CN108535263A - Glue Spreading Robot working state detection device - Google Patents
Glue Spreading Robot working state detection device Download PDFInfo
- Publication number
- CN108535263A CN108535263A CN201810243696.9A CN201810243696A CN108535263A CN 108535263 A CN108535263 A CN 108535263A CN 201810243696 A CN201810243696 A CN 201810243696A CN 108535263 A CN108535263 A CN 108535263A
- Authority
- CN
- China
- Prior art keywords
- detection data
- information
- acquisition
- matrix
- directional information
- 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- Withdrawn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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Analysis (AREA)
Abstract
Image is caused to be unfavorable for gluing Defect Detection in order to solve the problems, such as that image detecting apparatus is given a shock during processing of robots, the present invention provides a kind of Glue Spreading Robot working state detection devices, it is identified for the flaw to Glue Spreading Robot gluing finished product, including:Video acquisition subsystem, video acquisition is carried out to the position assembled in workpiece for being obtained by three video capture devices, Defect Detection data are obtained, and the position of the assembling is the position that workpiece assembled after different workpieces being assembled by robot is adjacent to each other or is connected;Processing subsystem, for carrying out damping and stabilisation to detection data;State-detection recognition subsystem, for carrying out assembling flaw identification by stabilized data.
Description
Technical field
The invention belongs to robot cell operation quality monitoring technology fields, work more particularly to a kind of Glue Spreading Robot
Condition checkout gear.
Background technology
Robot has been universally applied to all common robot of the occasions such as production line, including assembling, welding, gluing and has held
The figure of row operation.For example, the requirement that people are machined automobile etc. the performance and appearance of product is also higher and higher.Automobile
Painting effect is that automobile appearance gives people most direct impression.Vehicle coating process be four big technique of automobile making (punching press, be welded,
One of application, general assembly), quality directly affects first impression of the consumer for automobile brand.Due to painting dressing automobiles quality by
To the influence of many factors, such as:Coating itself, painting environment and each processing parameter setting etc. so that painting dressing automobiles become
One high-precision, highly difficult work, therefore there is still a need for carry out Defect Detection after body of a motor car spray painting drying.But due to automobile
The high light-reflecting property on surface so that Defect Detection is extremely difficult.
Currently, the application Defect Detection link in China processing of robots workshop is finished artificially, pass through stone of buying oil, light
The methods of according to, slight flaws are detected in conjunction with the modes such as observation and touch from different perspectives.In the production line, application is completed
After drying, generally flaw is detected by several workers, foundation is provided for follow-up link of repairing.This work not only needs to examine
Survey personnel have abundant working experience, and detection workman is required to remain the attention of high intensity, and assembly line is connected
The worker of continuous work, it is easy to generate visual fatigue, decline to easily lead to detection efficiency and Detection accuracy, inevitably
The phenomenon that will appear flase drop missing inspection.On the other hand, with the industrial transformation of the region of world economy adjustment and China's economic, manually
Cost is also higher and higher, and current high speed, accurate, automation production requirement can not be also adapted to using the method for artificial detection.Cause
How this improves the automatization level of Defect Detection, and it is also world's processing of robots that reduction production cost, which is China's automobile industry,
The pressing issues that industry faces.In addition, be technological means commonly used in the art by video detection flaw, but workshop is inevitably due to passing
Defeated band movement, machining operations etc. cause video acquisition to there are the vibrations to picture pick-up device in the process, influence to utilize its image
Carry out the quality of the coated face Defect Detection of gluing finished product.
Invention content
In order to improve robot assembling flaw monitor accuracy, solve during processing of robots image detecting apparatus by
The problem of causing image to be unfavorable for gluing Defect Detection to vibrations, the present invention provides a kind of detections of Glue Spreading Robot working condition
Device is identified for the flaw to Glue Spreading Robot gluing finished product, including:
Video acquisition subsystem carries out video for being obtained by three video capture devices to the position assembled in workpiece
Acquisition, obtain Defect Detection data, the position of the assembling be after different workpieces are assembled by robot be assembled workpiece that
This position that is adjacent or being connected;
Processing subsystem, for carrying out damping and stabilisation to detection data;
State-detection recognition subsystem, for carrying out assembling flaw identification by stabilized data.
Further, the shooting angle of the video capture device is different from each other.
Further, the focal length of the video capture device can be automatically adjusted, and respective adjusting range is each other
It is different.
Further, the frame data set that the detection data is made of multiple images.
