CN106636510A - Automatic brazing rod changing method based on machine vision - Google Patents
Automatic brazing rod changing method based on machine vision Download PDFInfo
- Publication number
- CN106636510A CN106636510A CN201610914755.1A CN201610914755A CN106636510A CN 106636510 A CN106636510 A CN 106636510A CN 201610914755 A CN201610914755 A CN 201610914755A CN 106636510 A CN106636510 A CN 106636510A
- Authority
- CN
- China
- Prior art keywords
- brazing rod
- image
- drill steel
- target
- pricker
- 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.)
- Granted
Links
Classifications
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B7/00—Blast furnaces
- C21B7/12—Opening or sealing the tap holes
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Materials Engineering (AREA)
- Metallurgy (AREA)
- Organic Chemistry (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an automatic brazing rod changing method based on machine vision. The automatic brazing rod changing method comprises the following steps of: mounting an industrial camera and an embedded type processor, wherein an image acquisition area of the industrial camera covers an area of a mechanical hand and a brazing rod; processing the brazing rod according to an image acquired by the industrial camera, and separating the position of the brazing rod from background and detecting the position of the brazing rod by utilizing steps such as Gaussian filter, sharpening, edge detecting and linear detecting; judging whether the brazing rod is in place or not; detecting the position of a mechanical gripping arm by utilizing a target-detecting and target-tracking algorithm, tracking a motion state of the mechanical hand in real time, judging a relative position relationship of the mechanical hand and the brazing rod, and judging whether the mechanical hand is in place or not according to shielding degree of the mechanical hand and the brazing rod; and transmitting detected and analyzed results to a field controller, and controlling each part of an automatic brazing rod changing device. The automatic brazing rod changing method based on machine vision disclosed by the invention meets generation requirements of automatic equipment, and improves mechanical and automatic level of casthouse equipment.
Description
Technical field
The present invention be it is a kind of pricker method is moved based on machine vision certainly, belong to technical field of metallurgical equipment machine vision skill
Art field.
Background technology
Iron notch drill is a kind of common equipment of iron-smelter, and its function mainly uses drill rod and opens blast furnace taphole.With
Development and the requirement of industry 4.0 that blast furnace maximizes, the requirement of mechanization, Automated water product to casthouse equipment is increasingly
It is high.And in current blast furnace steel-making factory, still use mostly and manually change/unload drill rod, minority large steel-making plant adopt from
Move pricker machine there are problems that during use below 3：(1) due to the factor such as the length and distribution of weight of drill steel is unbalanced
Drill steel is caused to occur crooked, so as to cause mechanical gripping arm normally to grip drill steel.(2) affected by live adverse circumstances, machine
Tool gripping arm may run not in place there is the phenomenon that uses leisure moments.(3) move certainly pricker machine be carried out fill pricker operation still unload pricker operation, at present
Still artificial judgment is used, fails to realize comprehensively automation, intellectuality.
The content of the invention
In order to improve the labour intensity and working environment of stokehold workman, further improve from move pricker machine reliability,
Intelligent, present invention offer is a kind of to move pricker method certainly based on machine vision, and the generation for not only meeting automation equipment will
Ask, also improve mechanization, the automatization level of casthouse equipment.
The technical scheme that the present invention takes is：
Step 1：Mounting industrial camera, flush bonding processor, the image acquisition region of industry camera cover manipulator,
The region of drill steel.
Step 2：It is processed according to the image that industry camera is collected, using gaussian filtering, sharpening, edge inspection
The steps such as survey, straight-line detection, separate, detect the position of drill steel from background.
Step 3：Whether judge drill steel " in place ".
Step 4：Using target detection and target tracking algorism, the position of mechanical gripping arm, and real-time tracking machinery are detected
The motion state of hand, judges the relative position relation of manipulator and drill steel, according to both coverage extents, whether judges manipulator
" in place ".
Step 5：The result of detection, analysis can be transferred to field controller, be carried out to moving pricker device all parts certainly
Control.
The present invention is a kind of to move pricker method certainly based on machine vision, and technique effect is as follows：
1：Whether drill steel has been provided with by Machine Vision Detection iron notch drill, for mechanical arm controller provide " dress pricker " and
" unloading pricker " operation provides foundation.
2：Whether the drill steel in detection " pricker frame " is in " in place " state；Signal is otherwise given so as to control system driving
" positioning " component working on " pricker frame " is to drill steel " in place ".
3：Mechanical gripping arm and the drill steel on " pricker frame " and the relative position between " in place " drill steel are detected, is mechanical gripping arm
Controller provides the action such as " start and stop " and " crawl " and provides foundation.
4：The present invention not only meets the generation requirement of automation equipment, also improves the mechanization, automatically of casthouse equipment
Change level.
Description of the drawings
Fig. 1 is the hardware configuration schematic diagram of the present invention.
Fig. 2 is the industry camera installation site and its visual field schematic diagram of the present invention.
Fig. 3 is the industry camera shooting image schematic diagram of the present invention.
Fig. 4 is manipulator " in place " schematic diagram of the present invention.
Fig. 5 be the present invention move pricker machine working condition transition graph certainly.
Fig. 6 is the embedded software system structure chart of the present invention.
Specific embodiment
It is a kind of that pricker method is moved based on machine vision certainly, comprise the following steps：
Step 1：Mounting industrial camera, flush bonding processor, the image acquisition region of industry camera cover manipulator,
The region of drill steel.
Step 2：It is processed according to the image that industry camera is collected, using gaussian filtering, sharpening, edge inspection
The steps such as survey, straight-line detection, separate, detect the position of drill steel 8 from background.
Gaussian filtering：It is a kind of linear smoothing filtering, it is adaptable to eliminate Gaussian noise, is widely used in subtracting for image procossing
Make an uproar process.Popular says, gaussian filtering is exactly that average process is weighted to entire image, the value of each pixel, all
Obtain after being weighted averagely by other pixel values in itself and neighborhood.The concrete operations of gaussian filtering are：With a mould
Each pixel in plate (or claiming convolution, mask) scan image, with the weighted average gray scale of pixel in the neighborhood of template determination
Value goes the value of alternate template central pixel point.
Sharpen：Using the instrument of sharpening can quick focus blur edge, in raising image the definition at a certain position or
Focal length degree, makes the color of image specific region distincter.
Rim detection：It is the basic problem in image procossing and computer vision, its objective is bright in reference numbers image
The obvious point of degree change.Significant changes in image attributes generally reflect the critical event and change of attribute, and these include (i)
Discontinuous, (ii) surface direction in depth is discontinuous, the change of (iii) material property and the change of (iv) scene lighting.Examine at edge
Survey is in image procossing and computer vision, especially a research field in feature extraction.
Straight-line detection：Straight line is detected in original image, here using Hough straight-line detections, its thought is, in original
A point under beginning image coordinate system has corresponded to the straight line in parameter coordinate system, the straight line pair of same parameter coordinate system
A point under original coordinate system is answered, then, the institute of straight line has been presented under original coordinate system a little, their slope and intercept has been
Identical, so they correspond to same point under parameter coordinate system.So by each spot projection under original coordinate system
After under parameter coordinate system, see under parameter coordinate system either with or without accumulation point, such accumulation point has just corresponded to original coordinate system
Under straight line.
Step 3：Whether judge drill steel " in place ".
Step 4：Using target detection and target tracking algorism, the position of mechanical gripping arm, and real-time tracking machinery are detected
The motion state of hand 4, in shooting area, judges the relative position of manipulator and drill steel to be taken, if manipulator is complete
Drill steel to be taken is blocked, then can be considered manipulator " in place ".
