving from workpiece.
The embodiment of the present invention is as follows, and automobile manufacture production line derives from workpiece visual identity method, raw for automobile manufacture
It derives from producing line before workpiece feeds intake and quickly identifies, include the following steps:
Step 1 derives from workpiece image acquisition
It determines first and derives from piece count, acquire all derivation workpiece when feeding intake state, visual angle is in the flat image of top.
In actual production, workpiece generally passes through truck and is delivered, workpiece be placed on the posture on truck will convenient for robot crawl and
Assembly.Flat image needs are acquired in strict accordance with posture of the workpiece on truck.It acquires arriving for the visual angle of image and derives from work
The distance of part is the operating distance of robot feeding.
Step 2, the analysis of feature point
Based on workpiece planarization image is derived from, analyze every kind derive from workpiece can visual discrimination difference, specifically can be that whether there is or not attached
Add part, whether there is or not reinforcer, have no hole slot, size difference, hole shape difference, every kind of derivation workpiece determines a special point, special
Sign point will preferentially select most obvious, difference most easy to identify, to lower subsequent image identification difficulty.
Step 3, identification of taking pictures
Camera is fixed on above the station that feeds intake, the view position of acquisition plane image in corresponding step 1, in place to delivery
Workpiece carry out it is whole takes pictures, after photo is transmitted to visual processing unit, visual processing unit is not necessarily to carry out at complexity picture
Reason.The feature point determined in step 2 shifts to an earlier date typing visual processing unit, and visual processing unit identifies searching in photo
These feature points.The recognizer of feature point, which recognizes any feature point, to be stopped, and do not need to complete whole picture photograph again
The identification of piece, can effectively save data processing time, improve recognition efficiency.
Biggish in special circumstances in workpiece volume, the camera of Image Acquisition can be more, the position point of every camera
Not Dui Ying it is possible that feature point region, it is only necessary to be respectively completed and take pictures to the regional area, and to being not in spy
The region for levying point, does not need identification of taking pictures.More secondary photos of more camera shootings are also executed and recognize any feature point
The method that can stop, it is not necessary to completing the identification of all photos.
The case where more concentrating further for feature point region can also be improved using the method locally taken pictures
The recognition efficiency of visual processing unit.
Step 4 judges to derive from workpiece model
The feature point and the feature point of each derivation workpiece determined in step 2 are checked one by one, feature point and
The derivation workpiece model that step 3 recognition result is consistent is exactly the model for feeding intake derive from workpiece on station at this time.
Described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on the present invention
In embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Claims (7)
1. automobile manufacture production line derives from workpiece visual identity method, fast before workpiece feeds intake for deriving from automobile manufacture production line
Speed identification, which comprises the steps of:
Step 1 derives from workpiece image acquisition
It determines and derives from piece count, acquire all derivation workpiece when feeding intake state, visual angle is in the flat image of top;
Step 2, the analysis of feature point
Based on derive from workpiece planarization image, analyze every kind derive from workpiece can visual discrimination difference, determine derive from workpiece spy
Different point;
Step 3, identification of taking pictures
Camera is fixed on above the station that feeds intake, and is taken pictures to the workpiece delivered in place, and visual processing unit identifies in photo
Existing feature point;
Step 4 judges to derive from workpiece model
The feature point and the feature point of each derivation workpiece determined in step 2 are checked one by one, determination is thrown at this time
Expect to derive from workpiece model on station.
2. automobile manufacture production line according to claim 1 derives from workpiece visual identity method, which is characterized in that step 1
The distance to derivation workpiece at the visual angle of middle acquisition plane image is the operating distance of robot feeding.
3. automobile manufacture production line according to claim 1 derives from workpiece visual identity method, which is characterized in that step 2
In can visual discrimination difference include no add-on parts, whether there is or not reinforcer, have no hole slot, size difference, hole shape difference.
4. automobile manufacture production line according to claim 1 derives from workpiece visual identity method, which is characterized in that step 2
In the derivation workpiece of every kind of vehicle determine a special point.
5. automobile manufacture production line according to claim 1 derives from workpiece visual identity method, which is characterized in that step 3
Middle camera is more, and every camera individually takes pictures to the regional area of workpiece.
6. automobile manufacture production line according to claim 1 derives from workpiece visual identity method, which is characterized in that step 3
Area-of-interest near middle visual processing unit identification feature point no longer executes the identification of photo remaining area.
7. automobile manufacture production line according to claim 5 derives from workpiece visual identity method, which is characterized in that at vision
Reason unit identifies that recognizing any feature point can stop, and no longer execute it by width to the multiple photos that more cameras are shot
The identification of its photo.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810914249.1A CN109034146A (en) | 2018-08-13 | 2018-08-13 | Automobile manufacture production line derives from workpiece visual identity method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810914249.1A CN109034146A (en) | 2018-08-13 | 2018-08-13 | Automobile manufacture production line derives from workpiece visual identity method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109034146A true CN109034146A (en) | 2018-12-18 |
Family
ID=64632951
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810914249.1A Pending CN109034146A (en) | 2018-08-13 | 2018-08-13 | Automobile manufacture production line derives from workpiece visual identity method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109034146A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110163305A (en) * | 2019-06-24 | 2019-08-23 | 上海大众祥源动力供应有限公司 | A kind of door identification device and door identification method for automobile assembly line |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101439452A (en) * | 2008-12-15 | 2009-05-27 | 广州本田汽车有限公司 | Device for preventing mounting vehicle parts by mistake |
WO2015115274A1 (en) * | 2014-01-31 | 2015-08-06 | オムロン株式会社 | Object assessment device, image sensor, and object assessment method |
CN105493143A (en) * | 2013-04-04 | 2016-04-13 | 日本电气株式会社 | Identification method, identification system, matching device, and program |
CN207641883U (en) * | 2017-11-13 | 2018-07-24 | 中国科学院合肥物质科学研究院 | A kind of part stream waterline automatic sorting boxing apparatus of view-based access control model |
-
2018
- 2018-08-13 CN CN201810914249.1A patent/CN109034146A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101439452A (en) * | 2008-12-15 | 2009-05-27 | 广州本田汽车有限公司 | Device for preventing mounting vehicle parts by mistake |
CN105493143A (en) * | 2013-04-04 | 2016-04-13 | 日本电气株式会社 | Identification method, identification system, matching device, and program |
WO2015115274A1 (en) * | 2014-01-31 | 2015-08-06 | オムロン株式会社 | Object assessment device, image sensor, and object assessment method |
CN207641883U (en) * | 2017-11-13 | 2018-07-24 | 中国科学院合肥物质科学研究院 | A kind of part stream waterline automatic sorting boxing apparatus of view-based access control model |
Non-Patent Citations (1)
Title |
---|
郭宝龙等: "《数字图像处理系统工程导论》", 31 July 2012, 西安电子科技大学出版社 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110163305A (en) * | 2019-06-24 | 2019-08-23 | 上海大众祥源动力供应有限公司 | A kind of door identification device and door identification method for automobile assembly line |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2780849B1 (en) | Automated identification of shoe parts | |
EP2780847B1 (en) | Automated identification and assembly of shoe parts | |
CN109186457B (en) | Binocular part identification method and device and production line using device | |
CN106251353A (en) | Weak texture workpiece and the recognition detection method and system of three-dimensional pose thereof | |
CN105225225B (en) | A kind of leather system for automatic marker making method and apparatus based on machine vision | |
CN109604187A (en) | A kind of part Automated Sorting System and method based on computer vision technique | |
CN109015637A (en) | Automobile manufacture production line vision guide charging method | |
CN109034146A (en) | Automobile manufacture production line derives from workpiece visual identity method | |
CN107424148A (en) | Product Visual automatic aligning method | |
CN104182728A (en) | Vehicle logo automatic location and recognition method based on pattern recognition | |
CN109454501A (en) | A kind of lathe on-line monitoring system | |
CN109604468A (en) | A kind of workpiece stamping system and its control method based on machine vision | |
CN108520533B (en) | Workpiece positioning-oriented multi-dimensional feature registration method | |
Luo et al. | A scalable modular architecture of 3D object acquisition for manufacturing automation | |
EP3567626B1 (en) | Mirror die image recognition system | |
CN107315392B (en) | Special-shaped workpiece discharging optimization algorithm of automatic feeder of punch press and tolerance algorithm thereof | |
CN109516130A (en) | Automate secondary boot piece uploading method and system | |
CN114104453A (en) | Non-ferrous metal automatic labeling method and device based on image processing | |
CN113977637A (en) | Robot vision identification grabbing system and method applicable to non-precision work bin | |
CN109457399B (en) | Automatic sewing method, system, device, terminal and medium for open-line editing | |
CN207057036U (en) | A kind of low-voltage air switch the System of Sorting Components based on Robot Visual Servoing | |
CN105405139A (en) | Monocular CCD (Charge Coupled Device) based method and system for rapidly positioning feeding of small-sized glass panel | |
US20240075513A1 (en) | Method for Producing a Device for Moving a Workpiece From a First Tool Into a Second Tool, Electronic Computing Unit, Computer Program Product and Computer-Readable Medium | |
Phyu et al. | Accuracy on photo-model-based clothes recognition | |
CN109551196A (en) | A kind of machine parts'precise assembly system and three-dimensional error measurement method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181218 |