CN115471547A - One-key calibration algorithm for visual detection of six-axis manipulator - Google Patents
One-key calibration algorithm for visual detection of six-axis manipulator Download PDFInfo
- Publication number
- CN115471547A CN115471547A CN202210775640.4A CN202210775640A CN115471547A CN 115471547 A CN115471547 A CN 115471547A CN 202210775640 A CN202210775640 A CN 202210775640A CN 115471547 A CN115471547 A CN 115471547A
- Authority
- CN
- China
- Prior art keywords
- workpiece
- visual
- processed
- coordinate system
- axis manipulator
- 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.)
- Withdrawn
Links
- 230000000007 visual effect Effects 0.000 title claims abstract description 66
- 238000001514 detection method Methods 0.000 title abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 38
- 239000003550 marker Substances 0.000 claims abstract description 25
- 238000003754 machining Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims abstract description 4
- 238000006243 chemical reaction Methods 0.000 claims abstract description 3
- 238000000605 extraction Methods 0.000 claims description 12
- 238000012937 correction Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 6
- 238000000926 separation method Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims 7
- 238000011179 visual inspection Methods 0.000 claims 3
- 238000000034 method Methods 0.000 description 12
- 239000000463 material Substances 0.000 description 5
- 230000009466 transformation Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Manipulator (AREA)
Abstract
The invention relates to the technical field of automation, in particular to a one-key calibration algorithm for visual detection of a six-axis manipulator. The specific algorithm flow of the invention is that S1, a workpiece to be processed is placed on a designated position of a processing table; s2, detecting the relative position and posture coordinates of a visual marker in the workpiece to be processed through a visual sensor; and S3, calibrating the machining point in the workpiece to be machined relative to a tool coordinate system under a machining table coordinate system through coordinate conversion. The one-key calibration algorithm for vision detection of the six-axis manipulator specifically comprises the steps of establishing a relative coordinate system by identifying a plurality of visual markers on a workpiece to be processed, and performing deviation matrix-based compensation on tool coordinates on a processing table by using relative coordinates in the visual markers, so that the size of the workpiece to be processed is not limited, and the processing quality and the automation degree of the workpiece are improved.
Description
Technical Field
The invention relates to the technical field of automation, wherein the IPC classification number is B25J13/08, and in particular relates to a one-key scaling algorithm for visual detection of a six-axis manipulator.
Background
Six-axis manipulator mainly used industrial production, the automated processing and the transportation operation of product, in the in-process that uses six-axis manipulator to handle the product material of reality, at first need carry out the location on the table surface of processing of product material, visual sensor in six-axis manipulator detects behind the product material position, just can carry out subsequent processing, but precision is not high in the tool coordinate system calibration of material work piece on the table surface at present stage, simultaneously because the influence of the unloading error of material work piece and the environmental factor in the actual course of processing, can cause the inaccurate problem of tool coordinate system location usually.
Patent CN201410311052 provides an auxiliary device for adjusting position of a robot arm, and this patent is through setting up position detector on the processing mesa of wafer for whether the position of placing of real-time supervision wafer is accurate, if the deviation that the wafer was placed is great, will automatic inform the engineer to adjust, but in this patent the position detector has restricted the size type of the wafer of waiting to detect, when the size of the wafer of waiting to detect changes, then can't carry out the detection of position data step by step, and the range of application is limited. Patent CN201510168995 provides a mechanical hand-eye calibration method based on active binocular vision, in this patent, a structured light generator is established at the end of a mechanical hand to perform feedback detection of the operation pose data at the end of the mechanical hand, so that it is not necessary to add a mark point at the end of the mechanical hand, and a control structure is simplified while realizing the detection of the motion pose of the mechanical hand.
Aiming at the problems existing in the prior six-axis manipulator workpiece position calibration, a manual secondary adjustment mode or a manual adjustment mode of a demonstrator is usually adopted in the factory processing process, but the manual point alignment calibration depends on the proficiency and the fidelity of technical workers, and the subjectivity is large, so that the precision of point alignment cannot be ensured due to manual naked eye errors in the calibration process, the teaching mode process of the demonstrator is complicated, the teaching needs to be adjusted repeatedly, and a set of automatic correction program is needed to correct the position points of a plurality of tool coordinate systems at the same time, so the invention provides a one-key calibration algorithm for the six-axis manipulator visual detection.