Further, the video acquisition subsystem includes:
First video capture device, for acquiring the first detection data and recording its first acquisition directional information;
Second video capture device, for acquiring the second detection data and recording its second acquisition directional information;
Third video capture device acquires directional information for acquiring third detection data and recording its third;
First detection data collection obtaining unit, for first detection data, the second detection data, third testing number
According to, first acquisition directional information, second acquisition directional information and third acquisition directional information carry out first collect, obtain the first inspection
Measured data collection, first collection includes adopting first detection data, the second detection data, third detection data, first
Collection directional information, the second acquisition directional information and third acquisition directional information are preserved, and the first detection data collection is obtained.
Further, the processing subsystem includes:
Second detection data collection obtaining unit, for according to the first detection data, the second detection data and third testing number
Direction is acquired according in different moments corresponding first acquisition direction, the second acquisition direction, third, to the first detection data collection
In each frame data reconfigured, obtain the second detection data collection;
Transmission unit, for the second detection data collection to be transferred to supervisory control of robot server.
Further, the first acquisition directional information includes horizontal information and posture information, and the horizontal information indicates
Straight line where the focal length of video capture device is towards the deflection in the direction in field data source, described in the posture information indicates
The three-dimensional acceleration vector of video capture device.
Further, the second detection data collection obtaining unit includes:
First level information difference computing unit, in the first moment t1, calculate separately the first acquisition directional information
The horizontal information and horizontal information of the second acquisition directional information, the horizontal information of the first acquisition directional information and third acquire direction
Difference between horizontal information the two of information, the difference correspond respectively to first level information difference α1With the second horizontal letter
Cease difference α2;
Second horizontal information difference computational unit, in the first moment t1The second moment t later2, calculate separately first
Acquire the horizontal information of the horizontal information of directional information and the horizontal information of the second acquisition directional information, the first acquisition directional information
The difference between horizontal information the two of directional information is acquired with third, which corresponds respectively to third horizontal information difference
α3With the 4th horizontal information difference α4;
Ratio calculation unit, for calculating the posture information of the first acquisition directional information, the appearance of the second acquisition directional information
These three posture informations of posture information of state information and third acquisition directional information are in the first moment t1With the second moment t2Between
The first posture information change rate g in period1, the second posture information change rate g2With third posture information change rate g3, wherein
The first, second, and third posture information change rate is the vector sum and the second moment and the first moment by three-dimensional acceleration
Between time difference between ratio calculation obtain;
First pixel matching matrix calculation unit is as follows for calculating pixel matching matrix A:
First Transition matrix calculation unit, is used for:It is located at the first moment t1, the corresponding pixel set pair of the first detection data
The matrix answered is m, and the corresponding matrix of the corresponding pixel set of the second detection data is n, the corresponding set of pixels of third detection data
It is p to close corresponding matrix;In the second moment t2, the corresponding matrix of the corresponding pixel set of the first detection data is x, the second detection
The corresponding matrix of the corresponding pixel set of data is y, and the corresponding matrix of the corresponding pixel set of third detection data is z, is calculated
First Transition Matrix C1For:
Wherein mod (t2-t1, 2) and it indicates to t2-t1The absolute value of difference take the remainder of the quotient relative to 2;
Second pixel matching matrix calculation unit, forAs overturning central point
Space coordinate, matrix A is symmetrically overturn, matrix A is obtained ';
Second transition matrix computing unit, for calculating the second transition matrix C2For:
Interpolation unit, for utilizing Matrix C2To matrix A ' interpolation is carried out, obtain matrix A ", and calculate the second detection data
It concentrates and the second moment t2The corresponding matrix q of the corresponding pixel set of corresponding detection data frame:
Cumulative unit for preserving q, and then constantly accumulates to obtain the second detection data collection.
Further, the state-detection recognition subsystem includes:Prewarning unit, for working as the second detection data collection
Frame data in, when the gray scales of frame data corresponding with the corresponding space coordinate in precalculated position is more than default gray threshold, hair
Go out to assemble flaw warning information.
Technical scheme of the present invention has the advantages that:
By the picture pick-up devices such as the camera of multiple and different focal lengths, camera it is collected to the position that is assembled in workpiece into
Row video acquisition, obtains Defect Detection data, and the position of the assembling is assembled after being assembled different workpieces by robot
Workpiece is adjacent to each other or the stabilization processes of position that are connected, improves during identification assembling flaw due to where robot
Assembly line operates the shake of workpiece, shakes the smudgy problem of the assembling Defect Detection image generated, improves for machine
The workpiece that people's monitoring server is assembled together by gray scale area detecting be connected to each other or adjacent surface
Between existing flaw detection identification accuracy and reliability.
