CN105654086A - Robot key member pose identification method and system based on monocular vision - Google Patents
Robot key member pose identification method and system based on monocular vision Download PDFInfo
- Publication number
- CN105654086A CN105654086A CN201410637167.9A CN201410637167A CN105654086A CN 105654086 A CN105654086 A CN 105654086A CN 201410637167 A CN201410637167 A CN 201410637167A CN 105654086 A CN105654086 A CN 105654086A
- Authority
- CN
- China
- Prior art keywords
- robot
- image
- vision
- pose
- anglec
- 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
Abstract
The invention relates to the technical field of robot vision, and specifically discloses a robot key member pose identification system based on monocular vision. The vision system comprises a camera and a vision measurement system. The camera acquires an image; the vision measurement system receives the image, obtains a pose result through image processing and image analysis and sends the pose result to a robot control system; and the robot control system drives a robot according to the pose result. According to the robot key member pose identification method and system based on the monocular vision, through adoption of a secondary positioning scheme, the influence exerted by lens distortion on the image processing due to non-collinearity between the optical axis of the camera and a measured object is reduced; through indirect measurement, accurate positioning of an object to be positioned which cannot be completely acquired by the camera is realized; through binary processing by use of a dynamic threshold, the influence exerted by illumination on the image processing is reduced; and through search for a positioning object by use of a communication domain, the problem of irregular shape after the object is imaged is solved.
Description
Technical field
The present invention relates to technical field of robot vision, particularly to a kind of robot key component method for recognizing position and attitude based on monocular vision and system.
Background technology
At present, vision localization algorithm cannot meet the industrial requirement to real-time and accuracy. Its core algorithm adopts Hough transformation to detect circle, but for big resolution (as: 2592x1944), successfully detect and once take time longer and stability is not high.
Existing vision localization algorithm there is a problem in that simultaneously:
1, due to camera optical axis and testee not conllinear time, lens distortion is bigger on the impact of image procossing;
2, video camera can not collect the image of target to be positioned completely;
3, illumination is bigger on the impact of image procossing;
4, problem in irregular shape after target imaging to be positioned.
List of references: Chen Baisheng, the new method of a kind of bianry image connected component labeling, the new method of a kind of bianry image connected component labeling, computer engineering and application 2O06.25:46��47.
Summary of the invention
It is contemplated that overcome the defect of existing vision localization algorithm, it is provided that a kind of robot key component method for recognizing position and attitude based on monocular vision and system, it is achieved the real-time of visual identity, accuracy and stability.
For achieving the above object, the present invention is by the following technical solutions:
The present invention provides a kind of robot key component pose identification system based on monocular vision, and including visual system, robot control system, robot, described visual system includes video camera and vision measurement system; Described camera acquisition image; Described vision measurement system receives described image, obtains pose result through image procossing, graphical analysis, and pose result is sent to described robot control system; Described robot control system drives described robot according to pose result.
In some embodiments, described vision measurement system includes image capture module, image processing module, image analysis module, error identification module;Described image capture module, is used for gathering image; Described image processing module, the image of image acquisition and background separation for processing collection; Described image analysis module, for the image of analytical separation and obtain position and the anglec of rotation of appearance target undetermined; Described error identification module, for position and the anglec of rotation of relatively appearance target undetermined, and judges to identify whether to meet error requirements.
In some embodiments, described image processing module includes binarization unit, connected area segmentation unit, thus obtaining the image with background separation.
In some embodiments, described image module, including connective region search unit, Feature Points Matching unit, thus obtaining position and the anglec of rotation of appearance target undetermined.
In some embodiments, described error identification module, the relatively position of appearance target undetermined and the anglec of rotation and standard appearance clarification of objective undetermined, position and the anglec of rotation, it is judged that identify whether to meet error requirements.
The present invention also provides for a kind of robot key component method for recognizing position and attitude based on monocular vision, including step, S1, is made feature holes and target hole and background separation by binaryzation; S2, by search for connected region obtain described feature holes; S3, detect described feature holes distribution whether be round, reject useless point; S4, the feature holes centers of more than three are become a circle by least square fitting; S5, data acquisition, and carry out binaryzation according to threshold values, it is thus achieved that chain code; S6, the assorted point filtered by consecutive points on described chain code, make chain code continuous; S7, chain code is grouped, it is thus achieved that reliable group, and angle-data is averaged, it is thus achieved that decelerator anglec of rotation D.