Target detection, is also Objective extraction, refers to and carries the foreground target of motion from background image from sequence image
Take out.Here target detection adopts frame difference method.Frame difference method mainly uses the change in video sequence between two continuous frames to examine
The moving target surveyed under static scene, it is assumed that fk(x, y) and fk+1(x, y) is respectively the kth frame in image sequence and the frame of kth+1
The pixel value of middle pixel (x, y), then the error image of this two field pictures is as shown in following equation：
Diffk+1(x, y)=| fk(x,y)-fk+1(x,y)| (1-1)
Difference does not represent the region (background pixel passed through by the motion of moving target by 0 image-region in above formula
Value is constant), but because adjacent video interframe time interval very little, target location changes also very little.So the motion of moving target
The region passed through also just represents the region that moving target is located in present frame.Target just can be extracted using this principle.Fig. 7
Give the basic procedure of frame difference method：(1) the error image Diff of kth frame and the frame of kth+1 is obtained using 1-1 formulask+1；(2) to institute
The error image Diff for obtainingk+1Binaryzation obtains Qk+1；
(3) be the interference that eliminates small noise, so as to get moving target it is more accurate, to Qk+1Carry out it is necessary filtering and
Denoising, post processing result is Mk+1。
Target following is to determine the process of position of the same object in the different frame of image sequence, and its tracking mode can divide
It is the tracking based on segmentation and the tracking based on model.Tracking based on segmentation is will to be partitioned into all of fortune in each frame
Moving-target, is then matched with tracked target, so as to reach the purpose of tracking.Tracking based on model only need to be in the first frame
In by manual or automatic mode select tracked target, then analyze target color, Texture eigenvalue, build target mould
Plate, in subsequent frames, it is not necessary to extract moving target again, only need to go out the To Template i.e. by certain decision search in the picture
Realize the tracking to target.Mean Shift are exactly a kind of tracking based on model, using color or grey level histogram
The statistical property of distribution is describing target signature.To all of pixel in initial frame objective area in image, feature is calculated empty
Between in each characteristic value probability, that is, set up target masterplate.To there may be the candidate regions of target in every two field picture afterwards
Domain calculates characteristic value, using similarity function metric objective masterplate and the similitude of present frame candidate region.By seeking similitude
Function maxima obtains vectorial with regard to the Mean Shift of target, and this vector is exactly that target turns from initial position to correct position
The vector of shifting, the correct position of target can be just found according to this vector.
The target in image is described with the distribution of color of a target, target's center is located at x0, then object module
It is expressed as vectorWherein：
WhereinTime is represented with y
The center of target is selected, then candidate family is expressed as vectorWherein：
Wherein
K [] is kernel function in two above formula, and m is the number of color characteristic in color space, and u is the spy of color characteristic
Value indicative, b (xi) it is x in imageiPixel value (color value) at point.Target following can be reduced to find optimum y so that q with
P (y) is at utmost similar.
Step 5：The result of detection, analysis can be transferred to field controller PLC, be entered to moving pricker device all parts certainly
Row control.
It is a kind of that pricker method is moved based on machine vision certainly, also include：By being industry camera configuration in the position of pricker frame 7
One head, after head moves to station, by field controller control head motion, by the shooting visual field 5 of industry camera
The drill steel installation site of iron notch drill is moved to, the picture that industry camera shoots is sent in flush bonding processor, passed through
Gaussian filtering, sharpening, Sobel operator edge detections and Hough straight-line detections.Image after a series of process, in figure
The straight line of presence can be labeled out, there is straight line at iron notch drill, that is, represent and assemble drill steel 8.
Gaussian filtering：It is a kind of linear smoothing filtering, it is adaptable to eliminate Gaussian noise, is widely used in subtracting for image procossing
Make an uproar process.Popular says, gaussian filtering is exactly that average process is weighted to entire image, the value of each pixel, all
Obtain after being weighted averagely by other pixel values in itself and neighborhood.The concrete operations of gaussian filtering are：With a mould
Each pixel in plate (or claiming convolution, mask) scan image, with the weighted average gray scale of pixel in the neighborhood of template determination
Value goes the value of alternate template central pixel point.
Sharpen：Using the instrument of sharpening can quick focus blur edge, in raising image the definition at a certain position or
Focal length degree, makes the color of image specific region distincter.
Sobel operator edge detections：In rim detection, conventional template is Sobel operators, and technically, it is one
Individual discrete first difference operator, for calculating the approximation of the First-order Gradient of brightness of image function.In any point of image
Using this operator, it will produce the corresponding gradient vector of point or its law vector.Matrix of the operator comprising two groups of 3x3, point
Not Wei transverse direction and longitudinal direction, it is made into planar convolution with image, you can draw the brightness difference approximation of transverse direction and longitudinal direction respectively.Such as
Fruit represents original image with A, and Gx and Gy represents respectively the image of Jing transverse direction and longitudinal direction rim detections, and its formula is as follows:
The transverse direction and longitudinal direction gradient approximation of each pixel of image can be combined to calculate gradient with below equation
Size.
Then below equation is can use to calculate gradient direction.
Hough straight-line detections：The general principle of Hough transform is the duality using point with line, and original image is empty
Between given curve negotiating curve representation form be changed into a point of parameter space, thus given curve in original image
Test problems be converted into the spike problem found in parameter space, namely that detection overall permanence is converted into detection local is special
Property.Hough straight-line detection thoughts are：A point under coordinates of original image coordinates system corresponded to one in parameter coordinate system it is straight
Line, the straight line of same parameter coordinate system has corresponded to a point under original coordinate system, then, presents under original coordinate system straight
A little, their slope and intercept is identical for the institute of line, so they correspond to same point under parameter coordinate system.So
After by under each spot projection under original coordinate system to parameter coordinate system, see under parameter coordinate system either with or without accumulation point, this
The accumulation point of sample has just corresponded to the straight line under original coordinate system.
Specific implementation step：
1st, hardware is installed according to structure principle chart, mainly including industry camera, flush bonding processor.Due to moving pricker certainly
Whether " in place " vision module of machine is mainly responsible for drill steel 8 and manipulator 4 in detection " pricker frame ", therefore image acquisition region should
This covers the region of manipulator 4 and drill steel as much as possible 8, as shown in Figure 2.
In Fig. 2：1 is trolley track, and 2 is industry camera, and 3 is dolly, and 4 is manipulator, and 5 is camera view region, 6
It is to take drill rod generation, 7 is pricker frame, and 8 is drill steel.
2nd, after industry camera is installed, the image that it is photographed is as shown in Figure 3：According to the figure that industry camera is collected
Picture, the graphical analysis of vision module and processes and processes storehouse using the computer vision of increasing income based on OpenCV, using gaussian filtering,
The steps such as sharpening, Sobel operator edge detections and Hough transform straight-line detection, separate from background, detect drill steel 8
Position.
3rd, Fig. 3 cathetus and the angle a with reference to dotted line are calculated, wherein being that system initialisation phase detection is in reference to straight line
The drill steel 8 of " in place " state has obtained initial position.If angle a<B (wherein b is experiment threshold value), then system thinks drill steel 8
" in place ", it is 1 then to put POLE_STATUS registers, is otherwise 0, as shown in table 1.
4th, using target detection and target tracking algorism, the position of mechanical gripping arm is detected, and real-time tracking manipulator 4
Motion state, judges the relative position relation (blocking) of manipulator 4 and drill steel 8, and according to both coverage extents manipulator 4 is judged
Whether " in place ", if mechanical gripping arm in place, puts MAIN_STATUS registers for 1, otherwise set to 0.Wherein mechanical arm is " just
Position " is if shown in schematic diagram 4：That is during d '-Δ≤d≤d ', system thinks mechanical gripping arm in place, and wherein d ' is experiment test
Optimal value, Δ is tolerance.
5th, the present invention is also equipped with monitoring the function that drill steel whether has been assembled with iron notch drill.By in pricker frame 7
Position is that camera configures a head, after walking carriage moves to station, by field controller control head motion, will be taken the photograph
The shooting visual field of camera moves to the drill steel installation site of iron notch drill, and the picture that camera shoots is sent into embedded processing
In device, by gaussian filtering, sharpening, Sobel operator edge detections and Hough straight-line detections, whether there is in detection image
Special characteristic straight line.
6th, MODBUS communication protocols：
The present invention uses MODBUS standard for Fieldbus, and the RTU data forms of MODBUS buses adopt the institute of table 2
Show.
The holding register explanation of table 1
The MODBUS RTU data frame formats of table 2
7th, apparatus of the present invention include two parts of hardware and software：Hardware components are mainly by industry camera and embedded
Reason device two parts composition；Software section mainly include IMAQ, MODBUS protocol stacks, graphical analysis, process, parameter setting,
Debugging and task scheduling.
Claims (2)
1. it is a kind of that pricker method is moved based on machine vision certainly, it is characterised in that to comprise the following steps：
Step 1：Mounting industrial camera, flush bonding processor, the image acquisition region of industry camera cover manipulator (4),
The region of drill steel (8)；
Step 2：It is processed according to the image that industry camera is collected, using gaussian filtering, sharpening, rim detection,
The steps such as straight-line detection, separate, detect the position of drill steel (8) from background；
Step 3：Judge drill steel (8) whether " in place "；
Step 4：Using target detection and target tracking algorism, the position of mechanical gripping arm, and real-time tracking manipulator (4) are detected
Motion state, judge the relative position relation of manipulator (4) and drill steel (8), according to both coverage extents, judge manipulator
(4) whether " in place "；
Step 5：The result of detection, analysis can be transferred to field controller, be controlled to moving pricker all parts certainly.
It is 2. a kind of according to claim 1 that pricker machine is moved based on machine vision certainly, it is characterised in that：By in pricker frame (7)
Position is that industry camera configures a head, after head moves to station, by field controller control head motion, by work
The shooting visual field (5) of industry camera moves to drill steel (8) installation site of iron notch drill, and the picture that industry camera is shot is sent out
In being sent to flush bonding processor, by gaussian filtering, sharpening, Sobel operator edge detections and Hough straight-line detections, detection
Whether there is special characteristic straight line in image, whether be assembled with drill steel on iron notch drill so as to monitor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610914755.1A CN106636510B (en) | 2016-10-20 | 2016-10-20 | It is a kind of that pricker method is moved based on machine vision certainly |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610914755.1A CN106636510B (en) | 2016-10-20 | 2016-10-20 | It is a kind of that pricker method is moved based on machine vision certainly |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106636510A true CN106636510A (en) | 2017-05-10 |
| CN106636510B CN106636510B (en) | 2019-07-30 |
Family
ID=58856892
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610914755.1A Active CN106636510B (en) | 2016-10-20 | 2016-10-20 | It is a kind of that pricker method is moved based on machine vision certainly |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106636510B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116083669A (en) * | 2023-01-03 | 2023-05-09 | 中冶南方工程技术有限公司 | Method for automatically replacing iron rod in front of blast furnace |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105759720A (en) * | 2016-04-29 | 2016-07-13 | 中南大学 | Mechanical arm tracking and positioning on-line identification and correction method based on computer vision |
| CN105930854A (en) * | 2016-04-19 | 2016-09-07 | 东华大学 | Manipulator visual system |
-
2016
- 2016-10-20 CN CN201610914755.1A patent/CN106636510B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930854A (en) * | 2016-04-19 | 2016-09-07 | 东华大学 | Manipulator visual system |
| CN105759720A (en) * | 2016-04-29 | 2016-07-13 | 中南大学 | Mechanical arm tracking and positioning on-line identification and correction method based on computer vision |
Non-Patent Citations (1)
| Title |
|---|
| 贾祥瑞: "高炉炉前结合工业机器人技术的控制设计", 《冶金电气电气应用》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116083669A (en) * | 2023-01-03 | 2023-05-09 | 中冶南方工程技术有限公司 | Method for automatically replacing iron rod in front of blast furnace |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106636510B (en) | 2019-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112233133B (en) | Power plant high-temperature pipeline defect detection and segmentation method based on OTSU and area growth method | |
| CN109325935B (en) | Power transmission line detection method based on unmanned aerial vehicle image | |
| CN108109137A (en) | The Machine Vision Inspecting System and method of vehicle part | |
| CN110315525A (en) | A kind of robot workpiece grabbing method of view-based access control model guidance | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN110807355A (en) | Pointer instrument detection and reading identification method based on mobile robot | |
| CN105184812B (en) | A kind of pedestrian based on target following hovers detection method | |
| CN109543587A (en) | A kind of high-temperature fusion fluid flow rate detection method and system | |
| CN105865329B (en) | The acquisition system and method for the bundled round steel end face center coordinate of view-based access control model | |
| CN109911481B (en) | Cabin frame target visual identification and positioning method and system for metallurgical robot plugging | |
| CN102542289A (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
| CN103870824B (en) | A kind of face method for catching and device during Face datection tracking | |
| CN109815822B (en) | Patrol diagram part target identification method based on generalized Hough transformation | |
| CN108491851A (en) | A kind of container lockhole based on machine vision is quick to be identified and suspender method for correcting error | |
| CN106780526A (en) | A kind of ferrite wafer alligatoring recognition methods | |
| CN108109154A (en) | A kind of new positioning of workpiece and data capture method | |
| CN109434251A (en) | A kind of weld image tracking based on particle filter | |
| CN111539927B (en) | Detection method of automobile plastic assembly fastening buckle missing detection device | |
| CN107301634A (en) | A kind of robot automatic sorting method and system | |
| CN111932490B (en) | Visual system grabbing information extraction method for industrial robot | |
| CN113822810A (en) | Method for positioning workpiece in three-dimensional space based on machine vision | |
| CN110751669A (en) | Novel CBOCP online infrared converter tapping steel flow automatic detection and tracking method and system | |
| CN105678737A (en) | Digital image corner point detection method based on Radon transform | |
| CN104966302B (en) | A kind of detection localization method of any angle laser cross | |
| CN109583306B (en) | Bobbin residual yarn detection method 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |