CN104730083B - Steel cable core conveying belt joint twitches automatic testing method - Google Patents
Steel cable core conveying belt joint twitches automatic testing method Download PDFInfo
- Publication number
- CN104730083B CN104730083B CN201510164949.XA CN201510164949A CN104730083B CN 104730083 B CN104730083 B CN 104730083B CN 201510164949 A CN201510164949 A CN 201510164949A CN 104730083 B CN104730083 B CN 104730083B
- Authority
- CN
- China
- Prior art keywords
- image
- point
- joint
- tap points
- connector
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of steel cable core conveying belt joint to twitch automatic testing method, comprises the following steps：A, treat detection tabs image and benchmark connector image carries out tap points detection；B, surf feature extractions are carried out to tap points image to be checked, characteristic matching is carried out according to characteristic value, seeks a mapping relations between connector image and benchmark connector image, obtains normalization coefficient；C, carry out twitch analysis again according to normalization coefficient and twitch distance is calculated, the steel cable core conveying belt joint of the present invention twitches automatic testing method, it can guarantee that normalization coefficient calculates reliability, even if therefore belt deflection or velocity variations, as long as tap points matching is correct, it is accurate to ensure that tap points distance calculates according to normalization coefficient, improve steel cable termina and twitch accuracy of detection, solve the problems, such as that conveyer belt inner wire rope socket is twitched and calculate inaccuracy, adapt to conveyer belt velocity variations and translation；Safe operation to ensureing transportation system is of great significance.
Description
Technical field
The present invention relates to a kind of conveying band joint detection method, more particularly to a kind of steel cable core conveying belt joint to twitch certainly
Dynamic detection method.
Background technology
In coal mine conveying belt transport field, the steel rope core conveying belt of long range has many vulcanized joints, and connector is whole
The most weak link of steel rope core conveying belt tension.It is most of from the point of view of steel rope core conveying belt fracture accident in existing coal enterprise
Be due to connector twitch do not found in time and caused by.Therefore, to steel rope core conveying belt connector twitch situation be detected with
And prediction, to avoiding broken belt accident from having important realistic meaning.Domestic multiple universities and company are defeated to radioscopy
Send belt lacing image to carry out twitch analysis, but all do not consider in these technologies in actual shipment environment, conveyer belt is due to delivery
Velocity variations, shake, sideslip can occur for load difference, and the same connector image of detector secondary acquisition will be translated, stretched
The deformation such as contracting.If directly carrying out connector twitch calculating to the connector image of secondary acquisition, very big error undoubtedly occurs.Cause
Appearance is reported by mistake or failed to report, and influences to keep the safety in production, in some instances it may even be possible to because the connector twitched is not handled by timely discovery, is conveyed
Band broken belt accident, causes economic loss.
The content of the invention
In view of this, the object of the present invention is to provide a kind of Steel cord for adapting to conveyer belt velocity variations, translating is defeated
Send belt lacing to twitch automatic testing method, treat detection tabs image and benchmark connector image carries out tap points detection and carried with feature
Take, characteristic matching is carried out according to characteristic value, seeks a connector mapping relations between connector image and benchmark connector image, obtains
Normalization coefficient, carries out twitching analysis being calculated twitch distance again according to normalization coefficient, therefore even if belt deflection or
Velocity variations, as long as tap points matching is correct, it is accurate to ensure that tap points distance calculates according to normalization coefficient.
The steel cable core conveying belt joint of the present invention twitches automatic testing method, comprises the following steps：A, connect to be detected
Head image and benchmark connector image carry out tap points detection；B, surf feature extractions are carried out to tap points image to be checked, according to spy
Value indicative carries out characteristic matching, seeks a mapping relations between connector image and benchmark connector image, obtains normalization coefficient；c、
Carry out twitch analysis again according to normalization coefficient and twitch distance is calculated；
Further, the step a specifically includes following steps：A1, gray level amendment；A2, gaussian filtering；A3, image point
Cut；A4, image filtering eliminate isolated point；A5, Contour extraction landing nipple point；
Further, the step b specifically includes following steps：B1, the description of tap points surf characteristic values；B2, benchmark image
With image to be detected Feature Points Matching；B3, calculate image to be detected joint length；B4, obtain normalization coefficient k；
Further, the step c specifically includes following steps：C1, calculate image to be detected butt joint distance；C2, basis
Butt joint distance and label, try to achieve twitch distance.