In some embodiments, the center that center of circle C is decelerator of described circle, radius is the radius of feature holes circular distribution; The region of described data acquisition is with center of circle C for the center of circle, and radius is the radius radius sum with feature holes circular distribution of target hole circular distribution.
The beneficial effects of the present invention is: the present invention, based on the robot key component method for recognizing position and attitude of monocular vision and system, adopts second positioning scheme, reduce the lens distortion impact on image procossing during due to camera optical axis and testee not conllinear; Adopt and indirectly measure, it is achieved video camera can not be collected being accurately positioned of target to be positioned completely; Adopt dynamic threshold to carry out binary conversion treatment, reduce the illumination impact on image procossing; Adopt connective region search location target, problem in irregular shape after solution target imaging.
Accompanying drawing explanation
Fig. 1 is the structural representation of the robot key component pose identification system of one embodiment of the invention.
Fig. 2 is the structural representation of the vision measurement system of one embodiment of the invention.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and the specific embodiments, the present invention is further elaborated. Should be appreciated that specific embodiment described herein is only in order to explain the present invention, and be not construed as limiting the invention.
In describing the invention, term " interior ", " outward ", " longitudinal direction ", " transverse direction ", " on ", D score, " top ", the orientation of the instruction such as " end " or position relationship be based on orientation shown in the drawings or position relationship, it is for only for ease of the description present invention rather than requires that the present invention with specific azimuth configuration and operation, therefore must be not considered as limiting the invention.
As it is shown in figure 1, the present invention proposes a kind of robot key component pose identification system based on monocular vision, including visual system 1, robot control system 2, robot 3, described visual system 1 includes video camera 11 and vision measurement system 12.
Described key component is base and an axle decelerator; Robot control system 2 is robot RC controller. Visual system of the present invention specifically includes industrial camera, permanent light source, vision processor, switch, robot, robot RC controller, handgrip, fast replacing device; Wherein, industrial camera, permanent light source, vision processor composition vision measurement system, vision processor is embedded in robot control cabinet, is communicated by switch between camera, vision processor and robot RC controller.
Described video camera 11 gathers image; Described vision measurement system 12 receives described image, obtains pose result through image procossing, graphical analysis, and pose result is sent to described robot control system 2; Described robot control system 2 drives described robot 3 according to pose result, completes vision assembling.
1st embodiment
As in figure 2 it is shown, in the present invention the 1st embodiment, described vision measurement system 12 includes image capture module 121, image processing module 122, image analysis module 123, error identification module 124.
Wherein,
Described image capture module 121, is used for gathering image.
Described image processing module 122, the image of image acquisition and background separation for processing collection.
Described image analysis module 123, for the image of analytical separation and obtain position and the anglec of rotation of appearance target undetermined.
Described error identification module 124, for position and the anglec of rotation of relatively appearance target undetermined, and judges to identify whether to meet error requirements.
In a preferred embodiment, described image processing module includes binarization unit, connected area segmentation unit, thus obtaining the image with background separation.
Described image module, including connective region search unit, Feature Points Matching unit, thus obtaining position and the anglec of rotation of appearance target undetermined.
Described error identification module, the position of comparison appearance target undetermined and the anglec of rotation and standard appearance clarification of objective undetermined, position and the anglec of rotation, it is judged that identify whether to meet error requirements.
The present invention also proposes a kind of robot key component method for recognizing position and attitude based on monocular vision, including step,
S1, made feature holes and target hole and background separation by binaryzation;
S2, by search for connected region obtain described feature holes;
S3, detect described feature holes distribution whether be round, reject useless point;
S4, the feature holes centers of more than three are become a circle by least square fitting;
S5, data acquisition, and carry out binaryzation according to threshold values, it is thus achieved that chain code;
S6, the assorted point filtered by consecutive points on described chain code, make chain code continuous;
S7, chain code is grouped, it is thus achieved that reliable group, and angle-data is averaged, it is thus achieved that decelerator anglec of rotation D.