Disclosure of Invention
Aiming at the existing problems, the invention provides a one-key calibration algorithm for vision detection of a six-axis manipulator, which comprises the specific algorithm processes of placing a workpiece to be processed on a specified position of a processing table top, detecting the relative position and posture coordinates of a visual marker in the workpiece to be processed through a vision sensor, and calibrating a processing point in the workpiece to be processed relative to a tool coordinate system under a processing table top coordinate system through coordinate conversion.
Preferably, in the visual sensor, a parameterized identification file about the visual markers is established, and the parameterized identification file comprises edge features, pixel features and identification number of the visual markers.
Specifically, the edge features of the visual marker are recognized in a form recognition mode, firstly, line features recognized on the surface of a workpiece to be processed are extracted, and then training and registration are carried out according to the shapes and orientation values of the line features.
Preferably, the parameterized identification file is subjected to feature registration with the visual marker features actually identified by the visual sensor, so as to extract the visual marker features.
Preferably, the feature registration is performed by adopting a mode of screening and training based on graphic features and pixel feature assisted registration to identify the visual markers.
Preferably, the parameterized identification file is used for circularly reading the workpiece to be processed by setting the identification times, and dynamically repositioning the pose of the visual marker in the processing process of the workpiece.
Preferably, the characteristic registration comprises the steps of firstly acquiring picture data of a workpiece to be processed through a visual sensor, and carrying out image processing on the picture data; wherein the image processing comprises the separation of foreground characteristics and background characteristics of the surface of the workpiece to be processed; and on the basis of the separation, performing edge feature extraction optimization on the workpiece to be processed.
Specifically, aiming at objects with large differences such as images acquired by the visual markers, background colors and the like, image data acquisition and visual marker identification can be directly carried out through a CCD camera, before separation of foreground features and background features on the surface of a workpiece to be processed is carried out, filtering and noise reduction of the image data are carried out by using a pcl-based point cloud library, exposure and gain values of the image data are adjusted on the basis, and therefore the influence of external ambient light transformation on a visual image processing algorithm is better eliminated.
Preferably, the feature registration is to register the to-be-processed workpiece with the edge features in the parameterized identification file after the edge feature extraction optimization, so as to perform classification extraction on the visual markers, and perform boundary contour line correction on the visual markers on the basis of the classification extraction.
Preferably, the centering calculation is performed after the boundary contour line is corrected.
Specifically, the boundary contour line correction is to perform binarization processing on picture data with boundary contour lines so as to acquire a data set based on the boundary contour line data, establish a data set optimization smoothing method based on Gaussian distribution in the data set, call an operator library based on opencv, and establish a central moment based on mean value symmetry measurement for the data set so as to calculate the central position in the edge contour data.
Preferably, the relative pose coordinate system of the visual marker is determined through the centralized calculation, the coordinate transformation is carried out on the relative coordinate systems and the processing table coordinate system, and a deviation matrix based on the relative pose coordinate system and the processing table coordinate system is established for carrying out position compensation on the actual tool coordinate system.
Preferably, the visual markers are in a plurality of cooperative calibration modes.
Compared with the prior art, the invention has the beneficial effects that:
(1) The one-key calibration algorithm for vision detection of the six-axis manipulator disclosed by the invention has the advantages that a relative coordinate system is established by identifying a plurality of vision markers on the workpiece to be machined, and compensation based on a deviation matrix is carried out on the tool coordinate on the machining table top by using the relative coordinates in the vision markers, so that the size of the workpiece to be machined is not limited, manual adjustment and teaching of a demonstrator are not needed, the tool coordinate system of the workpiece to be machined corresponding to the manipulator can be quickly and automatically calibrated at high precision, and the machining quality and the automation degree of the workpiece are improved.
(2) On the basis of (1), in the visual detection mode, the position of the workpiece processing point is corrected in real time in the process of processing the workpiece by circularly reading the visual markers of the workpiece to be processed in the processing process, the dynamic position and pose are relocated, and the relative position coordinates of the workpiece processing point relative to the plurality of visual markers and the preset processing coordinates are calculated to calculate the deviation, so that the real-time adjustment of the position and pose of the processing point is realized, and the processing and positioning precision of the workpiece is improved.
(3) On the basis of (1), in the processing process of the workpiece to be processed, due to the influence of various factors such as blanking deformation of the workpiece to be processed, processing temperature or processing pressure and the like, the workpiece to be processed may be deformed, and if the feature extraction of the visual marker is directly carried out on the deformed workpiece, the distortion of a feature image is caused, so that the boundary contour line correction is added to the part, the centralized calculation is carried out on the basis, and the accuracy of the calculation of the object position posture coordinate of the visual marker is improved.