The beneficial effects of the invention are as follows：The steel cable core conveying belt joint of the present invention twitches automatic testing method, using connecing
Head points correspondence mode, even if belt deflection or velocity variations, as long as tap points matching is correct, calculates just according to normalization
It is accurate to can ensure that tap points distance calculates, improves steel cable termina and twitches accuracy of detection, solve conveyer belt inner cable
Connector, which is twitched, calculates inaccurate problem, adapts to conveyer belt velocity variations and translation；Safe operation to ensureing transportation system has
It is significant.
Brief description of the drawings
Fig. 1 is connector image schematic diagram；
Fig. 2 isolates dot image schematic diagram to eliminate；
Fig. 3 is connector apart from schematic diagram.
Embodiment
Fig. 1 is connector image schematic diagram, and for Fig. 2 to eliminate isolated dot image schematic diagram, Fig. 3 is connector apart from schematic diagram；Such as
Shown in figure：The steel cable core conveying belt joint of the present embodiment twitches automatic testing method, comprises the following steps：A, connect to be detected
Head image and benchmark connector image carry out tap points detection；B, surf feature extractions are carried out to tap points image to be checked, according to spy
Value indicative carries out characteristic matching, seeks a mapping relations between connector image and benchmark connector image, obtains normalization coefficient；c、
Carry out twitch analysis again according to normalization coefficient and twitch distance is calculated.
In the present embodiment, the step a specifically includes following steps：A1, gray level amendment；A2, gaussian filtering；A3, figure
As segmentation；A4, image filtering eliminate isolated point；A5, Contour extraction landing nipple point；In the step a1, due to detector picture
Member is irradiated strong and weak different, intensity profile inequality by X-ray light, and image both sides gray value is relatively low, it is necessary to which butt joint image carries out ash
Degree is corrected, and is mapped using piecewise linear transform in a certain tonal range, and original image f (x, y) tonal range is [min, max], is become
Change rear image g (x, y) tonal range and extend to [a, b], its greyscale transformation is expressed as：
Then noise is eliminated using Gaussian smoothing filter；In step a2, connector image shown in Figure 1, tap points exist
Adjacent thereto gray value differences of vertical direction are larger, and wherein top connection point gray value is less than neighbor grayscale value, lower contact point gray scale
Value is more than neighbor grayscale value, makees Y-direction difference to each pixel and image segmentation can be achieved, note vertical direction difference is DY=
P (x, y)-P (x, y+1), P (x, y) are a certain pixel gray value, and P (x, y+1) is vertically adjacent pixel gray value,
Thr is a threshold value, order
Gray value is considered top connection point for 128, and gray value is considered lower contact point for 255, and thr is by gray-level correction
Rear joint gradation of image scope and picture contrast determine that, if connector picture contrast is larger, threshold value can be chosen larger；
In practical applications, affected by noise, brightness/gray scale change is usually there are transition portion, if the ash directly with neighborhood point in itself
Angle value calculates, threshold value determine it is more difficult, in order to effectively avoid such case, before applying equation (1) calculating, the values of tap points and
An its neighbor pixel gray scale is averaged, and is averaged using three neighbor pixel gray scales, then carries out Y-direction difference, note
In DY=(P (x-1, y)+P (x, y)+P (x+1, y))/3-P (x, y+1), step a3, after the segmentation of butt joint difference, also exist very
For the smaller noise isolated point of many areas, it is necessary to which these isolated points are excluded, the method for eliminating isolated point is very much, can such as carry out 5 × 5
Median filter process, but tap points area is also smaller sometimes, is taken as isolated point to eliminate, by observing tap points picture
Element distribution, is only filtered processing to isolated point X-direction, makes P (x, y)=0, such as P (x-1, y)=0 and P (x+1, y)=0, ginseng
Dot image is isolated in elimination as shown in Figure 2；Finally by all tap points up and down in Contour extraction target area, connector is calculated
The center-of-mass coordinate of point.