2nd embodiment
In the present invention the 2nd embodiment,
S1, made feature holes and target hole and background separation by binaryzation;
S2, by search for connected region obtain described feature holes;
S3, detect described feature holes distribution whether be round, reject useless point;
S4, the feature holes centers of more than three are become a circle by least square fitting; The center that center of circle C is decelerator of described circle, radius is the radius of feature holes circular distribution;
S5, data acquisition, its region be with center of circle C be the center of circle, radius be target hole circular distribution the radius sum of radius and feature holes circular distribution, and carry out binaryzation according to threshold values, it is thus achieved that chain code;
S6, the assorted point filtered by consecutive points on described chain code, make chain code continuous;
S7, chain code is grouped, it is thus achieved that reliable group, and angle-data is averaged, it is thus achieved that decelerator anglec of rotation D.
In concrete application, the embodiment of the present invention adopts second positioning, first determines whether the position of target to be positioned; Secondly, mobile camera is to the surface of target to be positioned; Again, position and the anglec of rotation of target to be positioned are oriented.
Indirectly measure, by the local feature of target to be positioned, infer the global feature of target to be positioned.
Carry out binary conversion treatment by dynamic threshold, namely carry out binary conversion treatment by the mean flow rate around current point.
Connective region search location target, first, screens according to the area of connected domain; Secondly, screen according to the radius of connected domain; Again, the shape according to connected domain is screened; Finally, handsome choosing is carried out according to the position relationship between connected domain.
The robot key component method for recognizing position and attitude based on monocular vision of present invention proposition and system, adopt second positioning scheme, reduces the lens distortion impact on image procossing during due to camera optical axis and testee not conllinear; Adopt and indirectly measure, it is achieved video camera can not be collected being accurately positioned of target to be positioned completely; Adopt dynamic threshold to carry out binary conversion treatment, reduce the illumination impact on image procossing; Adopt connective region search location target, problem in irregular shape after solution target imaging.
The detailed description of the invention of present invention described above, is not intended that limiting the scope of the present invention. Any technology according to the present invention is conceived done various other and is changed accordingly and deformation, should be included in the protection domain of the claims in the present invention.
Claims (7)
1., based on a robot key component pose identification system for monocular vision, including visual system, robot control system, robot, described visual system includes video camera and vision measurement system;
Described camera acquisition image; Described vision measurement system receives described image, obtains pose result through image procossing, graphical analysis, and pose result is sent to described robot control system; Described robot control system drives described robot according to pose result.
2. robot as claimed in claim 1 key component pose identification system, it is characterised in that described vision measurement system includes image capture module, image processing module, image analysis module, error identification module;
Described image capture module, is used for gathering image;
Described image processing module, the image of image acquisition and background separation for processing collection;
Described image analysis module, for the image of analytical separation and obtain position and the anglec of rotation of appearance target undetermined;
Described error identification module, for position and the anglec of rotation of relatively appearance target undetermined, and judges to identify whether to meet error requirements.
3. robot as claimed in claim 2 key component pose identification system, it is characterised in that described image processing module includes binarization unit, connected area segmentation unit, thus obtaining the image with background separation.
4. robot as claimed in claim 2 key component pose identification system, it is characterised in that described image module, including connective region search unit, Feature Points Matching unit, thus obtaining position and the anglec of rotation of appearance target undetermined.
5. robot as claimed in claim 2 key component pose identification system, it is characterized in that, described error identification module, the position of comparison appearance target undetermined and the anglec of rotation and standard appearance clarification of objective undetermined, position and the anglec of rotation, it is judged that identify whether to meet error requirements.
6. based on the robot key component method for recognizing position and attitude of monocular vision, it is characterised in that include step,
S1, made feature holes and target hole and background separation by binaryzation;
S2, by search for connected region obtain described feature holes;
S3, detect described feature holes distribution whether be round, reject useless point;
S4, the feature holes centers of more than three are become a circle by least square fitting;
S5, data acquisition, and carry out binaryzation according to threshold values, it is thus achieved that chain code;
S6, the assorted point filtered by consecutive points on described chain code, make chain code continuous;
S7, chain code is grouped, it is thus achieved that reliable group, and angle-data is averaged, it is thus achieved that decelerator anglec of rotation D.