(4) On the basis of the step (3), the image data acquired by the vision sensor is separated based on the foreground characteristic and the background characteristic, and the extraction and optimization of the edge characteristic of the foreground characteristic are carried out after the separation, so that the condition that the edge of the fusion part of the vision marker and the scene is fuzzy and noisy is avoided, the quality of the extraction of the edge characteristic of the vision marker is improved, and the pose positioning precision of the vision marker is further improved.
(5) On the basis of (1), the invention establishes a collaborative pose calibration mode based on a plurality of visual markers, and compared with the problem of inaccurate positioning caused by single visual marker positioning, the invention firstly fuses the relative coordinate systems recognized by the visual markers, and fuses and corrects the deviation of the fused relative coordinate system and the tool coordinate system, thereby improving the positioning accuracy of the visual markers.
Drawings
FIG. 1 is a flow chart of a neighborhood connected heterogeneous design method based on an adaptive chip.
Detailed Description
Example 1:
in the embodiment, as shown in fig. 1, a specific algorithm flow is that a workpiece to be processed is placed at an appointed position on a processing table, a relative pose coordinate of a visual marker in the workpiece to be processed is detected by a visual sensor, and a processing point in the workpiece to be processed is calibrated relative to a tool coordinate system under a coordinate system of the processing table through coordinate transformation.
The machining table is internally provided with a mechanical positioning device serving as a primary positioning position area of a machined workpiece, then a visual marker on the machined workpiece is identified through a visual image processing algorithm in a visual sensor, and the position of the workpiece to be machined is further calibrated on the basis of the primary positioning area through the visual marker.
Specifically, the position calibration includes the following main steps:
a1, initializing a visual sensor and setting internal parameters;
a2, establishing a visual marker calibration model in the visual sensor on the basis of the S1, and updating parameters of the visual sensor by using the visual marker model after outputting a visual marker model handle;
a3, placing a workpiece to be processed in a station, and identifying an object to be identified in the workpiece to be processed through a parameterized identification file;
a4, acquiring initialization parameters of a processing bedplate through a vision sensor, and acquiring image data of a visual marker in a workpiece to be processed;
a5, on the basis of S4, carrying out filtering processing on the image data and separating surface foreground characteristics from background characteristics;
a6, extracting a boundary contour line of the visual marker on the basis of the S6, and performing boundary contour line correction on the boundary contour line, wherein the boundary contour line correction is performed, on the basis of the boundary contour line correction, nearest neighbor edge fitting of an edge contour line is performed, the central position of a convenient contour line to be fitted is calculated, and the central position is taken as a relative position coordinate system;
and A7, converting the relative coordinate systems into a tool coordinate system, and registering the tool coordinate system and a visual coordinate system in the visual sensor to realize visual identification.
Claims (10)
1. A one-key calibration algorithm for vision inspection of six-axis manipulators is characterized in that the specific algorithm flow is S1, a workpiece to be machined is placed at an appointed position of a machining table top; s2, detecting the relative pose coordinates of the visual markers in the workpiece to be processed through a visual sensor; and S3, calibrating the machining point in the workpiece to be machined relative to a tool coordinate system under a machining table coordinate system through coordinate conversion.
2. The key scaling algorithm for six-axis robot vision inspection as claimed in claim 1, wherein the vision sensor is configured to create a parametric identification file for the visual markers, the parametric identification file comprising edge features, pixel features and identification number of the visual markers.
3. The key scaling algorithm for six-axis manipulator visual inspection according to claim 2, wherein the parameterized identification file is used for visual marker feature extraction by feature registration with the visual marker features actually identified by the visual sensor.
4. The one-key calibration algorithm for six-axis manipulator visual inspection according to claim 3, wherein the feature registration is performed by means of feature screening training based on graphics and pixel feature assisted registration to identify visual markers.
5. The key scaling algorithm for vision inspection of six-axis manipulator according to claim 3, wherein the parameterized identification file is used for reading the workpiece to be processed cyclically by setting the identification times, and dynamically repositioning the visual markers during the processing of the workpiece.
6. The one-key scaling algorithm for vision inspection of six-axis robot as claimed in claim 3, wherein said feature registration comprises firstly obtaining picture data of a workpiece to be processed by a vision sensor, and performing image processing on the picture data; wherein the image processing comprises the separation of foreground characteristics and background characteristics of the surface of the workpiece to be processed; and on the basis of the separation, performing edge feature extraction optimization on the workpiece to be processed.
7. The one-key scaling algorithm for vision inspection of six-axis manipulator as claimed in claim 6, wherein the feature registration is to register the edge feature of the workpiece to be processed after the optimization of edge feature extraction with the edge feature in the parameterized identification file, so as to perform classified extraction on the visual markers, and perform boundary contour correction on the visual markers based on the classified extraction.
8. A key scaling algorithm for six-axis manipulator visual inspection according to claim 7, wherein centering is performed after the boundary contour is rectified.
9. The key scaling algorithm for vision inspection of six-axis manipulator as claimed in claim 8, wherein the relative pose coordinate system of the vision marker is determined by said centering calculation, the relative coordinate systems are coordinate transformed with the table top coordinate system, and a deviation matrix based on the relative pose coordinate system and the table top coordinate system is established for position compensation of the actual tool coordinate system.
10. The key scaling algorithm for six-axis robot vision inspection according to claim 1, wherein the visual markers are a plurality of coordinated scaling modes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210775640.4A CN115471547A (en) | 2022-07-01 | 2022-07-01 | One-key calibration algorithm for visual detection of six-axis manipulator |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210775640.4A CN115471547A (en) | 2022-07-01 | 2022-07-01 | One-key calibration algorithm for visual detection of six-axis manipulator |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115471547A true CN115471547A (en) | 2022-12-13 |
Family
ID=84366557
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210775640.4A Withdrawn CN115471547A (en) | 2022-07-01 | 2022-07-01 | One-key calibration algorithm for visual detection of six-axis manipulator |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115471547A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117324995A (en) * | 2023-12-01 | 2024-01-02 | 宁波肆典零科技有限公司 | Workpiece feeding and centering method |
-
2022
- 2022-07-01 CN CN202210775640.4A patent/CN115471547A/en not_active Withdrawn
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117324995A (en) * | 2023-12-01 | 2024-01-02 | 宁波肆典零科技有限公司 | Workpiece feeding and centering method |
CN117324995B (en) * | 2023-12-01 | 2024-03-15 | 宁波肆典零科技有限公司 | Workpiece feeding and centering method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106780623B (en) | Rapid calibration method for robot vision system | |
EP3105016B1 (en) | Automatic calibration method for robot systems using a vision sensor | |
EP2045772B1 (en) | Apparatus for picking up objects | |
US11972589B2 (en) | Image processing device, work robot, substrate inspection device, and specimen inspection device | |
CN110580725A (en) | Box sorting method and system based on RGB-D camera | |
US20190184582A1 (en) | Imaging device including vision sensor capturing image of workpiece | |
CN108907526A (en) | A kind of weld image characteristic recognition method with high robust | |
CN110293559B (en) | Installation method for automatically identifying, positioning and aligning | |
CN106809730A (en) | The container automatic butt tackling system and hoisting method of a kind of view-based access control model | |
CN112497219B (en) | Columnar workpiece classifying and positioning method based on target detection and machine vision | |
CN113878576B (en) | Robot vision sorting process programming method | |
CN112729112B (en) | Engine cylinder bore diameter and hole site detection method based on robot vision | |
JP2019057250A (en) | Work-piece information processing system and work-piece recognition method | |
CN115471547A (en) | One-key calibration algorithm for visual detection of six-axis manipulator | |
CN113822810A (en) | Method for positioning workpiece in three-dimensional space based on machine vision | |
CN115131268A (en) | Automatic welding system based on image feature extraction and three-dimensional model matching | |
CN113500593A (en) | Method for grabbing designated part of shaft workpiece for loading | |
CN112109072A (en) | Method for measuring and grabbing accurate 6D pose of large sparse feature tray | |
CN117885096A (en) | Method and device for controlling welding operation of robot tail end welding gun | |
CN114918723B (en) | Workpiece positioning control system and method based on surface detection | |
CN116912316A (en) | Visual positioning method for dispensing machine | |
CN111563935B (en) | Visual positioning method for honeycomb holes of honeycomb sectional material | |
CN114882108A (en) | Method for estimating grabbing pose of automobile engine cover under two-dimensional image | |
CN107020545A (en) | The apparatus and method for recognizing mechanical workpieces pose | |
CN118314138B (en) | Laser processing method and system based on machine vision |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20221213 |
|
WW01 | Invention patent application withdrawn after publication |