Description of the drawings
Fig. 1 shows the detection device composition frame chart of the present invention.
Specific implementation mode
According to a preferred embodiment of the invention, Glue Spreading Robot working state detection device as shown in Figure 1, for applying
The flaw of glue robot coating finished product is identified, including:
Video acquisition subsystem carries out video for being obtained by three video capture devices to the position assembled in workpiece
Acquisition, obtain Defect Detection data, the position of the assembling be after different workpieces are assembled by robot be assembled workpiece that
This position that is adjacent or being connected;
Processing subsystem, for carrying out damping and stabilisation to detection data;
State-detection recognition subsystem, for carrying out assembling flaw identification by stabilized data.
Preferably, the shooting angle of the video capture device is different from each other.
Preferably, the focal length of the video capture device can be automatically adjusted, and respective adjusting range is each other not
Together.
Preferably, the frame data set that the detection data is made of multiple images.
Preferably, the video acquisition subsystem includes:
First video capture device, for acquiring the first detection data and recording its first acquisition directional information;
Second video capture device, for acquiring the second detection data and recording its second acquisition directional information;
Third video capture device acquires directional information for acquiring third detection data and recording its third;
First detection data collection obtaining unit, for first detection data, the second detection data, third testing number
According to, first acquisition directional information, second acquisition directional information and third acquisition directional information carry out first collect, obtain the first inspection
Measured data collection, first collection includes adopting first detection data, the second detection data, third detection data, first
Collection directional information, the second acquisition directional information and third acquisition directional information are preserved, and the first detection data collection is obtained.
Preferably, the processing subsystem includes:
Second detection data collection obtaining unit, for according to the first detection data, the second detection data and third testing number
Direction is acquired according in different moments corresponding first acquisition direction, the second acquisition direction, third, to the first detection data collection
In each frame data reconfigured, obtain the second detection data collection;
Transmission unit, for the second detection data collection to be transferred to supervisory control of robot server.
Preferably, the first acquisition directional information includes horizontal information and posture information, and the horizontal information expression regards
Straight line where the focal length of frequency collecting device towards the direction in field data source deflection, the posture information indicate described in regard
The three-dimensional acceleration vector of frequency collecting device.
Preferably, the second detection data collection obtaining unit includes:
First level information difference computing unit, in the first moment t1, calculate separately the first acquisition directional information
The horizontal information and horizontal information of the second acquisition directional information, the horizontal information of the first acquisition directional information and third acquire direction
Difference between horizontal information the two of information, the difference correspond respectively to first level information difference α1With the second horizontal letter
Cease difference α2;
Second horizontal information difference computational unit, in the first moment t1The second moment t later2, calculate separately first
Acquire the horizontal information of the horizontal information of directional information and the horizontal information of the second acquisition directional information, the first acquisition directional information
The difference between horizontal information the two of directional information is acquired with third, which corresponds respectively to third horizontal information difference
α3With the 4th horizontal information difference α4;
Ratio calculation unit, for calculating the posture information of the first acquisition directional information, the appearance of the second acquisition directional information
These three posture informations of posture information of state information and third acquisition directional information are in the first moment t1With the second moment t2Between
The first posture information change rate g in period1, the second posture information change rate g2With third posture information change rate g3, wherein
The first, second, and third posture information change rate is the vector sum and the second moment and the first moment by three-dimensional acceleration
Between time difference between ratio calculation obtain;
First pixel matching matrix calculation unit is as follows for calculating pixel matching matrix A:
First Transition matrix calculation unit, is used for:It is located at the first moment t1, the corresponding pixel set pair of the first detection data
The matrix answered is m, and the corresponding matrix of the corresponding pixel set of the second detection data is n, the corresponding set of pixels of third detection data
It is p to close corresponding matrix;In the second moment t2, the corresponding matrix of the corresponding pixel set of the first detection data is x, the second detection
The corresponding matrix of the corresponding pixel set of data is y, and the corresponding matrix of the corresponding pixel set of third detection data is z, is calculated
First Transition Matrix C1For:
Wherein mod (t2-t1, 2) and it indicates to t2-t1The absolute value of difference take the remainder of the quotient relative to 2;
Second pixel matching matrix calculation unit, forAs overturning central point
Space coordinate, matrix A is symmetrically overturn, matrix A is obtained ';
Second transition matrix computing unit, for calculating the second transition matrix C2For:
Interpolation unit, for utilizing Matrix C2To matrix A ' interpolation is carried out, obtain matrix A ", and calculate the second detection data
It concentrates and the second moment t2The corresponding matrix q of the corresponding pixel set of corresponding detection data frame:
Cumulative unit for preserving q, and then constantly accumulates to obtain the second detection data collection.