In the present embodiment, the step b specifically includes following steps：B1, the description of tap points surf characteristic values；B2, benchmark
Image and image to be detected Feature Points Matching；B3, calculate image to be detected joint length；B4, obtain normalization coefficient k；Step
In b1, using the tap points extracted as key feature points, the distribution of the response of single order Haar small echos in the x and y direction is established
(local message integration), algorithm steps are as follows：
1. 24 × 24 regions are as interest region around selected characteristic point, and subregion of this region segmentation into 4 × 4,
In order to retain some spatial informations, subregion can overlap more, and subregion is 9 × 9 sizes；
2. in each sub-regions, Haar small echos response dx, dy are calculated in 9 × 9 dot matrix, and carry out Gauss weighting；
3. respectively to dx, the dy of subregion, | dx |, | dy | response summation and normalized vector.Obtain feature description [∑
Dx, ∑ dy, ∑ | dx |, ∑ | dy |], so each characteristic point is exactly 64 dimensional vectors；
In step b2, the Euclidean distance of all joint characteristics points of calculating benchmark image and all tap points of image to be detectedWherein, (x1,x2…x64) on the basis of image some joint characteristics point
Feature vector, (x1′,x2′…x64') feature vector of image to be detected some joint characteristics point；Calculate its arest neighbors and time
Neighbour's Euclidean distance ratio carrys out match query point, when ratio is less than some threshold value, then it is assumed that 2 points of matchings；
In step b3, after the matching of b1 steps, the matching point set of image to be detected, [(x are obtained1,y1),(x2,
y2)…(xn,yn)], the connector image after observation segmentation, sets space constraints, it is easy to obtain connector the top tap points
With bottom tap points, the top a line white point in the top tap points, that is, Fig. 2, they are all contained in matching point set [(x1,
y1),(x2,y2)…(xn,yn)] in, make the top connection point be
Lower contact point K is top connection point number, and l connects under being
Head point number；Average multiple joint characteristics point coordinates seek joint length, and joint length to be detected is expressed as：Lc=((yb1+yb2+…
+ybl)/l-(yt1+yt2+…+ytk)/k) × PixWidth, wherein PixWidth be pel spacing, it can also similarly try to achieve benchmark
The order of image joint length is Lr, finally, normalization coefficient k=L is calculated in step b4c/r。
In the present embodiment, the step c specifically includes following steps：C1, calculate image to be detected butt joint distance；c2、
According to butt joint distance and label, twitch distance is tried to achieve；In wherein step c1, top connection is obtained by tool joint monitor and is denoted as U (xi,
yi), lower contact is denoted as D (xj,yj), wherein xi,yiRepresent the coordinate of some connector；
If meet following space length constraints, then it is assumed that corresponding 2 tap points up and down are a butt joint；|xi-
xj| ＜ σx (2)
|yi-yj| ＜ σy (3)
σxAnd σyFor level thresholds and vertical threshold, determined by conveyer belt specification in concrete application, usual σxSlightly larger than two
Average distance between bar Steel cord, σySlightly larger than the standard vertical distance between upper and lower tap points；After obtaining matching butt joint, meter
The distance between connector is calculated, is expressed as：Wherein PixWidth is pel spacing,
Determined by detector precision；
Described according to the feature of the two tap points, dock first unique label to this, so ensure that benchmark
All butt joints are corresponded with the butt joint in image to be detected in reference picture；In step c2, difference calculating benchmark image
The distance of label butt joint identical with image to be detected, referring to Fig. 3, is denoted as R respectivelyiAnd Ci, wherein (i=0,1 ... N), N is pair
Connector number, i number for butt joint, then distance L is twitched in butt jointi=Ci×k-Ri, average distance of twitching is Li/N。
Finally illustrate, the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted, although with reference to compared with
The present invention is described in detail in good embodiment, it will be understood by those of ordinary skill in the art that, can be to the skill of the present invention
Art scheme technical scheme is modified or replaced equivalently, without departing from the objective and scope of technical solution of the present invention, it should all cover at this
Among the right of invention.
Claims (2)
1. a kind of steel cable core conveying belt joint twitches automatic testing method, it is characterised in that：Comprise the following steps：A, to be checked
Survey connector image and benchmark connector image carries out tap points detection；B, surf feature extractions, root are carried out to tap points image to be checked
Characteristic matching is carried out according to characteristic value, seeks a mapping relations between connector image and benchmark connector image, obtains normalization system
Number；C, carry out twitch analysis again according to normalization coefficient and twitch distance is calculated；
The step a specifically includes following steps：A1, gray level amendment；A2, gaussian filtering；A3, image segmentation；A4, image filter
Ripple eliminates isolated point；A5, Contour extraction landing nipple point；In step a2, noise is eliminated using Gaussian smoothing filter, tap points
Value and its a neighbor pixel gray scale be averaged, be averaged using three neighbor pixel gray scales, then to carry out Y-direction poor
Point, note DY=(P (x-1, y)+P (x, y)+P (x+1, y))/3-P (x, y+1) is a certain pixel gray value, and P (x, y+1) is vertical
For Nogata to neighbor pixel gray value, thr is a threshold value, order
The step b specifically includes following steps：B1, the description of tap points surf characteristic values；B2, benchmark image and image to be detected
Feature Points Matching；B3, calculate image to be detected joint length；B4, obtain normalization coefficient k；
In step b1, using the tap points extracted as key feature points, the sound of single order Haar small echos in the x and y direction is established
The distribution answered, algorithm steps are as follows：
1. 24 × 24 regions are as interest region around selected characteristic point, and this region segmentation into 4 × 4 subregion, sub-district
Domain is 9 × 9 sizes；
2. in each sub-regions, Haar small echos response dx, dy are calculated in 9 × 9 dot matrix, and carry out Gauss weighting；
3. respectively to dx, the dy of subregion, | dx |, | dy | response summation and normalized vector；Obtain feature description [∑ dx, ∑
dy,∑|dx|,∑|dy|]；
In step b2, the Euclidean distance of all joint characteristics points of calculating benchmark image and all tap points of image to be detectedWherein, (x1,x2…x64) on the basis of image some joint characteristics point
Feature vector, (x1′,x2′…x64') feature vector of image to be detected some joint characteristics point；Calculate its arest neighbors and time
Neighbour's Euclidean distance ratio carrys out match query point, when ratio is less than threshold value, then it is assumed that 2 points of matchings；
In step b3, after the matching of b1 steps, the matching point set of image to be detected, [(x are obtained1,y1),(x2,y2)…
(xn,yn)], the connector image after observation segmentation, sets space constraints, obtains connector the top tap points and bottom connects
Head point, makes the top connection point beLower contact point K is top connection point number, and l is lower contact point number；
Average multiple joint characteristics point coordinates seek joint length, and joint length to be detected is expressed as：Lc=((yb1+yb2+…+ybl)/l-
(yt1+yt2+…+ytk)/k)) × PixWidth, wherein PixWidth be pel spacing, it can also similarly try to achieve benchmark image and connect
Head length, makes as Lr；
Normalization coefficient k=L is calculated in step b4c/Lr。
2. steel cable core conveying belt joint according to claim 1 twitches automatic testing method, it is characterised in that：The step
Rapid c specifically includes following steps：C1, calculate image to be detected butt joint distance；C2, according to butt joint distance and label, try to achieve
Twitch distance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510164949.XA CN104730083B (en) | 2015-04-09 | 2015-04-09 | Steel cable core conveying belt joint twitches automatic testing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510164949.XA CN104730083B (en) | 2015-04-09 | 2015-04-09 | Steel cable core conveying belt joint twitches automatic testing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104730083A CN104730083A (en) | 2015-06-24 |
| CN104730083B true CN104730083B (en) | 2018-05-01 |
Family
ID=53454181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510164949.XA Active CN104730083B (en) | 2015-04-09 | 2015-04-09 | Steel cable core conveying belt joint twitches automatic testing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104730083B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106296700B (en) * | 2016-08-15 | 2019-02-15 | 南京工程学院 | A kind of steel cord conveyor belt connector twitch detection method |
