CN113343834A - Belt deviation diagnosis method based on machine vision and laser line - Google Patents
Belt deviation diagnosis method based on machine vision and laser line Download PDFInfo
- Publication number
- CN113343834A CN113343834A CN202110614892.4A CN202110614892A CN113343834A CN 113343834 A CN113343834 A CN 113343834A CN 202110614892 A CN202110614892 A CN 202110614892A CN 113343834 A CN113343834 A CN 113343834A
- Authority
- CN
- China
- Prior art keywords
- belt
- laser line
- picture
- laser
- area
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000003745 diagnosis Methods 0.000 title claims abstract description 10
- 238000012706 support-vector machine Methods 0.000 claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000000605 extraction Methods 0.000 claims abstract description 9
- 238000003709 image segmentation Methods 0.000 claims abstract description 9
- 238000006073 displacement reaction Methods 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 7
- 239000000463 material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The belt deviation diagnosis method based on machine vision and laser lines comprises the following steps: installing a high-energy laser generator above the belt, wherein the color of a laser ray emitted by the laser generator is selected based on the criterion of large color contrast with the belt and objects transported on the belt; installing a high-definition camera in front of or behind the laser line above the belt, so that the laser line and the belt are positioned in the center of a picture shot by the high-definition camera; the method comprises the steps of utilizing a high-definition camera to capture a scene picture of belt operation in real time, applying algorithms such as image segmentation and characteristic extraction to the captured picture, training a characteristic sample through a support vector machine, completing training and learning of a belt motion scene, detecting the moving displacement of a laser line in an image in real time, namely the moving displacement of the belt, and judging whether the belt deviates or not and the direction of the deviation. The belt of the invention does not need to be specially designed, has strong universality, furthest protects the service life of the belt and has high detection precision.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a belt deviation diagnosis method based on machine vision and laser lines.
Background
The groove-type belt conveyor is widely applied to industries such as power stations, coal mines, ports and the like, the phenomenon of belt deviation is a common fault in the operation process, and if blanking points are not centered or the load is unbalanced in the conveying process, the phenomenon of irregular deviation of the belt of the conveyor can be caused. When the offset exceeds each technical index, the conveyor equipment is likely to have belt derailment, which causes the material in the conveying system to be poured in a large amount, and the belt with high value is thoroughly damaged, thereby further causing the fault shutdown of the whole material conveying system and the shutdown of the device.
At present, belt deviation detection is mainly completed by matching large control station units, the system is complex and high in cost, and a traditional deviation switch detection system cannot judge the deviation state in real time, so that the belt cannot be timely corrected and is pulled, and the economic loss is large. CN105197537B discloses a belt off tracking detecting system and method based on color detection, which adopts the mode of positioning a color ribbon, a color detection sensor and a programmable controller to detect belt off tracking, but in the detecting system, the positioning color ribbon is printed on a conveying belt, no corresponding finished product can not be applied in a large range in the actual belt conveying process, and even if the belt is printed by adopting a special manufacturing process and the positioning color ribbon is not clear after long-term abrasion, the detecting can not be detected or the detecting result is not accurate.
Disclosure of Invention
The invention aims to solve the technical problem that the belt deviation diagnosis method based on machine vision and laser lines aims at the defects of the existing belt deviation detection, realizes real-time belt deviation detection by combining image depth learning and laser lines, extracts the laser lines on the belt in pictures, judges whether the belt deviates or not according to the position movement of the laser lines, can further analyze the belt deflection direction to diagnose whether the belt deviates or not, protects the service life of the belt to the maximum extent, and has high detection precision.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the belt deviation diagnosis method based on machine vision and laser lines comprises the following steps:
s1, installing a high-energy laser generator above the belt, wherein the color of the laser generator is selected based on the principle that the color contrast with the belt and the objects transported on the belt is relatively large;
s2, installing a high-definition camera above the belt in front of or behind the laser line, so that the laser line and the belt are positioned in the center of a picture shot by the high-definition camera;
s3, a scene picture of belt operation is captured in real time by the aid of a high-definition camera through the laser transmitter additionally arranged above the belt, the captured picture is trained through a support vector machine by means of algorithms such as image segmentation and characteristic extraction, and accordingly training and learning of a belt motion scene are completed, movement displacement of a laser line in an image, namely movement displacement of the belt, is detected in real time, and whether the belt deviates or not and the deviation direction are judged.
According to the above scheme, the step S3 specifically includes:
s31, collecting samples
Collecting high-definition scene pictures of belt operation under different load conditions according to a belt motion scene; the scene pictures are divided into two categories, one category is a high-definition picture of which the position relation of the left side of the laser line, the belt and the right side of the laser line is within a deviation threshold range under the condition that the belt is under different loads (including no load), and an effective label is attached to the high-definition picture; the other type is to collect pictures of the belt under different load conditions (including no load), wherein the running direction of the belt is deviated to the left side of the laser line or the running direction is deviated to the right side of the laser line, and an invalid label is attached;
s32, image segmentation
Based on Mean Shift algorithm (Mean Shift algorithm, a feature space clustering algorithm), a captured scene picture is segmented, and three blocks are divided, namely a laser line left side area, a belt area and a laser line right side area;
s33, feature extraction
Extracting the shape characteristics of the image and retrieving the outline by adopting a paired geometric histogram, and taking a plurality of homogeneous regions as the same connecting region by a histogram-based method;
extracting the image shape characteristics of the left area, the belt area and the right area of the laser line, and defining the following variables as characteristic samples:
Sidebdl: a laser line left boundary;
Sidebdr: the right border of the right region of the laser line;
Sidebel: a belt left boundary;
Sideber: a belt right boundary;
XAL-BEL: left Side left boundary Side of laser linebdlSide to belt left boundarybelThe pixel point of (2);
XBEL-BER: left Side edge of beltbelSide to right Side of beltberThe pixel point of (2);
XBR-BER: belt right border SideberSide of right boundary with right Side of laser linebdrThe pixel point of (2);
PA-BELT: the pixel ratio of the left area of the laser line to the belt area is calculated according to the following formula: pA-BELT＝XAL-BEL/XBEL-BER；
PB-BELT: the pixel ratio of the right side area of the laser line to the belt area is calculated according to the following formula: pB-BELT＝XBR-BER/XBEL-BER；
Extracting characteristic values of the left side of the laser line, the color area of the belt and the right side of the laser line, and determining whether the contour pixel ratio of the left side of each laser line and the color area of the belt is normal or not:
in the case of normal belt operation, PA-BELTAnd PB-BELTThe change in value of (A) is dependent only on the belt load, P being relatively constantA-BELTAnd PB-BELTThe change in value of (a) is relatively stable;
if the belt is deflected during belt operation, PA-BELTAnd PB-BELTThe value of (A) is obviously changed, and the belt is judged to be abnormally operated;
s34 support vector machine training
Before a support vector machine is used for training, the characteristic samples (scene pictures of belt motion) collected and processed in front are divided into two types, wherein one type is an effective picture, namely a picture with a normal belt running track, and the two types comprise a load condition and an idle condition; the other type is an invalid picture, namely the picture that the running track of the belt is abnormal also comprises two situations of belt loading and no-load; and then, respectively taking out one part of the two types of pictures to be used for training the support vector machine model, and taking the other part of the two types of pictures to be used as a test set, wherein the test set also comprises valid pictures and invalid pictures so as to verify the classification effect of the support vector machine model after training.