Preferably, the state-detection recognition subsystem includes:Prewarning unit, for when the second detection data collection
In frame data, when the gray scale of frame data corresponding with the corresponding space coordinate in precalculated position is more than default gray threshold, send out
Assemble flaw warning information.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement etc., should all be included in the protection scope of the present invention made by within refreshing and principle.
Claims (9)
1. a kind of Glue Spreading Robot working state detection device is identified for the flaw to Glue Spreading Robot gluing finished product,
It is characterised in that it includes:
Video acquisition subsystem adopts the position progress video assembled in workpiece for being obtained by three video capture devices
Collection, obtain Defect Detection data, the position of the assembling be after different workpieces are assembled by robot be assembled workpiece each other
Position that is adjacent or being connected;
Processing subsystem, for carrying out damping and stabilisation to detection data;
State-detection recognition subsystem, for carrying out assembling flaw identification by stabilized data.
2. Glue Spreading Robot working state detection device according to claim 1, which is characterized in that the video acquisition is set
Standby shooting angle is different from each other.
3. Glue Spreading Robot working state detection device according to claim 1, which is characterized in that the video acquisition is set
Standby focal length can be automatically adjusted, and respective adjusting range is different from each other.
4. Glue Spreading Robot working state detection device according to claim 3, which is characterized in that the detection data is
The frame data set being made of multiple images.
5. Glue Spreading Robot working state detection device according to claim 4, which is characterized in that video acquisition
System includes:
First video capture device, for acquiring the first detection data and recording its first acquisition directional information;
Second video capture device, for acquiring the second detection data and recording its second acquisition directional information;
Third video capture device acquires directional information for acquiring third detection data and recording its third;
First detection data collection obtaining unit, for first detection data, the second detection data, third detection data,
First acquisition directional information, the second acquisition directional information and third acquisition directional information carry out first and collect, and obtain the first detection
Data set, first collection include by first detection data, the second detection data, third detection data, the first acquisition
Directional information, the second acquisition directional information and third acquisition directional information are preserved, and the first detection data collection is obtained.
6. Glue Spreading Robot working state detection device according to claim 5, which is characterized in that the processing subsystem
Including:
Second detection data collection obtaining unit, for being existed according to the first detection data, the second detection data and third detection data
Different moments corresponding first acquisition direction, the second acquisition direction, third acquire direction, are concentrated to the first detection data
Each frame data are reconfigured, and the second detection data collection is obtained;
Transmission unit, for the second detection data collection to be transferred to supervisory control of robot server.
7. Glue Spreading Robot working state detection device according to claim 6, which is characterized in that the first acquisition side
Include horizontal information and posture information to information, straight line where the horizontal information indicates the focal length of video capture device is towards now
The deflection in the direction in field data source, the posture information indicate the three-dimensional acceleration vector of the video capture device.
8. Glue Spreading Robot working state detection device according to claim 7, which is characterized in that second testing number
Include according to collection obtaining unit:
First level information difference computing unit, in the first moment t1, calculate separately the horizontal letter of the first acquisition directional information
Breath and the horizontal information of the second acquisition directional information, horizontal information and the third acquisition directional information of the first acquisition directional information
Difference between horizontal information the two, the difference correspond respectively to first level information difference α1With the second horizontal information difference
α2;
Second horizontal information difference computational unit, in the first moment t1The second moment t later2, calculate separately the first acquisition
The horizontal information and the of the horizontal information of directional information and the horizontal information of the second acquisition directional information, the first acquisition directional information
Difference between horizontal information the two of three acquisition directional informations, the difference correspond respectively to third horizontal information difference α3With
4th horizontal information difference α4;
Ratio calculation unit, for calculating the posture information of the first acquisition directional information, the posture letter of the second acquisition directional information
Breath and these three posture informations of posture information of third acquisition directional information are in the first moment t1With the second moment t2Between time
The first posture information change rate g in section1, the second posture information change rate g2With third posture information change rate g3, wherein described
First, second, and third posture information change rate is by the vector sum of three-dimensional acceleration and between the second moment and the first moment
Time difference between ratio calculation obtain;
First pixel matching matrix calculation unit is as follows for calculating pixel matching matrix A:
First Transition matrix calculation unit, is used for:It is located at the first moment t1, the corresponding pixel set of the first detection data is corresponding
Matrix is m, and the corresponding matrix of the corresponding pixel set of the second detection data is n, the corresponding pixel set pair of third detection data
The matrix answered is p;In the second moment t2, the corresponding matrix of the corresponding pixel set of the first detection data is x, the second detection data
The corresponding matrix of corresponding pixel set is y, and the corresponding matrix of the corresponding pixel set of third detection data is z, calculates first
Transition matrix C1For:
Wherein mod (t2-t1, 2) and it indicates to t2-t1The absolute value of difference take the remainder of the quotient relative to 2;
Second pixel matching matrix calculation unit, forSky as overturning central point
Between coordinate, matrix A is symmetrically overturn, matrix A is obtained ';
Second transition matrix computing unit, for calculating the second transition matrix C2For:
Interpolation unit, for utilizing Matrix C2To matrix A ' interpolation is carried out, obtain matrix A ", and calculate the second detection data and concentrate
With the second moment t2The corresponding matrix q of the corresponding pixel set of corresponding detection data frame:
Cumulative unit for preserving q, and then constantly accumulates to obtain the second detection data collection.
9. Glue Spreading Robot working state detection device according to claim 8, which is characterized in that the state-detection is known
Small pin for the case system includes:Prewarning unit is used in the frame data of the second detection data collection, sky corresponding with precalculated position
Between the gray scale of the corresponding frame data of coordinate when being more than default gray threshold, send out assembling flaw warning information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810243696.9A CN108535263A (en) | 2018-03-23 | 2018-03-23 | Glue Spreading Robot working state detection device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810243696.9A CN108535263A (en) | 2018-03-23 | 2018-03-23 | Glue Spreading Robot working state detection device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108535263A true CN108535263A (en) | 2018-09-14 |
Family
ID=63483657
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810243696.9A Withdrawn CN108535263A (en) | 2018-03-23 | 2018-03-23 | Glue Spreading Robot working state detection device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108535263A (en) |
-
2018
- 2018-03-23 CN CN201810243696.9A patent/CN108535263A/en not_active Withdrawn
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106239476B (en) | Robot and the robot controller for possessing the display function of power | |
CN111013883A (en) | Robot control method for intelligent spraying of multiple vehicle types | |
DE102018009815B4 (en) | Imaging device comprising a visual sensor for capturing the image of a workpiece | |
CN106271081B (en) | Three coordinate rectangular robot line laser seam tracking systems and its tracking | |
CN112010024B (en) | Automatic container grabbing method and system based on laser and vision fusion detection | |
CN103776378A (en) | Non-contact type flexible on-line dimension measurement system | |
CN114011608B (en) | Spraying process optimization system based on digital twinning and spraying optimization method thereof | |
Wang et al. | A real-time weld line detection for derusting wall-climbing robot using dual cameras | |
CN108572625A (en) | The control system of lathe | |
US20100246899A1 (en) | Method and Apparatus for Dynamic Estimation of Feature Depth Using Calibrated Moving Camera | |
CN105563481A (en) | Robot vision guide method used for shaft hole assembling | |
Xia et al. | Workpieces sorting system based on industrial robot of machine vision | |
CN108435456A (en) | Industrial robot spraying control system | |
CN108453441A (en) | industrial welding robot welding quality identification method | |
CN108535263A (en) | Glue Spreading Robot working state detection device | |
CN108067725A (en) | A kind of new robotic laser vision weld joint detecting system and method | |
CN108422122A (en) | Assembly line work condition inspection system based on image recognition | |
CN108436333A (en) | Welding robot based on machine vision | |
CN108489999A (en) | The assembling flaw detection method of workpiece | |
Yesin et al. | Robust CAD model based visual tracking for 3D microassembly using image space potentials | |
JP3340143B2 (en) | Polishing tool control device | |
DE112021004660T5 (en) | robotic system | |
CN114926531A (en) | Binocular vision based method and system for autonomously positioning welding line of workpiece under large visual field | |
Zou et al. | Features extraction of butt joint for tailored blank laser welding based on three-line stripe laser vision sensor | |
CN206912816U (en) | Identify the device of mechanical workpieces pose |
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: 20180914 |
|
WW01 | Invention patent application withdrawn after publication |