| CN106706238B (en) * | 2016-11-22 | 2019-01-29 | 山西大学 | Steel cable core conveying belt joint overlap joint label and recognition methods |
| CN107423744A (en) * | 2017-03-23 | 2017-12-01 | 北京环境特性研究所 | The Seam tracking and damage positioning method of steel rope core conveying belt |
| CN107728223A (en) * | 2017-09-13 | 2018-02-23 | 太原理工大学 | A kind of steel rope core conveying belt joint twitches online test method |
| CN108550135B (en) * | 2018-03-07 | 2021-07-30 | 天津工业大学 | Automatic detection method for elongation of steel wire rope core conveyor belt joint based on X-ray image |
| CN110288562B (en) * | 2019-05-16 | 2023-01-17 | 枣庄学院 | Method for detecting joint twitching of steel wire rope core conveying belt based on X-ray image |
| CN114235825B (en) * | 2022-02-24 | 2022-05-17 | 武汉祥文钢材制品有限公司 | Steel wire rope quality detection method based on computer vision |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679703A (en) * | 2013-11-25 | 2014-03-26 | 河海大学 | Hyperspectral remote sensing image dimensionality reduction method based on conformal geometric algebra |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3214364B2 (en) * | 1996-08-14 | 2001-10-02 | 富士電機株式会社 | Inter-vehicle distance measuring device |
| KR100932210B1 (en) * | 2007-10-12 | 2009-12-16 | 광주과학기술원 | Method and apparatus for image feature extraction, method and apparatus for content-based image retrieval using the same, and recording medium recording program for performing the methods |
| JP4752918B2 (en) * | 2009-01-16 | 2011-08-17 | カシオ計算機株式会社 | Image processing apparatus, image collation method, and program |
-
2015
- 2015-04-09 CN CN201510164949.XA patent/CN104730083B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679703A (en) * | 2013-11-25 | 2014-03-26 | 河海大学 | Hyperspectral remote sensing image dimensionality reduction method based on conformal geometric algebra |
Non-Patent Citations (1)
| Title |
|---|
| 强力输送带接头抽动检测算法研究;樊荣 等;《工矿自动化》;20140228;第40卷(第2期);第31页第1-2节，第32页第3节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104730083A (en) | 2015-06-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104730083B (en) | Steel cable core conveying belt joint twitches automatic testing method | |
| CN111079747B (en) | Railway wagon bogie side frame fracture fault image identification method | |
| CN101770582B (en) | Image matching system and method | |
| US9558403B2 (en) | Chemical structure recognition tool | |
| CN107169489B (en) | Method and apparatus for tilt image correction | |
| CN106296700B (en) | A kind of steel cord conveyor belt connector twitch detection method | |
| CN108369650A (en) | The method that candidate point in the image of calibrating pattern is identified as to the possibility characteristic point of the calibrating pattern | |
| CN109060842B (en) | Citrus surface defect detection method based on surface fitting correction | |
| CN107169933B (en) | Edge reflection pixel correction method based on TOF depth camera | |
| CN115375588B (en) | Power grid transformer fault identification method based on infrared imaging | |
| CN116805317A (en) | Rotary furnace inner wall defect detection method based on artificial intelligence | |
| CN114331986A (en) | Dam crack identification and measurement method based on unmanned aerial vehicle vision | |
| CN109447036A (en) | A kind of segmentation of image digitization and recognition methods and system | |
| CN105283750B (en) | Method for handling the digital picture of surface of tyre to detect abnormal | |
| CN110889874B (en) | Error evaluation method for binocular camera calibration result | |
| CN105891231A (en) | Carrot surface defect detection method based on image processing | |
| CN104112123A (en) | Defect characteristic extraction and identification method of AOI system used for bullet apparent defect detection | |
| CN105205829B (en) | Substation's Infrared Image Segmentation based on improved two dimension Otsu | |
| CN114359251A (en) | Automatic identification method for concrete surface damage | |
| CN111160339B (en) | License plate correction method, image processing equipment and device with storage function | |
| US20210090260A1 (en) | Deposit detection device and deposit detection method | |
| CN108205641B (en) | Gesture image processing method and device | |
| CN105740796B (en) | Lane line image binaryzation method after a kind of perspective transform based on grey level histogram | |
| CN110956616A (en) | Target detection method and system based on stereoscopic vision | |
| CN114663433B (en) | Method and device for detecting running state of roller cage shoe, computer equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |