CN103051872A - Method for detecting conveyor belt deviation based on image edge extraction - Google Patents

Method for detecting conveyor belt deviation based on image edge extraction Download PDF

Info

Publication number
CN103051872A
CN103051872A CN2012105505770A CN201210550577A CN103051872A CN 103051872 A CN103051872 A CN 103051872A CN 2012105505770 A CN2012105505770 A CN 2012105505770A CN 201210550577 A CN201210550577 A CN 201210550577A CN 103051872 A CN103051872 A CN 103051872A
Authority
CN
China
Prior art keywords
edge
image
conveyer belt
rotating shaft
sideslip
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
Application number
CN2012105505770A
Other languages
Chinese (zh)
Other versions
CN103051872B (en
Inventor
马艳华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qingdao University of Science and Technology
Original Assignee
Qingdao University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Qingdao University of Science and Technology filed Critical Qingdao University of Science and Technology
Priority to CN201210550577.0A priority Critical patent/CN103051872B/en
Publication of CN103051872A publication Critical patent/CN103051872A/en
Application granted granted Critical
Publication of CN103051872B publication Critical patent/CN103051872B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention discloses a method for detecting conveyor belt deviation based on image edge extraction. The method comprises the following steps of: collecting original images; acquiring image information; detecting edge profiles; repairing the edge profiles; extracting edges; marking the edges; calculating a deviation distance; and determining the deviation degree. The edge profiles of the original images collected by video equipment are detected and repaired until conveyor belts in the images and the edges of rotating shafts are extracted, and whether the conveyor belts are deviated is detected through the extracted edges; and the accuracy of deviation detection is improved.