According to the above scheme, in step S34, according to the characteristics of the scene picture, the performance of the training algorithm parameters of the support vector machine is optimized, and it can be known from the analysis of the support vector machine principle that the penalty factor C and the kernel parameter r in the support vector machine are key factors affecting the performance of the support vector machine, the penalty factor C is used to determine the confidence interval range for adjusting the learning machine in the sample data subspace, the optimal C in different data subspaces is different, and the change of the kernel parameter r actually changes the mapping function implicitly, so as to change the complexity of the sample data subspace distribution, i.e., the maximum VC dimension of the linear classification, and thus determine the minimum error reached by the linear classification.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the invention, the high-energy laser generator is arranged above the belt, and after a camera above the belt is combined to capture a picture of belt motion in real time, a laser line on the belt in the picture is extracted, and whether the belt deviates or not is judged according to the position movement of the laser line; the real-time detection of the belt deviation is realized, the belt itself does not need to be specially designed, whether the belt deviates or not can be diagnosed in real time, and the service life of the belt is protected to the greatest extent;
2. the belt deviation direction can be known at a glance according to the pixel ratio of the left side of the laser line, the right side of the laser line and the belt area in the characteristic sample data, and the detection precision is high.
Drawings
FIG. 1 is a schematic diagram of an embodiment of a belt deviation diagnosis method based on machine vision and laser lines;
FIG. 2 is a flow chart of a belt deviation diagnosis method based on machine vision and laser lines according to the invention;
FIG. 3 is a schematic view of a belt area of a belt shot by a high definition camera according to an embodiment of the present invention during normal operation;
FIG. 4 is a schematic diagram of the structure of the ratio of the left side of the laser line to the pixel of the belt area when the belt is in normal operation according to the embodiment of the present invention;
FIG. 5 is a schematic diagram of the structure of the ratio of the right side of the laser line to the pixel of the belt area when the belt is in normal operation according to the embodiment of the present invention;
fig. 6 is a schematic diagram of a belt area when a belt shot by a high definition camera according to an embodiment of the present invention deviates;
FIG. 7 is a schematic diagram illustrating the influence of fixed parameters r and C as variables on SVM performance in SVM training according to an embodiment of the present invention;
FIG. 8 is a diagram illustrating the influence of a fixed parameter C, r as a variable on SVM performance in SVM training according to an embodiment of the present invention;
FIG. 9 is a diagram illustrating the influence of the equivalence of the parameters r and C on the SVM performance in the SVM training of the present invention.
Detailed Description
The technical solution of the present invention is described in detail below with reference to the accompanying drawings and examples.
The invention relates to a belt deviation diagnosis method based on machine vision and laser lines, which comprises the following steps:
s1, referring to the figure 1, firstly, installing a high-energy laser generator above the belt, wherein the color of the laser generator is selected based on the principle of large color contrast (obvious visual difference) with the belt and the objects transported on the belt;
s2, secondly, installing a high-definition camera above the belt and in front of or behind the laser line, so that the laser line and the belt are positioned in the center of a picture shot by the high-definition camera; when the light is dim or works at night, a light supplement lamp is arranged beside the high-definition camera;
s3, through the laser emitter that increases above the belt, utilize high definition digtal camera to snapshot the scene picture of belt operation in real time, to the picture of taking a candid photograph, the application image is cut apart, algorithms such as characterized by extraction, and train the characteristic sample through the support vector machine, mark, accomplish the training study of belt motion scene, thereby detect out the displacement of laser line in the image in real time, also be exactly the displacement of belt, and judge whether the belt off tracking, and the direction of off tracking, specifically as shown in FIG. 2, include:
s31, collecting samples
According to the motion scene of the belt, collecting high-definition scene pictures of the belt running under different load conditions, and clearly distinguishing a left line belt and a belt area of a laser line through pictures shot by a high-definition camera as shown in figure 3; scene pictures are divided into two categories, one is a high-definition picture in which the position relationship among the left side of a laser line, the belt and the right side of the laser line is within a deviation threshold range under the condition that a belt is under different loads (including no load), as shown in figure 3, and an effective label is attached; the other is to collect pictures of the belt under different load conditions (including no load), wherein the running direction of the belt is deviated to the left side of the laser line or the running direction is deviated to the right side of the laser line, as shown in figure 6, and an invalid label is attached;
s32, image segmentation
Firstly, a scene picture is segmented, three blocks are divided, namely a laser line left side area, a belt area and a laser line right side area, a snap shot scene picture is segmented mainly by means of a Mean Shift algorithm, namely a Mean Shift algorithm, the image segmentation algorithm based on Mean Shift is a feature space clustering algorithm, and the algorithm steps are as follows:
assume that image data of a scene picture is { x }ij1,2, n, j 1,2yAn iterative process of obtaining a Mean Shift algorithm in image segmentation is as follows, wherein k is 1,2.
Wherein g (x) is a negative derivative of a nuclear section function of an image feature space, and if M is the minimum number of pixels in an independent area, image segmentation of the Mean Shift algorithm is described as follows:
the first step is as follows: selecting an Epanechov kernel as a kernel function;
the second step is that: for each point on the image, the convergence point is calculated and is noted as zij；
The third step: in the data set zijPerforming feature clustering on the coordinate space with the Euclidean distance smaller than h, wherein i is 1,2, and n, j is 1,2cAnd the color space distance is less than hLThe data points of (a) are grouped into one class and all feature classes are taken as: l cp|p＝1，2...q；
The fourth step:
1,2, m, class label of image data space lij＝{p|xij∈cp}；
The fifth step: eliminating classes with the number of elements smaller than M;
s33, characteristic extraction
Extracting the shape characteristics of the image and retrieving the outline by adopting a paired geometric histogram, and taking a plurality of homogeneous regions as the same connecting region by a histogram-based method;
extracting the image shape characteristics of the left area, the belt area and the right area of the laser line, and defining the following variables as characteristic samples:
Sidebdl: a laser line left boundary;
Sidebdr: the right edge of the right side of the laser line;
Sidebel: a belt left boundary;
Sideber: a belt right boundary;
XAL-BEL: left Side left boundary Side of laser linebdlSide to belt left boundarybelThe pixel point of (2);
XBEL-BER: left Side edge of beltbelSide to right Side of beltberThe pixel point of (2);
XBR-BER: belt right border SideberSide of right boundary with right Side of laser linebarThe pixel point of (2);
PA-BELT: the pixel ratio of the left area of the laser line to the belt area is calculated according to the following formula: pA-BELT＝XAL-BEL/XBEL-BER；
PB-BELT: the pixel ratio of the right side area of the laser line to the belt area is calculated according to the following formula: pB-BELT＝XBR-BER/XBEL-BER；
Through the characteristic extraction algorithm, the characteristic values on the left side of the laser line A, the belt color area and the right side of the laser line B are extracted, so that whether the contour pixel ratio of the left side of each laser line A and the belt color area is normal or not is confirmed:
in the case of normal belt operation, PA-BELTAnd PB-BELTThe change in value of (A) is dependent only on the belt load, P being relatively constantA-BELTAnd PB-BELTThe change of the value of (c) is relatively stable, as shown in fig. 4 and 5.
If the belt is deflected during belt operation, PA-BELTAnd PB-BELTThe value of (a) is significantly changed, and as shown in fig. 6, the conveyor belt is shifted toward the laser line B side.
S34 support vector machine training
Before a support vector machine is used for training, the characteristic samples (scene pictures of belt motion) collected and processed in front are divided into two types, wherein one type is an effective picture, namely a picture with a normal belt running track, and the two types comprise a load condition and an idle condition; the other type is an invalid picture, namely the picture that the running track of the belt is abnormal also comprises two situations of belt loading and no-load; and then, respectively taking out one part of the two types of pictures to be used for training the support vector machine model, and taking the other part of the two types of pictures to be used as a test set, wherein the test set also comprises valid pictures and invalid pictures so as to verify the classification effect of the support vector machine model after training.
According to the characteristics of the scene, the performance of the parameters of the training algorithm of the support vector machine is optimized, and the punishment factors C and r in the support vector machine are key factors influencing the performance of the SVM which can be known from the analysis of the support vector machine principle. The function of the parameter C is to determine the confidence interval range of the adjustment learning machine in the data subspace, the optimal C in different data subspaces is different, and the change of the kernel parameter r actually implicitly changes the mapping function so as to change the complexity of the sample data subspace distribution, i.e. the maximum VC dimension of the linear classification, and thus the minimum error reached by the linear classification is determined. Fig. 7 and 8 are performance curves predicted by a model trained according to a certain factor of C and r, respectively, and reflect the influence of parameters C and r on the performance of the SVM.
Fig. 7 shows the influence of C on the performance of the SVM after the model is trained with the fixed kernel parameters r and C as variables. It can be seen that when C is larger and larger, after the C is more than 12, the model reaches the optimum from the identification rate, the recall rate and the comprehensive evaluation score, and as C is increased further, the curve is stable and even unchanged.
Fig. 8 shows a fixed penalty factor C, a kernel parameter r is used as a variable, and after a model is trained, r influences the performance of the SVM, and we can see that the performance of the SVM model reaches the optimum when the value of r gradually rises in an interval of approximately 7 to 10, and then as the value of r increases again, the performance of the SVM decreases, and finally, the performance tends to be stable after 20.
From fig. 9, it can be seen that under the condition that C and r are equivalent, the performance of the SVM gradually increases and reaches the optimum in the interval from 14 to 22, so the value range of SVM parameters C and r is further narrowed, reference can be made to the final value, and it is extremely important to optimize the image processing effect and the feature extraction while obtaining the optimum C and r for training.
It should be understood that the above examples are only for clearly illustrating the present invention and are not intended to limit the embodiments of the present invention. Obvious variations or modifications to the present invention, which would be apparent to a person of ordinary skill in the art, would nevertheless fall within the scope of the present invention.
Claims (3)
1. A belt deviation diagnosis method based on machine vision and laser lines is characterized by comprising the following steps:
s1, installing a high-energy laser generator above the belt, wherein the color of the laser generator is selected based on the principle that the color contrast with the belt and the objects transported on the belt is relatively large;
s2, installing a high-definition camera above the belt in front of or behind the laser line, so that the laser line and the belt are positioned in the center of a picture shot by the high-definition camera;
s3, a scene picture of belt operation is captured in real time by the aid of a high-definition camera through the laser transmitter additionally arranged above the belt, the captured picture is trained through a support vector machine by means of algorithms such as image segmentation and characteristic extraction, and accordingly training and learning of a belt motion scene are completed, movement displacement of a laser line in an image, namely movement displacement of the belt, is detected in real time, and whether the belt deviates or not and the deviation direction are judged.
2. The method for diagnosing belt deviation based on machine vision and laser line as claimed in claim 1, wherein said step S3 specifically includes:
s31, collecting samples
Collecting high-definition scene pictures of belt operation under different load conditions according to a belt motion scene; the scene pictures are divided into two categories, one category is a high-definition picture with the position relation of the left side of the laser line, the belt and the right side of the laser line within a deviation threshold range under different load conditions of the belt, and an effective label is attached to the high-definition picture; the other type is to collect pictures of which the running direction of the belt deviates to the left side of the laser line or the running direction deviates to the right side of the laser line under different loading conditions, and to paste an invalid label;
s32, image segmentation
Segmenting a captured scene picture based on a Mean Shift algorithm to divide three blocks, namely a laser line left side area, a belt area and a laser line right side area;
s33, feature extraction
Extracting the shape characteristics of the image and retrieving the outline by adopting a paired geometric histogram, and taking a plurality of homogeneous regions as the same connecting region by a histogram-based method;
extracting the image shape characteristics of the left area, the belt area and the right area of the laser line, and defining the following variables as characteristic samples:
Sidebdl: a laser line left boundary;
Sidebdr: the right edge of the right side of the laser line;
Sidebel: a belt left boundary;
Sideber: a belt right boundary;
XAL-BEL: left Side left boundary Side of laser linebdlSide to belt left boundarybelThe pixel point of (2);
XBEL-BER: left Side edge of beltbelSide to right Side of beltberThe pixel point of (2);
XBR-BER: belt right border SideberSide of right boundary with right Side of laser linebdrThe pixel point of (2);
PA-BELT: the pixel ratio of the left area of the laser line to the belt area is calculated according to the following formula: pA-BELT＝XAL-BEL/XBEL-BER；
PB-BELT: the pixel ratio of the right side area of the laser line to the belt area is calculated as follows：PB-BELT＝XBR-BER/XBEL-BER；
Extracting characteristic values of the left side of the laser line, the color area of the belt and the right side of the laser line, and determining whether the contour pixel ratio of the left side of each laser line and the color area of the belt is normal or not:
in the case of normal belt operation, PA-BELTAnd PB-BELTThe change in value of (A) is dependent only on the belt load, P being relatively constantA-BELTAnd PB-BELTThe change in value of (a) is relatively stable;
if the belt is deflected during belt operation, PA-BELTAnd PB-BELTThe value of (A) is obviously changed, and the belt is judged to be abnormally operated;
s34 support vector machine training
Before a support vector machine is used for training, the characteristic samples collected and processed in front are divided into two types, wherein one type is an effective picture, namely a picture with a normal belt running track, and the two types comprise a load condition and an idle condition; the other type is an invalid picture, namely the picture that the running track of the belt is abnormal also comprises two situations of belt loading and no-load; and then, respectively taking out one part of the two pictures to be used for training the support vector machine model, and then taking the other part of the two pictures to be used as a test set, wherein the test set also comprises valid pictures and invalid pictures so as to verify the classification effect of the support vector machine model after being trained.
3. The method as claimed in claim 2, wherein in step S34, based on the characteristics of the scene picture, the performance of the parameters of the training algorithm of the support vector machine is optimized, and from the analysis of the principle of the support vector machine, the penalty factor C and the kernel parameter r in the support vector machine are key factors affecting the performance of the support vector machine, the penalty factor C is used to determine the confidence interval range of the learning machine in the sample data subspace, where the optimal C is different, and the change of the kernel parameter r actually implicitly changes the mapping function to change the complexity of the sample data subspace distribution, i.e. the maximum VC dimension of the linear classification, i.e. the minimum error reached by the linear classification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110614892.4A CN113343834A (en) | 2021-06-02 | 2021-06-02 | Belt deviation diagnosis method based on machine vision and laser line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110614892.4A CN113343834A (en) | 2021-06-02 | 2021-06-02 | Belt deviation diagnosis method based on machine vision and laser line |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113343834A true CN113343834A (en) | 2021-09-03 |
Family
ID=77474782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110614892.4A Pending CN113343834A (en) | 2021-06-02 | 2021-06-02 | Belt deviation diagnosis method based on machine vision and laser line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113343834A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114084613A (en) * | 2021-11-18 | 2022-02-25 | 北京华能新锐控制技术有限公司 | Coal conveying belt deviation detecting system |
| 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 |
| CN116573366A (en) * | 2023-07-07 | 2023-08-11 | 江西小马机器人有限公司 | Belt deviation detection method, system, equipment and storage medium based on vision |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104316593A (en) * | 2014-10-30 | 2015-01-28 | 徐州工程学院 | Real-time online non-destructive steel rope core conveyor belt testing device |
| CN108584352A (en) * | 2018-06-15 | 2018-09-28 | 湖北凯瑞知行智能装备有限公司 | Rubber belt from deviating diagnostic system based on machine vision and method |
| CN110902315A (en) * | 2019-12-10 | 2020-03-24 | 浙江蓝卓工业互联网信息技术有限公司 | Belt deviation state detection method and system |
| CN112193763A (en) * | 2020-10-29 | 2021-01-08 | 河南理工大学 | Belt deviation detection system of belt conveyor |
-
2021
- 2021-06-02 CN CN202110614892.4A patent/CN113343834A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104316593A (en) * | 2014-10-30 | 2015-01-28 | 徐州工程学院 | Real-time online non-destructive steel rope core conveyor belt testing device |
| CN108584352A (en) * | 2018-06-15 | 2018-09-28 | 湖北凯瑞知行智能装备有限公司 | Rubber belt from deviating diagnostic system based on machine vision and method |
| CN110902315A (en) * | 2019-12-10 | 2020-03-24 | 浙江蓝卓工业互联网信息技术有限公司 | Belt deviation state detection method and system |
| CN112193763A (en) * | 2020-10-29 | 2021-01-08 | 河南理工大学 | Belt deviation detection system of belt conveyor |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114084613A (en) * | 2021-11-18 | 2022-02-25 | 北京华能新锐控制技术有限公司 | Coal conveying belt deviation detecting system |
| CN114084613B (en) * | 2021-11-18 | 2023-11-07 | 北京华能新锐控制技术有限公司 | Coal conveying belt deviation detecting system |
| 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 |
| CN115116010B (en) * | 2022-08-29 | 2022-11-18 | 山东千颐科技有限公司 | Belt deviation-preventing visual identification system based on image processing |
| CN116573366A (en) * | 2023-07-07 | 2023-08-11 | 江西小马机器人有限公司 | Belt deviation detection method, system, equipment and storage medium based on vision |
| CN116573366B (en) * | 2023-07-07 | 2023-11-21 | 江西小马机器人有限公司 | Belt deviation detection method, system, equipment and storage medium based on vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108584352B (en) | Machine vision-based adhesive tape deviation diagnosis system and method | |
| CN113343834A (en) | Belt deviation diagnosis method based on machine vision and laser line | |
| CN105067638B (en) | Tire fetal membrane face character defect inspection method based on machine vision | |
| EP1808018B1 (en) | Image sensor annotation method and apparatus | |
| CN108765373A (en) | A kind of insulator exception automatic testing method based on integrated classifier on-line study | |
| CN113724258B (en) | Conveyor belt tearing detection method and system based on image processing | |
| CN111080691A (en) | Infrared hot spot detection method and device for photovoltaic module | |
| CN115018850B (en) | Method for detecting burrs of punched hole of precise electronic part based on image processing | |
| CN111354026B (en) | Improved train image accurate registration method and system based on SURF (speeded up robust features) | |
| CN105044122A (en) | Copper part surface defect visual inspection system and inspection method based on semi-supervised learning model | |
| JP5168215B2 (en) | Appearance inspection device | |
| CN113724231A (en) | Industrial defect detection method based on semantic segmentation and target detection fusion model | |
| CN113139572B (en) | Image-based train air spring fault detection method | |
| CN115144399B (en) | Assembly quality detection method and device based on machine vision | |
| CN114359253A (en) | Image pollution detection method and system based on convolutional neural network | |
| CN113191394A (en) | Machine vision-based conveyor belt deviation diagnosis method and system | |
| CN111398294B (en) | Straw particle defect detection system and detection method based on machine vision | |
| CN116330319B (en) | State monitoring system and method for carrying manipulator | |
| CN117094609A (en) | Intelligent management system for aluminum profile production quality based on machine vision | |
| CN115352832A (en) | Belt tearing detection method | |
| CN112907524B (en) | Method for detecting fault of fire-proof plate of rail wagon based on image processing | |
| CN112330633B (en) | Jumper wire adhesive tape damage fault image segmentation method based on self-adaptive band-pass filtering | |
| CN112837285B (en) | Edge detection method and device for panel image | |
| CN114693652A (en) | Fabric defect detection method based on Gaussian mixture model | |
| CN113283339A (en) | Machine vision-based method for monitoring material shortage abnormity of industrial mineral aggregate conveying belt |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210903 |
|
| RJ01 | Rejection of invention patent application after publication |