CN206912816U - Identify the device of mechanical workpieces pose - Google Patents
Identify the device of mechanical workpieces pose Download PDFInfo
- Publication number
- CN206912816U CN206912816U CN201720471254.0U CN201720471254U CN206912816U CN 206912816 U CN206912816 U CN 206912816U CN 201720471254 U CN201720471254 U CN 201720471254U CN 206912816 U CN206912816 U CN 206912816U
- Authority
- CN
- China
- Prior art keywords
- mechanical workpieces
- fixture
- pose
- model
- utility
- 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
- Length Measuring Devices By Optical Means (AREA)
Abstract
Digital control processing and detection device application field are the utility model is related to, to propose the method for identification mechanical workpieces pose, realizes processing or the intellectuality of detection device and automation.The technical solution adopted in the utility model is, identify the device of mechanical workpieces pose, it is made up of metal fixture, colored industrial camera, tested mechanical workpieces are fixed on metal fixture, colored industrial camera is fixed on the revolving body of three coordinate measuring machine, and revolving body drives camera to realize that diverse location different angle carries out high-precision detection to mechanical workpieces.The metal fixture is copper fixture.The utility model is mainly used in digital control processing and design of testing device occasions that manufacture.
Description
Technical field
Digital control processing and detection device application field are the utility model is related to, specifically, is related to the knowledge of mechanical workpieces pose
Other apparatus.
Background technology
In order to improve the security of numerically controlled machine equipment and intellectuality, to the part position processing or detect and posture
What the problem of identification is a major issue for needing to solve, and the research work of early stage is primarily directed in two-dimensional scene was carried out.
Rafiq Ahmad et al. are based on two three dimensions methods of machine vision applications, and Virtual Space emulates to real space,
Anticollision detection and the generation of safe trajectory are carried out in Virtual Space.Tian et al. propose using video camera in Virtual Space and
The method that same position is taken pictures to part and carries out image recognition in real space, while propose one kind and be based on image
The scaling method of identification, but the space that the scaling method is applicable is small, the three-dimensional information of extraction is few, and application compares limitation.
Zhao Jincai carries out pose identification to the part on three-dimensional coordinates measurement machine platform using geometric moment, but just with single in testing
Video camera, extraction of spatial information is less, and to the concerns mandate between camera coordinate system and three coordinate measuring machine coordinate system
Compare harsh, the identification of the various poses of part can not be realized.Parts of the Han Ning using two cameras to three coordinate measuring machine
Pose identification is carried out, can extract to certain limit three-dimensional spatial information, but still can not realize complete to the various postures of part
The extraction of director space information, positioning precision be not also high.
In actual processing or measurement process, many mechanical workpieces are all to need fixture to be fixed on measurement or processing platform.
It can so cause many workpiece and metal fixture can not be distinguished from black and white industrial camera, also None- identified is specifically detected
Be what mechanical workpieces and specific position and attitude, be also just difficult to processing or the intellectuality of detection device and automatic
Change.
The content of the invention
For overcome the deficiencies in the prior art, the utility model is directed to the method for identifying mechanical workpieces pose, realizes and add
The intellectuality and automation of work or detection device.The technical solution adopted in the utility model is to identify the dress of mechanical workpieces pose
Put, be made up of metal fixture, colored industrial camera, tested mechanical workpieces are fixed on metal fixture, and colored industrial camera is fixed
On the revolving body of three coordinate measuring machine, it is high-precision that revolving body drives camera to realize that diverse location different angle is carried out to mechanical workpieces
The detection of degree.
The metal fixture is copper fixture.
The characteristics of the utility model and beneficial effect are:
1st, fixture and mechanical workpieces can be directly distinguished from the color of workpiece using colored industrial camera, is follow-up knowledge
The position of other mechanical workpieces and posture are laid a good foundation.
2nd, the copper fixture used, it ensure that fixture has obvious brassy so that fixture and common mechanical workpieces
The material such as material such as aluminium, steel has obvious color distinction so that fixture and workpiece can be preferably distinguished.
3rd, colored industrial camera is fixed on the revolving body of three coordinate measuring machine, and camera can be driven to realize diverse location not
High-precision detection is carried out to mechanical workpieces with angle.
4th, vision-based detection model is established by structure light and camera collective effect, can effectively extracts the three of various workpiece
Dimensional feature point cloud information, laid a good foundation for follow-up identification.
5th, copper fixture has obvious color, while has stronger mechanical strength again, it is not easy to deform upon, and
And cost is also relatively suitable.
6th, device of the present utility model can realize the pose identification of common mechanical workpiece, have preferable practicality.
Brief description of the drawings:
Fig. 1 vision system installation drawings.
Fig. 2 mechanical workpieces and fixture schematic diagram.
Embodiment
In view of the above-mentioned problems, the utility model proposes a kind of mechanical workpieces pose identification device.The present apparatus mainly has one
Platform colour industrial camera and copper fixture composition.Illustrated below by taking three coordinate measuring machine as an example, but it is of the present utility model
Using being not limited to three coordinate measuring machine.Colored industrial camera, which is fixed on the revolving body of three coordinate measuring machine, sees Fig. 1, by returning
Swivel drives camera to realize and the mechanical workpieces on measuring table is detected in diverse location different angle, so as to extract more
Comprehensive three-dimensional feature information.More stable be placed on measuring table could be realized when mechanical workpieces need fixture to fix
When, the utility model proposes mechanical workpieces are fixed using copper fixture, copper fixture present brassy can with it is tested
The mechanical workpieces of most of materials of amount are distinguished from color.When colored industrial camera from some position and angle to machine
When tool workpiece is detected, two width pictures are shot respectively first, a width picture does not have structure light such as laser etc., and a width has in addition
Structure light.Never determine that those pixels belong to brassy in the picture of structure light, it is the imaging of fixture to illustrate these pixels,
And then determine the imaging region of mechanical workpieces.Finally utilize mechanical workpieces imaging and the information meter of other width structure light imaging
The three-dimensional point cloud information of extraction mechanical workpieces is calculated, for example is calculated using models such as laser triangulations.
There are three kinds of three-dimensional system of coordinates in the real space of three coordinate measuring machine collision avoidance system, be three-dimensional coordinates measurement respectively
Machine coordinate system, visual coordinate system, part coordinate system.Visual coordinate system establishes part coordinate system and three coordinate measuring machine coordinate system
Between contact.All it is by demarcating the transformational relation established between three coordinate systems.
It is described in detail with reference to the accompanying drawings and examples to of the present utility model.
The utility model proposes a kind of mechanical workpieces pose identification device.The device design of utility model is with three coordinates
Measuring machine is platform, but the vision system application shown in Fig. 1 is not limited to three-dimensional coordinates measurement machine platform, is widely used in other numerical controls
Processing and detection device.The present apparatus is mainly made up of a colored industrial camera and copper fixture.Below with three-dimensional coordinates measurement
Illustrated exemplified by machine, but application of the present utility model is not limited to three coordinate measuring machine.Colored industrial camera is fixed on three seats
See Fig. 1 on the revolving body of co-ordinate measuring machine, drive camera to realize in diverse location different angle on measuring table by revolving body
Mechanical workpieces detected, so as to extract more comprehensive three-dimensional feature information.When mechanical workpieces need fixture to fix ability
Realize that the utility model proposes mechanical workpieces are consolidated using copper fixture when being placed on measuring table more stablely
Fixed, copper fixture is presented brassy and can distinguished with the mechanical workpieces of measured most of materials from color, such as Fig. 2
Shown, argenteous metal ball position workpiece for measurement, brassy below is fixture, and application fixture and workpiece of the present utility model are not
It is limited to shown in Fig. 2.When colored industrial camera detects from some position and angle to mechanical workpieces, shoot respectively first
Two width pictures, a width picture do not have structure light such as laser etc., and a width has structure light in addition.Never in the picture of structure light really
Make those pixels and belong to brassy, illustrate that these pixels are the imaging of fixture, and then determine the imaging region of mechanical workpieces.
The three-dimensional point cloud information of extraction mechanical workpieces is finally calculated using the information of mechanical workpieces imaging and other width structure light imaging,
For example calculated using models such as laser triangulations.
Specific implementation step is:
1) determine that mechanical workpieces are placed on the particular location of measurement or processing space by structure light surface sweeping first.Such as figure
1 three coordinate measuring machine revolving body drives the whole measuring table of vision system surface sweeping, removes the point cloud of measurement this TV station in itself, residue is just
It is the point cloud of mechanical workpieces and fixture.
2) according to the mechanical workpieces of (1) detection and the particular location of fixture, then go out around mechanical workpieces and fixture automatic design
Different test positions and angle.
3) two width pictures will be gathered for each position and angle colour industrial camera, a width is no structure light as swashed
The picture of light, a width are the pictures for having structure light.Distinguished from the first width picture those pixel regions be mechanical workpieces and
Those pixel regions are mechanical workpieces.Then the structure optical information in mechanical workpieces region is carried out threedimensional model calculating, for example swashed
Light trigonometry etc., calculate the three-dimensional feature cloud data of mechanical workpieces.
4) known according to the cloud data of the three dimensional point cloud of the mechanical workpieces calculated in (3) and CAD model generation
Do not match, for example be identified according to Curvature Operator, or registration etc. is carried out according to ICP algorithm.Finally determine to be measured or processing
Mechanical workpieces and position and attitude.
As shown in figure 1, the utility model includes (being to detailed description of the present utility model below, being please comprehensively intactly situated between
Continue whole technology contents of the present utility model, can be more for several embodiments, or illustrate that if more than one embodiment
A little positions can be replaced using other structures, the word of every description, correlation should all be found in Figure of description
Content, the number of being numbered with preferably on accompanying drawing, in explanatory note, also put on corresponding with accompanying drawing number).
Claims (2)
1. a kind of device for identifying mechanical workpieces pose, it is characterized in that, it is made up of metal fixture, colored industrial camera, is tested machine
Tool workpiece is fixed on metal fixture, and colored industrial camera is fixed on the revolving body of three coordinate measuring machine, and revolving body drives phase
Machine realizes that diverse location different angle carries out high-precision detection to mechanical workpieces.
2. the device of identification mechanical workpieces pose as claimed in claim 1, it is characterized in that, the metal fixture is copper folder
Tool.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201720471254.0U CN206912816U (en) | 2017-04-30 | 2017-04-30 | Identify the device of mechanical workpieces pose |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201720471254.0U CN206912816U (en) | 2017-04-30 | 2017-04-30 | Identify the device of mechanical workpieces pose |
Publications (1)
Publication Number | Publication Date |
---|---|
CN206912816U true CN206912816U (en) | 2018-01-23 |
Family
ID=61306381
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201720471254.0U Expired - Fee Related CN206912816U (en) | 2017-04-30 | 2017-04-30 | Identify the device of mechanical workpieces pose |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN206912816U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110701999A (en) * | 2018-07-09 | 2020-01-17 | 鸿富锦精密电子(郑州)有限公司 | Vision measurement positioning system, vision measurement positioning method and storage device |
-
2017
- 2017-04-30 CN CN201720471254.0U patent/CN206912816U/en not_active Expired - Fee Related
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110701999A (en) * | 2018-07-09 | 2020-01-17 | 鸿富锦精密电子(郑州)有限公司 | Vision measurement positioning system, vision measurement positioning method and storage device |
CN110701999B (en) * | 2018-07-09 | 2021-08-31 | 郑州富联智能工坊有限公司 | Vision measurement positioning system, vision measurement positioning method and storage device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108555908B (en) | Stacked workpiece posture recognition and pickup method based on RGBD camera | |
CN107590836B (en) | Kinect-based charging pile dynamic identification and positioning method and system | |
CN106524945B (en) | A kind of plane included angle On-line Measuring Method based on mechanical arm and structure light vision | |
Tang et al. | 3D mapping and 6D pose computation for real time augmented reality on cylindrical objects | |
CN110751691B (en) | Automatic pipe fitting grabbing method based on binocular vision | |
CN110211180A (en) | A kind of autonomous grasping means of mechanical arm based on deep learning | |
CN108492017B (en) | Product quality information transmission method based on augmented reality | |
Šuligoj et al. | Object tracking with a multiagent robot system and a stereo vision camera | |
Momeni-k et al. | Height estimation from a single camera view | |
CN116157837A (en) | Calibration method and device for robot | |
Hsu et al. | Development of a faster classification system for metal parts using machine vision under different lighting environments | |
CN112518748A (en) | Automatic grabbing method and system of vision mechanical arm for moving object | |
CN112419429A (en) | Large-scale workpiece surface defect detection calibration method based on multiple viewing angles | |
CN111598172A (en) | Dynamic target grabbing posture rapid detection method based on heterogeneous deep network fusion | |
Wang et al. | Error analysis and improved calibration algorithm for LED chip localization system based on visual feedback | |
Nissler et al. | Evaluation and improvement of global pose estimation with multiple apriltags for industrial manipulators | |
CN206912816U (en) | Identify the device of mechanical workpieces pose | |
Lin et al. | Vision based object grasping of industrial manipulator | |
CN107020545A (en) | The apparatus and method for recognizing mechanical workpieces pose | |
Oh et al. | Development of structured light based bin picking system using primitive models | |
Fan et al. | An automatic robot unstacking system based on binocular stereo vision | |
Yang et al. | A coaxial vision assembly algorithm for un-centripetal holes on large-scale stereo workpiece using multiple-dof robot | |
Kim et al. | Construction and inspection management system using mobile augmented reality | |
JP2011174891A (en) | Device and method for measuring position and attitude, and program | |
Fröhlig et al. | Three-dimensional pose estimation of deformable linear object tips based on a low-cost, two-dimensional sensor setup and AI-based evaluation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180123 Termination date: 20180430 |