Description

Belt deflection detection method based on Edge extraction
Technical field
The invention belongs to technical field of image processing, specifically, relate to a kind of belt deflection detection method of processing based on image, more particularly, relate to a kind of belt deflection detection method based on Edge extraction.
Background technology
Conveyer belt is the key equipment of conveying system, and its safety, stable operation directly have influence on production operation.The sideslip of conveyer belt is the most common failure of ribbon conveyer, and sideslip gently then can cause spreading, affect manufacturing schedule and infringement conveyer belt and associated machines; If long distance is carried, in the mine use procedure, also can because belt deflection increases the conveyer belt running resistance, may cause the serious accidents such as mine fire, casualties.
Along with the development of the improving constantly of automatic technology, communication and control technology, the conveyer belt monitoring technique is also progressively improving, and is bringing into play great role such as automatic correcting error device for conveying belt etc. in conveying system.Exactly because and the also raising of automation and intellectualized technology, the often unmanned guard that brings causes a certain function in case lost efficacy and can not in time find.In addition, in conveying system, although video monitoring equipment is also moving, its practical significance only is video monitor, also needs the manual observation monitoring image to find and judges whether sideslip of conveyer belt.Therefore, automatically, in time find whether sideslip of conveyer belt, and can notify the related personnel in accurate, appropriate mode, for prevention and reduce serious accident generation, to eliminate potential faults, guarantee the person and device security, raise labour productivity be vital.
Summary of the invention
The object of the present invention is to provide a kind of belt deflection detection method based on Edge extraction, carry out edge contour detection, edge contour reparation by the original image to the video equipment collection, until extract the edge of conveyer belt and rotating shaft in the image, and utilize whether sideslip of the rim detection conveyer belt extract, improved the accuracy that sideslip detects.
For achieving the above object, the present invention adopts following technical proposals to be achieved:
A kind of belt deflection detection method based on Edge extraction, described method comprises the steps:
A1, collection original image: the original image that obtains the rotating shaft that includes conveyer belt and the conveyer belt left and right sides;
A2, obtain image information: the width that obtains original image
Figure 222790DEST_PATH_IMAGE001
, highly
Figure 202247DEST_PATH_IMAGE002
And gray level image
Figure 795034DEST_PATH_IMAGE003
A3, edge contour detect: for gray level image
Figure 190243DEST_PATH_IMAGE003
, the definition running direction of conveyor belt is
Figure 66932DEST_PATH_IMAGE004
Direction is calculated gray value poor of arbitrary neighborhood two row, obtains error image
Figure 889395DEST_PATH_IMAGE005
Set gray threshold
Figure 969477DEST_PATH_IMAGE006
, according to the following equation to error image
Figure 168377DEST_PATH_IMAGE007
Make binary conversion treatment, obtain the edge contour image of binaryzation
Figure 899573DEST_PATH_IMAGE008
,
For arbitrarily
Figure 697197DEST_PATH_IMAGE009
,
Figure 451526DEST_PATH_IMAGE010
,
Figure 250855DEST_PATH_IMAGE011
Figure 774240DEST_PATH_IMAGE012
A4, edge contour reparation: adopt the Filtering Template based on shape and directivity With the edge contour image
Figure 321076DEST_PATH_IMAGE008
Make convolution, obtain image
Figure 596200DEST_PATH_IMAGE014
, definition , then edge contour reparation image is
Figure 387887DEST_PATH_IMAGE016
A5, edge extracting: repair image from edge contour according to following formula
Figure 179125DEST_PATH_IMAGE017
Middle extraction edge obtains edge image
Figure 70989DEST_PATH_IMAGE018
?
Figure 568966DEST_PATH_IMAGE019
；
A6, edge labelling: use two-dimensional matrix
Figure 402930DEST_PATH_IMAGE020
, ,
Figure 46194DEST_PATH_IMAGE022
With
Figure 398678DEST_PATH_IMAGE023
Rotating shaft edge, difference mark conveyer belt left side, conveyer belt left side edge, belt right lateral edges and rotating shaft edge, conveyer belt right side, the element in the matrix is the coordinate figure of the point on the respective edges;
A7, sideslip are apart from calculating: calculate rotating shaft edge, conveyer belt left side
Figure 403543DEST_PATH_IMAGE020
With the conveyer belt left side edge
Figure 654527DEST_PATH_IMAGE021
Between average distance
Figure 75144DEST_PATH_IMAGE024
, the belt right lateral edges With rotating shaft edge, conveyer belt right side
Figure 723480DEST_PATH_IMAGE023
Between average distance
Figure 461760DEST_PATH_IMAGE025
, calculate
Figure 686068DEST_PATH_IMAGE026
With
Figure 809881DEST_PATH_IMAGE027
Range deviation
Figure 907282DEST_PATH_IMAGE028
And intermediate distance
Figure 319808DEST_PATH_IMAGE029
A8, sideslip degree are determined: according to formula Determine conveyer belt whether sideslip and sideslip degree;
Wherein,
Figure 591707DEST_PATH_IMAGE031
Be the position of pixel in the gray level image,
Figure 597359DEST_PATH_IMAGE009
,
Figure 497181DEST_PATH_IMAGE010
Figure 125609DEST_PATH_IMAGE032
Be the coordinate position of Filtering Template,
Figure 240327DEST_PATH_IMAGE033
,
Figure 600901DEST_PATH_IMAGE034
Figure 50337DEST_PATH_IMAGE035
With
Figure 420138DEST_PATH_IMAGE036
Be respectively the length at conveyer belt rotating shaft edge, left side and rotating shaft edge, conveyer belt right side,
Figure 123783DEST_PATH_IMAGE037
For rotating shaft edge, conveyer belt left side exists
Figure 983155DEST_PATH_IMAGE004
Starting point on the direction and conveyer belt left side edge exist Starting point edge on the direction
Figure 578532DEST_PATH_IMAGE004
Distance on the direction,
Figure 589214DEST_PATH_IMAGE038
For rotating shaft edge, conveyer belt right side exists Starting point on the direction and belt right lateral edges exist
Figure 715618DEST_PATH_IMAGE004
Starting point edge on the direction
Figure 237342DEST_PATH_IMAGE004
Distance on the direction.
Aforesaid method, described Filtering Template Can be 1 benchmark template for following width:
Figure 303704DEST_PATH_IMAGE039
。
Further, for improving the repair ability of edge contour, described Filtering Template
Figure 700182DEST_PATH_IMAGE013
Be preferably described benchmark template
Figure 949897DEST_PATH_IMAGE040
Expanding expansion templates after several values are 1 pixel in left and right sides bilateral symmetry on the vertical line direction of each point on the described line, also is width greater than 1 Filtering Template.
Aforesaid method, for improving data processing speed, the original image that described step a1 gathers is the RGB coloured image
Figure 997488DEST_PATH_IMAGE041
, its gray level image is
Aforesaid method, in described step a3, gray threshold
Figure 191020DEST_PATH_IMAGE006
Be preferably
Aforesaid method also comprises the steps: after described step a8
A9, according to the sideslip degree
Figure 146523DEST_PATH_IMAGE044
The alarm signal of size output different stage.
Compared with prior art, advantage of the present invention and good effect are: the present invention carries out edge contour detection, edge contour reparation by the original image to the video equipment collection, until extract the edge of conveyer belt and rotating shaft in the image, whether the rim detection conveyer belt that utilization is extracted sideslip, guarantee continuity and the accuracy at edge, improved the reliability and stability of sideslip testing result.
After reading the specific embodiment of the present invention by reference to the accompanying drawings, other characteristics of the present invention and advantage will become clearer.
Description of drawings
Fig. 1 is the flow chart that the present invention is based on an embodiment of belt deflection detection method of Edge extraction;
Fig. 2 to Fig. 6 is the image after different step is processed among Fig. 1 embodiment;
Fig. 7 (a) and Fig. 7 (b) are that Fig. 1 embodiment is used, width is 1 benchmark Filtering Template schematic diagram;
Fig. 8 (a) and Fig. 8 (b) are that Fig. 1 embodiment is used, width is 3 extended filtering template schematic diagram.
Embodiment
Below in conjunction with the drawings and specific embodiments technical scheme of the present invention is described in further detail.
At first, brief description technical thought of the present invention: for the conveyer belt of production scene with long distance, for guaranteeing the conveyer belt even running, below the conveyer belt left and right sides, be respectively arranged with many countershafts, be referred to as conveyer belt left side rotating shaft and the rotating shaft of conveyer belt right side.The left hand edge of left side rotating shaft and the right hand edge of right side rotating shaft lay respectively at the outside of conveyer belt left side edge and belt right lateral edges, and the position of rotating shaft is changeless.Thus, just can be with the rotating shaft of conveyer belt left side and the rotating shaft of conveyer belt right side as the reference edge, obtain actual left side edge and right side edge in the conveyer belt running, detect whether sideslip of conveyer belt according to the change of distance between the rotating shaft edge of belt edges and respective side, and the sideslip degree during sideslip.
Please refer to an embodiment of the belt deflection detection method that the present invention is based on Edge extraction shown in Fig. 1 to Fig. 6, this embodiment transports the conveyer belt of coal as example in the detection colliery.Wherein, Fig. 1 is the flow chart of this embodiment, and Fig. 2 to Fig. 6 is the image after different step is processed, and is used for the concrete result of key diagram 1 each step of flow process, and Fig. 2 to Fig. 6 is the result images after the parts of images in the original image is processed.
This embodiment detects conveyer belt, and the concrete processing procedure of sideslip and sideslip degree size is as follows:
Step 101: flow process begins.
Step 102: gather original image.
Utilization is arranged on the camera photographic images at belt conveyance scene as original image, by adjusting the setting position of camera, guarantees not only to include conveyer belt in the captured original image, also includes the rotating shaft that is arranged on the conveyer belt left and right sides.
Step 103: obtain image information.
Consider that original image is generally larger, all processing speed is slower, and, can there be the added text information such as shooting time, spot for photography in the original image, easily detection is caused larger interference.Given this, this embodiment takes to intercept from original image and comprises conveyer belt and both sides rotating shaft thereof and disturb less parts of images (concrete intercepting can be determined according to riding position, the position of added text information in image of camera) as the real image of subsequent treatment.Real image is carried out data analysis, obtain the width of image
Figure 440233DEST_PATH_IMAGE001
, highly
Figure 732674DEST_PATH_IMAGE002
, judge that simultaneously image is coloured image or gray level image.At present, the captured image of camera generally is coloured image, with the RGB trichromatic specification is
Figure 652088DEST_PATH_IMAGE041
For coloured image, need to be converted into gray level image:
Figure 346375DEST_PATH_IMAGE042
, wherein,
Figure 825634DEST_PATH_IMAGE031
Be the position of pixel in the gray level image,
Figure 933267DEST_PATH_IMAGE009
,
Figure 328476DEST_PATH_IMAGE010
Image after greyscale transformation
Figure 690319DEST_PATH_IMAGE003
As shown in Figure 2.
Except adopting above-mentioned formula to calculate after the gray level image, can also adopt other formula to calculate, for example, and the different account form of RGB three look proportions, this embodiment is not construed as limiting this.
Step 104: the edge contour in the detected image.
For gray level image
Figure 778360DEST_PATH_IMAGE003
, the definition running direction of conveyor belt is the edge
Figure 107710DEST_PATH_IMAGE004
Direction.Then, calculate gray value poor of arbitrary neighborhood two row, obtain error image
Figure 306611DEST_PATH_IMAGE005
Error image
Figure 522959DEST_PATH_IMAGE007
As shown in Figure 3.
Then, set gray threshold
Figure 844219DEST_PATH_IMAGE006
, according to the following equation to error image
Figure 598549DEST_PATH_IMAGE007
Make binary conversion treatment, the edge contour image after the acquisition binaryzation as shown in Figure 4
Figure 148610DEST_PATH_IMAGE008
:
For arbitrarily
Figure 671995DEST_PATH_IMAGE009
,
Figure 164156DEST_PATH_IMAGE010
,
Figure 950322DEST_PATH_IMAGE011
Figure 491025DEST_PATH_IMAGE012
Wherein, gray threshold
Figure 196813DEST_PATH_IMAGE006
Can select according to image size and accuracy of detection.Preferably, gray threshold Be error image
Figure 73950DEST_PATH_IMAGE007
The mean value of middle gray value of having a few, namely
After above-mentioned processing, conveyer belt left side rotating shaft edge contour, conveyer belt left side edge profile, belt right lateral edges profile and the conveyer belt right side rotating shaft edge contour that can obtain to be arranged in order from left to right among Fig. 4.And, from this edge contour image, it can also be seen that the edge contour image
Figure 713059DEST_PATH_IMAGE008
For having an edge frame of one fixed width, and the image border has defective (edge is discontinuous) and/or unnecessary burr.This is because when taking original image, be subjected to the refraction of light of transportation thing in the reflection, conveyer belt of light light etc. and reflection etc. impact and so that the belt edges that detects or conveyer belt rotating shaft edge produce defective or burr.For guaranteeing detection accuracy, this embodiment adopts the filtering of Filtering Template edge contour images, and to repair edge contour, specific as follows step is described.
Step 105: adopt Filtering Template edge contour images to repair.
Consider that belt edges and conveyer belt rotating shaft edge all are the linear pattern shapes, and because of these edges of problem of shooting angle be the edge basically
Figure 32176DEST_PATH_IMAGE004
Direction or depart from a little, therefore, this embodiment adopts the Filtering Template based on shape and directivity
Figure 248394DEST_PATH_IMAGE013
The edge contour images is repaired.Concrete repair process is as follows:
At first, utilize Filtering Template
Figure 193216DEST_PATH_IMAGE013
With the edge contour image
Figure 545700DEST_PATH_IMAGE008
Make convolution, obtain image
Figure 301297DEST_PATH_IMAGE014
Then, definition
Figure 739232DEST_PATH_IMAGE015
, then edge contour reparation image is
Figure 222166DEST_PATH_IMAGE016
In this formula,
Figure 510714DEST_PATH_IMAGE032
Be the coordinate position of Filtering Template, ,
Figure 346132DEST_PATH_IMAGE034
For Filtering Template , its size is can not be excessive also unsuitable too small, oversize, and not only processing speed is slow, may miss the real edge pixel point of filtering, and undersized, then very likely thinks noise spot by mistake marginal point and keeps, and has reduced filter effect.It is 1 benchmark template that Filtering Template can adopt width, and its expression formula is:
Figure 710566DEST_PATH_IMAGE045
。
And consider the uncertainty of edge contour direction, and the preferred expansion templates with larger width that adopts, this expansion templates is by the benchmark template
Figure 57233DEST_PATH_IMAGE040
In value be on the vertical line direction of the each point on the line that consists of of each pixel of 1 left and right sides bilateral symmetry to expand several values be a Filtering Template that obtains behind 1 the pixel.Fig. 7 shows and is of a size of 5*5, width is a benchmark Filtering Template of 1, and Figure 8 shows that size is that 7*3, width are extended filtering templates of 3.In this embodiment, preferably adopt the extended filtering template edge contour images shown in Fig. 8 Carry out filtering, thereby the edge contour that obtains is as shown in Figure 5 repaired image
Figure 45229DEST_PATH_IMAGE017
Step 106: from edge contour reparation image, extract the edge.
The edge contour that step 105 obtains is repaired image
Figure 226812DEST_PATH_IMAGE047
General can not be real edge, can have certain width.For ease of calculating, guarantee simultaneously the rim detection accuracy, need to repair from edge contour and extract real edge the image.
Consider that belt edges and conveyer belt rotating shaft edge generally are the edges
Figure 2012105505770100002DEST_PATH_IMAGE049
Direction or
Figure 744381DEST_PATH_IMAGE049
Slightly inclined to one side
Figure 457253DEST_PATH_IMAGE051
Direction, therefore, adopt the edge
Figure 85680DEST_PATH_IMAGE051
The method of direction finding central point is repaired the image from edge contour and is extracted the edge.Specifically, be to repair image according to following formula from edge contour
Figure 121770DEST_PATH_IMAGE047
Middle extraction edge obtains edge image
?
Figure 945162DEST_PATH_IMAGE055
。
The edge image that extracts
Figure 111701DEST_PATH_IMAGE057
As shown in Figure 6, include altogether four edges edge line, be followed successively by from left to right rotating shaft edge, conveyer belt left side, conveyer belt left side edge, belt right lateral edges and rotating shaft edge, conveyer belt right side.Comparison diagram 5 and Fig. 6 can find out that the edge contour line width of Fig. 5 is larger, and the edge of Fig. 6 be width is 1 line, and consistent with the centre line shape of edge wheel profile, illustrate that the edge through said method extracts is accurately.
Step 107: marker edge, and calculate the sideslip distance.
With four two-dimensional matrixs
Figure 80925DEST_PATH_IMAGE059
, ,
Figure 549132DEST_PATH_IMAGE063
With
Figure 722625DEST_PATH_IMAGE065
Rotating shaft edge, difference mark conveyer belt left side, conveyer belt left side edge, belt right lateral edges and rotating shaft edge, conveyer belt right side, the element in the matrix is the coordinate figure of the point on the respective edges.
With rotating shaft edge, conveyer belt left side
Figure 546355DEST_PATH_IMAGE059
Be example,
Figure 514311DEST_PATH_IMAGE067
, wherein: On the rotating shaft edge, expression conveyer belt left side the
Figure 931834DEST_PATH_IMAGE069
Individual pixel, namely of matrix
Figure 62602DEST_PATH_IMAGE069
OK,
Figure 263776DEST_PATH_IMAGE071
,
Figure RE-DEST_PATH_IMAGE073
Be the length at rotating shaft edge, conveyer belt left side,
Figure DEST_PATH_IMAGE075
Expression
Figure 651464DEST_PATH_IMAGE059
Matrix column,
Figure DEST_PATH_IMAGE077
The time,
Figure DEST_PATH_IMAGE079
Element value in the matrix is pixel on the rotating shaft edge, conveyer belt left side The axial coordinate value,
Figure DEST_PATH_IMAGE081
The time,
Figure DEST_PATH_IMAGE083
Element value in the matrix is pixel on the rotating shaft edge, conveyer belt left side
Figure DEST_PATH_IMAGE085
The axial coordinate value.
At mark after each edge, calculate respectively rotating shaft edge, conveyer belt left side
Figure 637186DEST_PATH_IMAGE059
With the conveyer belt left side edge
Figure 9261DEST_PATH_IMAGE061
Between average distance
Figure DEST_PATH_IMAGE087
, the belt right lateral edges
Figure 893035DEST_PATH_IMAGE063
With rotating shaft edge, conveyer belt right side
Figure 8758DEST_PATH_IMAGE065
Between average distance
Figure DEST_PATH_IMAGE089
, and calculate
Figure DEST_PATH_IMAGE091
With
Figure DEST_PATH_IMAGE093
Range deviation
Figure DEST_PATH_IMAGE095
And intermediate distance
Figure DEST_PATH_IMAGE097
In the following formula,
Figure DEST_PATH_IMAGE099
For rotating shaft edge, conveyer belt left side exists
Figure 534024DEST_PATH_IMAGE049
Starting point on the direction and conveyer belt left side edge exist
Figure 77001DEST_PATH_IMAGE049
Starting point edge on the direction
Figure 369442DEST_PATH_IMAGE049
Distance on the direction,
Figure DEST_PATH_IMAGE101
For rotating shaft edge, conveyer belt right side exists
Figure 39589DEST_PATH_IMAGE049
Starting point on the direction and belt right lateral edges exist
Figure 796192DEST_PATH_IMAGE049
Starting point edge on the direction
Figure 447754DEST_PATH_IMAGE049
Distance on the direction.That is to say,
Figure 306119DEST_PATH_IMAGE099
Satisfy , and
Figure 763645DEST_PATH_IMAGE101
Satisfy
Figure 2012105505770100002DEST_PATH_IMAGE105
The below illustrates the acquisition methods of above-mentioned four two-dimensional matrixs take edge image shown in Figure 6 as example:
For highly being
Figure DEST_PATH_IMAGE107
Image, row bottom is
Figure DEST_PATH_IMAGE109
, with the first row of this row as two-dimensional matrix, then, from
Figure 190734DEST_PATH_IMAGE109
Beginning up, scan line by line, every row is successively scanning from left to right.For bottom
Figure 91825DEST_PATH_IMAGE109
Row has four edges edge line, the two-dimensional matrix that every edge is corresponding
Figure 358859DEST_PATH_IMAGE069
Be 1, and
Figure DEST_PATH_IMAGE111
, also be Each two-dimensional matrix the 1st row The axial coordinate value is as follows:
Figure DEST_PATH_IMAGE115
Figure 898741DEST_PATH_IMAGE109
Line scanning is complete, again scanning OK.For
Figure 970734DEST_PATH_IMAGE117
, the two-dimensional matrix that every edge is corresponding
Figure 459484DEST_PATH_IMAGE069
Be 2, namely
Figure DEST_PATH_IMAGE119
Each two-dimensional matrix the 2nd row
Figure 524392DEST_PATH_IMAGE085
The axial coordinate value is as follows:
Figure DEST_PATH_IMAGE121
Line by line scan successively according to the method described above corresponding pixel in four two-dimensional matrixs when obtaining every delegation and all having the four edges edge
Figure 857897DEST_PATH_IMAGE085
The axial coordinate value.
When continuing the edge When direction upwards scans, the only situation at remaining three edges can occur, generally speaking, therefore belt edges length, need judge first in the case greater than the rotating shaft edge length
Figure DEST_PATH_IMAGE123
Pixel whether belong to
Figure 139154DEST_PATH_IMAGE059
The edge, method is as follows:
Figure DEST_PATH_IMAGE125
Line by line scan successively according to the method described above corresponding pixel in the respective two-dimensional matrix when obtaining every delegation and all having three edges
Figure 742173DEST_PATH_IMAGE085
The axial coordinate value.
Work as the edge
Figure 933114DEST_PATH_IMAGE049
When direction upwards continues scanning, the only situation at two edges of left and right sides of remaining conveyer belt can occur, at this moment, can proceed mark, also can finish at this point, not affect testing result.
Step 108: determine the sideslip degree.
Definition belt deflection degree is
Figure DEST_PATH_IMAGE127
, judge the whether degree of sideslip and sideslip of conveyer belt according to the size of sideslip degree numerical value.The sideslip degree is larger, illustrates that sideslip is more serious.
For example, can be according to the sideslip degree
Figure DEST_PATH_IMAGE129
Size be divided into Three Estate,
Figure DEST_PATH_IMAGE131
Be the first order,
Figure DEST_PATH_IMAGE133
Be the second level,
Figure DEST_PATH_IMAGE135
Be the third level,
Figure DEST_PATH_IMAGE137
Less or the sideslip not of expression belt deflection degree.
Step 109: according to the alarm signal of sideslip degree output different stage.
When the belt deflection degree
Figure 471543DEST_PATH_IMAGE129
When reaching corresponding rank, the alarm signal of output appropriate level can also be taked corresponding control measure.If
Figure 762583DEST_PATH_IMAGE137
, it is not serious to judge that then conveyer belt departs from, and need not to report to the police.
Above steps is the processing procedure to piece image, and after handling, above steps is carried out in circulation, realizes the continuous detecting to conveyer belt.
Above-described embodiment carries out data analysis and process by the original image to Real-time Collection, and whether the automatic decision conveyer belt sideslip occurs, and presses the sideslip grading.When belt deflection, other are different and start corresponding warning function with its level, avoid the human error that only depends on artificial video monitor to occur, in time find aspect the fault more reliable, stable.
Above embodiment is only in order to illustrating technical scheme of the present invention, but not limits it; Although with reference to previous embodiment the present invention is had been described in detail, for the person of ordinary skill of the art, still can make amendment to the technical scheme that previous embodiment is put down in writing, perhaps part technical characterictic wherein is equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution break away from the spirit and scope of the present invention's technical scheme required for protection.

Claims (6)

1. the belt deflection detection method based on Edge extraction is characterized in that described method comprises the steps:
A1, collection original image: the original image that obtains the rotating shaft that includes conveyer belt and the conveyer belt left and right sides;
A2, obtain image information: the width that obtains original image
Figure 420614DEST_PATH_IMAGE001
, highly
Figure 670330DEST_PATH_IMAGE002
And gray level image
Figure 216455DEST_PATH_IMAGE003
A3, edge contour detect: for gray level image
Figure 526214DEST_PATH_IMAGE003
, the definition running direction of conveyor belt is the edge
Figure 659255DEST_PATH_IMAGE004
Direction is calculated gray value poor of arbitrary neighborhood two row, obtains error image
Figure 774978DEST_PATH_IMAGE005
Set gray threshold
Figure 614758DEST_PATH_IMAGE006
, according to the following equation to error image
Figure 157735DEST_PATH_IMAGE007
Make binary conversion treatment, obtain the edge contour image of binaryzation ,
For arbitrarily
Figure 871056DEST_PATH_IMAGE009
, ,
Figure 279220DEST_PATH_IMAGE011
Figure 324537DEST_PATH_IMAGE012
A4, edge contour reparation: adopt the Filtering Template based on shape and directivity
Figure 782063DEST_PATH_IMAGE013
With the edge contour image
Figure 393173DEST_PATH_IMAGE008
Make convolution, obtain image , definition
Figure 326678DEST_PATH_IMAGE015
, then edge contour reparation image is
Figure 259999DEST_PATH_IMAGE016
A5, edge extracting: repair image from edge contour according to following formula Middle extraction edge obtains edge image
? ；
A6, edge labelling: use two-dimensional matrix
Figure 866112DEST_PATH_IMAGE020
,
Figure 953279DEST_PATH_IMAGE021
,
Figure 117545DEST_PATH_IMAGE022
With
Figure 421487DEST_PATH_IMAGE023
Rotating shaft edge, difference mark conveyer belt left side, conveyer belt left side edge, belt right lateral edges and rotating shaft edge, conveyer belt right side, the element in the matrix is the coordinate figure of the point on the respective edges;
A7, sideslip are apart from calculating: calculate rotating shaft edge, conveyer belt left side
Figure 962190DEST_PATH_IMAGE020
With the conveyer belt left side edge
Figure 402398DEST_PATH_IMAGE021
Between average distance
Figure 65461DEST_PATH_IMAGE024
, the belt right lateral edges
Figure 528803DEST_PATH_IMAGE022
With rotating shaft edge, conveyer belt right side Between average distance
Figure 932026DEST_PATH_IMAGE025
, calculate With
Figure 716629DEST_PATH_IMAGE027
Range deviation And intermediate distance
Figure 13935DEST_PATH_IMAGE029
A8, sideslip degree are determined: according to formula
Figure 753221DEST_PATH_IMAGE030
Determine conveyer belt whether sideslip and sideslip degree;
Wherein,
Figure 191155DEST_PATH_IMAGE031
Be the position of pixel in the gray level image, ,
Figure 710441DEST_PATH_IMAGE010
Be the coordinate position of Filtering Template,
Figure 811438DEST_PATH_IMAGE033
,
Figure 35746DEST_PATH_IMAGE034
Figure 425139DEST_PATH_IMAGE035
With
Figure 709489DEST_PATH_IMAGE036
Be respectively the length at conveyer belt rotating shaft edge, left side and rotating shaft edge, conveyer belt right side,
Figure 423148DEST_PATH_IMAGE037
For rotating shaft edge, conveyer belt left side exists
Figure 451147DEST_PATH_IMAGE004
Starting point on the direction and conveyer belt left side edge exist
Figure 695047DEST_PATH_IMAGE004
Starting point edge on the direction
Figure 212616DEST_PATH_IMAGE004
Distance on the direction,
Figure 112439DEST_PATH_IMAGE038
For rotating shaft edge, conveyer belt right side exists
Figure 740866DEST_PATH_IMAGE004
Starting point on the direction and belt right lateral edges exist
Figure 776955DEST_PATH_IMAGE004
Starting point edge on the direction
Figure 966890DEST_PATH_IMAGE004
Distance on the direction.
2. method according to claim 1 is characterized in that, described Filtering Template
Figure 88430DEST_PATH_IMAGE013
Be following benchmark template:
Figure 520549DEST_PATH_IMAGE039
。
3. method according to claim 2 is characterized in that, described Filtering Template
Figure 676723DEST_PATH_IMAGE013
Be described benchmark template
Figure 536095DEST_PATH_IMAGE040
Expand expansion templates after several values are 1 pixel in left and right sides bilateral symmetry on the vertical line direction of each point on the described line.
4. method according to claim 1 is characterized in that, the original image that described step a1 gathers is the RGB coloured image
Figure 144931DEST_PATH_IMAGE041
, its gray level image is
Figure 115161DEST_PATH_IMAGE042
5. method according to claim 1 is characterized in that, in described step a3, and gray threshold
Figure 125842DEST_PATH_IMAGE006
Be error image The mean value of middle gray value of having a few.
6. each described method in 5 according to claim 1 is characterized in that, also comprises the steps: after described step a8
A9, according to the sideslip degree
Figure 813099DEST_PATH_IMAGE043
The alarm signal of size output different stage.
CN201210550577.0A 2012-12-18 2012-12-18 Based on the belt deflection detection method of Edge extraction Expired - Fee Related CN103051872B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210550577.0A CN103051872B (en) 2012-12-18 2012-12-18 Based on the belt deflection detection method of Edge extraction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210550577.0A CN103051872B (en) 2012-12-18 2012-12-18 Based on the belt deflection detection method of Edge extraction

Publications (2)

Publication Number Publication Date
CN103051872A true CN103051872A (en) 2013-04-17
CN103051872B CN103051872B (en) 2015-08-12

Family

ID=48064359

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210550577.0A Expired - Fee Related CN103051872B (en) 2012-12-18 2012-12-18 Based on the belt deflection detection method of Edge extraction

Country Status (1)

Country Link
CN (1) CN103051872B (en)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105083912A (en) * 2015-07-07 2015-11-25 青岛科技大学 Conveyer belt deviation detecting method based on image identification
CN106423913A (en) * 2016-09-09 2017-02-22 华侨大学 Construction waste sorting method and system
CN107690421A (en) * 2015-05-29 2018-02-13 肖特股份有限公司 Method and apparatus for reducing lateral bending on thin glass
CN109724776A (en) * 2017-10-30 2019-05-07 中冶长天国际工程有限责任公司 A kind of determination method and device of the grid section damaged condition of sintering pallet
CN110641947A (en) * 2019-10-16 2020-01-03 山东中衡光电科技有限公司 Intelligent inspection robot system for bulk cargo conveyor and detection method thereof
CN110697373A (en) * 2019-07-31 2020-01-17 湖北凯瑞知行智能装备有限公司 Conveying belt deviation fault detection method based on image recognition technology
CN111908060A (en) * 2020-08-31 2020-11-10 国电浙能宁东发电有限公司 Power plant coal conveying belt deviation monitoring and early warning device and method
CN112320266A (en) * 2020-11-19 2021-02-05 太仓武港码头有限公司 Management method of intelligent management system of belt conveyor
CN112919050A (en) * 2021-02-04 2021-06-08 华润电力技术研究院有限公司 Conveyor belt monitoring method, device, equipment and computer readable storage medium
CN113112485A (en) * 2021-04-20 2021-07-13 中冶赛迪重庆信息技术有限公司 Belt conveyor deviation detection method, system, equipment and medium based on image processing
CN113344905A (en) * 2021-06-28 2021-09-03 燕山大学 Strip deviation amount detection method and system
CN113762283A (en) * 2021-08-30 2021-12-07 中铁工程装备集团有限公司 Method and device for monitoring deviation of conveying belt
CN113763376A (en) * 2021-09-17 2021-12-07 深圳市赛为智能股份有限公司 Conveyor belt deviation detection method and device, computer equipment and storage medium
CN115116010A (en) * 2022-08-29 2022-09-27 山东千颐科技有限公司 Belt deviation-preventing visual identification system based on image processing

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1532499B1 (en) * 2002-08-07 2007-10-24 Kimberly-Clark Worldwide, Inc. System and method for guiding a web in a web converting manufacturing process
JP2008230711A (en) * 2007-03-16 2008-10-02 Fuji Xerox Co Ltd Recording material conveying device and image forming device using the same
CN102275723A (en) * 2011-05-16 2011-12-14 天津工业大学 Machine-vision-based online monitoring system and method for conveyer belt
CN102319743A (en) * 2011-05-24 2012-01-18 重庆大学 Band steel deflection and floating quantity laser scanning detection method and deflection correction system
CN102360083A (en) * 2011-08-19 2012-02-22 上海高晶影像科技有限公司 Method for removing belt artifacts in images for line scanning X-ray security inspection equipment
CN102602681A (en) * 2012-01-13 2012-07-25 天津工业大学 Machine vision based online deviation fault detecting method for conveying belts

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1532499B1 (en) * 2002-08-07 2007-10-24 Kimberly-Clark Worldwide, Inc. System and method for guiding a web in a web converting manufacturing process
JP2008230711A (en) * 2007-03-16 2008-10-02 Fuji Xerox Co Ltd Recording material conveying device and image forming device using the same
CN102275723A (en) * 2011-05-16 2011-12-14 天津工业大学 Machine-vision-based online monitoring system and method for conveyer belt
CN102319743A (en) * 2011-05-24 2012-01-18 重庆大学 Band steel deflection and floating quantity laser scanning detection method and deflection correction system
CN102360083A (en) * 2011-08-19 2012-02-22 上海高晶影像科技有限公司 Method for removing belt artifacts in images for line scanning X-ray security inspection equipment
CN102602681A (en) * 2012-01-13 2012-07-25 天津工业大学 Machine vision based online deviation fault detecting method for conveying belts

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107690421B (en) * 2015-05-29 2021-09-21 肖特股份有限公司 Method and apparatus for reducing side bending on thin glass
CN107690421A (en) * 2015-05-29 2018-02-13 肖特股份有限公司 Method and apparatus for reducing lateral bending on thin glass
CN105083912A (en) * 2015-07-07 2015-11-25 青岛科技大学 Conveyer belt deviation detecting method based on image identification
CN106423913A (en) * 2016-09-09 2017-02-22 华侨大学 Construction waste sorting method and system
CN109724776A (en) * 2017-10-30 2019-05-07 中冶长天国际工程有限责任公司 A kind of determination method and device of the grid section damaged condition of sintering pallet
CN110697373A (en) * 2019-07-31 2020-01-17 湖北凯瑞知行智能装备有限公司 Conveying belt deviation fault detection method based on image recognition technology
CN110641947A (en) * 2019-10-16 2020-01-03 山东中衡光电科技有限公司 Intelligent inspection robot system for bulk cargo conveyor and detection method thereof
CN111908060A (en) * 2020-08-31 2020-11-10 国电浙能宁东发电有限公司 Power plant coal conveying belt deviation monitoring and early warning device and method
CN112320266A (en) * 2020-11-19 2021-02-05 太仓武港码头有限公司 Management method of intelligent management system of belt conveyor
CN112320266B (en) * 2020-11-19 2022-04-12 太仓武港码头有限公司 Management method of intelligent management system of belt conveyor
CN112919050A (en) * 2021-02-04 2021-06-08 华润电力技术研究院有限公司 Conveyor belt monitoring method, device, equipment and computer readable storage medium
CN113112485A (en) * 2021-04-20 2021-07-13 中冶赛迪重庆信息技术有限公司 Belt conveyor deviation detection method, system, equipment and medium based on image processing
CN113344905A (en) * 2021-06-28 2021-09-03 燕山大学 Strip deviation amount detection method and system
CN113762283A (en) * 2021-08-30 2021-12-07 中铁工程装备集团有限公司 Method and device for monitoring deviation of conveying belt
CN113762283B (en) * 2021-08-30 2024-04-09 中铁工程装备集团有限公司 Conveyor belt deviation monitoring method and device
CN113763376A (en) * 2021-09-17 2021-12-07 深圳市赛为智能股份有限公司 Conveyor belt deviation detection method and device, computer equipment and storage medium
CN113763376B (en) * 2021-09-17 2024-03-01 深圳市赛为智能股份有限公司 Conveyor belt offset detection method, conveyor belt offset detection device, computer equipment and storage medium
CN115116010A (en) * 2022-08-29 2022-09-27 山东千颐科技有限公司 Belt deviation-preventing visual identification system based on image processing

Also Published As

Publication number Publication date
CN103051872B (en) 2015-08-12

Similar Documents

Publication Publication Date Title
CN103051872A (en) Method for detecting conveyor belt deviation based on image edge extraction
CN104504388B (en) A kind of pavement crack identification and feature extraction algorithm and system
US11003940B2 (en) System and methods for automatic solar panel recognition and defect detection using infrared imaging
CN104297255B (en) A kind of dixie cup defective vision detection method and system and device
CN106920219B (en) Article defect detection method, image processing system and computer readable recording medium
CN104361314B (en) Based on infrared and transformer localization method and device of visual image fusion
CN104537651B (en) Proportion detecting method and system for cracks in road surface image
CN110793587B (en) Efficient and safe operation monitoring method for coal mine working surface belt conveyor
CN104506857A (en) Camera position deviation detection method and device
CN104608799A (en) Information fusion technology based train wheel set tread damage online detection and recognition method
CN104135660A (en) Detection method of contamination of image pickup module and detection system
CN102509075A (en) Remnant object detection method and device
CN106228546A (en) The detection method of a kind of board and device
CN201890600U (en) Machine vision belt tearing detecting device
CN110838097A (en) Conveyor belt offset measurement method based on machine vision
CN103927553B (en) Coal and rock recognition method based on multi-scale micro-lamination and contrast ratio joint distribution
CN104197836A (en) Vehicle lock assembly size detection method based on machine vision
CN103913149B (en) A kind of binocular range-measurement system and distance-finding method thereof based on STM32 single-chip microcomputer
CN106504231A (en) Component defects detection method and system
CN115144399B (en) Assembly quality detection method and device based on machine vision
CN103010258A (en) System and method for detecting cracks of fasteners of high-speed rails and subways
CN102609957A (en) Method and system for detecting picture offset of camera device
CN110310275A (en) A kind of chain conveyor defect inspection method based on image procossing
CN102663781A (en) Sub-pixel level welding center extraction method based on visual sense
CN112801965A (en) Sintering belt foreign matter monitoring method and system based on convolutional neural network

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20150812

Termination date: 20191218