7. the robot key component method for recognizing position and attitude based on monocular vision as claimed in claim 6, it is characterised in that
The center that center of circle C is decelerator of described circle, radius is the radius of feature holes circular distribution;
The region of described data acquisition is with center of circle C for the center of circle, and radius is the radius radius sum with feature holes circular distribution of target hole circular distribution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410637167.9A CN105654086A (en) | 2014-11-12 | 2014-11-12 | Robot key member pose identification method and system based on monocular vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410637167.9A CN105654086A (en) | 2014-11-12 | 2014-11-12 | Robot key member pose identification method and system based on monocular vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105654086A true CN105654086A (en) | 2016-06-08 |
Family
ID=56479309
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410637167.9A Pending CN105654086A (en) | 2014-11-12 | 2014-11-12 | Robot key member pose identification method and system based on monocular vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105654086A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109238142A (en) * | 2018-09-27 | 2019-01-18 | 嘉善苏尼电子科技有限公司 | A kind of device for visual identification for the positioning of self-lubricating bearing shaft sleeve workpiece graphite pores |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101893894A (en) * | 2010-06-30 | 2010-11-24 | 上海交通大学 | Reconfigurable miniature mobile robot cluster locating and tracking system |
CN102284769A (en) * | 2011-08-05 | 2011-12-21 | 上海交通大学 | System and method for initial welding position identification of robot based on monocular vision sensing |
CN103604426A (en) * | 2013-12-02 | 2014-02-26 | 苏州大学张家港工业技术研究院 | Estimation method and apparatus for poses of mobile robot |
-
2014
- 2014-11-12 CN CN201410637167.9A patent/CN105654086A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101893894A (en) * | 2010-06-30 | 2010-11-24 | 上海交通大学 | Reconfigurable miniature mobile robot cluster locating and tracking system |
CN102284769A (en) * | 2011-08-05 | 2011-12-21 | 上海交通大学 | System and method for initial welding position identification of robot based on monocular vision sensing |
CN103604426A (en) * | 2013-12-02 | 2014-02-26 | 苏州大学张家港工业技术研究院 | Estimation method and apparatus for poses of mobile robot |
Non-Patent Citations (2)
Title |
---|
李婷 等: "基于机器视觉的圆定位技术研究", 《计算机工程与应用》 * |
李阳: "基于视觉的移动机器人运动目标跟踪技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109238142A (en) * | 2018-09-27 | 2019-01-18 | 嘉善苏尼电子科技有限公司 | A kind of device for visual identification for the positioning of self-lubricating bearing shaft sleeve workpiece graphite pores |
CN109238142B (en) * | 2018-09-27 | 2024-01-19 | 浙江图元智能装备科技有限公司 | Visual identification device for graphite hole positioning of self-lubricating bearing shaft sleeve workpiece |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109300161B (en) | Positioning method and device based on binocular vision | |
US9087258B2 (en) | Method for counting objects and apparatus using a plurality of sensors | |
US20190034714A1 (en) | System and method for detecting hand gestures in a 3d space | |
CN105631390B (en) | Method and system for spatial finger positioning | |
CN112352149A (en) | System and method for automatically analyzing air samples | |
CN104506857A (en) | Camera position deviation detection method and device | |
CN104021382A (en) | Eye image collection method and system | |
US11371851B2 (en) | Method and system for determining landmarks in an environment of a vehicle | |
CN109271844A (en) | Electrical cabinet electrical symbol recognition methods based on OpenCV | |
CN102818544A (en) | On-line measurement method for pitch circle center of automobile hub bolt hole and central eccentric distance of central hole | |
CN107610085A (en) | A kind of welding point defect detecting system based on computer vision | |
CN102622614B (en) | Knife switch closing reliability judging method based on distance between knife switch arm feature point and fixing end | |
CN109409395A (en) | Using the method for template matching method identification target object region electrical symbol in power monitoring | |
CN104156951A (en) | Leukocyte detecting method aiming at bronchoalveolar lavage smear | |
CN105022999A (en) | Man code company real-time acquisition system | |
CN105957300B (en) | A kind of wisdom gold eyeball identification is suspicious to put up masking alarm method and device | |
CN111639568A (en) | Navigation patrol method for intelligent robot of transformer substation | |
CN104700620A (en) | Traffic checkpoint-based method and device for recognizing fake-licensed vehicles | |
CN109460757A (en) | Seal location recognition method and device | |
CN114820475B (en) | Edge identification method and system, wafer processing device and method for determining concentric state of wafer and processing table | |
CN114155557B (en) | Positioning method, positioning device, robot and computer-readable storage medium | |
CN104715250A (en) | Cross laser detection method and device | |
JP2020109542A (en) | Deposition substance detection device and deposition substance detection method | |
CN101571953B (en) | Object detection method, system and stereoscopic vision system | |
CN105654086A (en) | Robot key member pose identification method and system based on monocular vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into 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: 20160608 |