CN115535525A - Conveyor belt longitudinal tearing detection system and method based on image matching - Google Patents
Conveyor belt longitudinal tearing detection system and method based on image matching Download PDFInfo
- Publication number
- CN115535525A CN115535525A CN202211092351.0A CN202211092351A CN115535525A CN 115535525 A CN115535525 A CN 115535525A CN 202211092351 A CN202211092351 A CN 202211092351A CN 115535525 A CN115535525 A CN 115535525A
- Authority
- CN
- China
- Prior art keywords
- image
- characteristic
- conveying belt
- conveyor belt
- tearing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000001514 detection method Methods 0.000 title claims description 64
- 238000012545 processing Methods 0.000 claims abstract description 32
- 230000003760 hair shine Effects 0.000 claims abstract description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 55
- 238000001914 filtration Methods 0.000 claims description 35
- 238000012937 correction Methods 0.000 claims description 23
- 230000011218 segmentation Effects 0.000 claims description 18
- 230000000877 morphologic effect Effects 0.000 claims description 10
- 230000009467 reduction Effects 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 238000012423 maintenance Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 7
- 238000005457 optimization Methods 0.000 description 6
- 239000002689 soil Substances 0.000 description 6
- 230000008901 benefit Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 241000255969 Pieris brassicae Species 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001364 causal effect Effects 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000191 radiation effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G43/00—Control devices, e.g. for safety, warning or fault-correcting
- B65G43/02—Control devices, e.g. for safety, warning or fault-correcting detecting dangerous physical condition of load carriers, e.g. for interrupting the drive in the event of overheating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G15/00—Conveyors having endless load-conveying surfaces, i.e. belts and like continuous members, to which tractive effort is transmitted by means other than endless driving elements of similar configuration
- B65G15/30—Belts or like endless load-carriers
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G2203/00—Indexing code relating to control or detection of the articles or the load carriers during conveying
- B65G2203/02—Control or detection
- B65G2203/0266—Control or detection relating to the load carrier(s)
- B65G2203/0275—Damage on the load carrier
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
- B65G2203/00—Indexing code relating to control or detection of the articles or the load carriers during conveying
- B65G2203/04—Detection means
- B65G2203/041—Camera
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Image Processing (AREA)
Abstract
The utility model provides a vertical tear detecting system of conveyer belt and method based on image matching, relates to machine vision, image processing technical field, including conveyer belt and tear detecting system, tear detecting system and include a word red laser generator, high-speed CCD camera, siren, computer processing software, a word red laser generator slope sets up the below of conveyer belt, the running direction of laser line perpendicular to conveyer belt that a word red laser generator shines, high-speed CCD camera setting is used for catching below the conveyer belt the laser line, high-speed CCD camera signal connection computer processing software, computer processing software signal connection the siren. By means of the characteristics of good monochromaticity, high contrast, strong directivity and the like of the linear red laser line, the surface of the conveying belt is a high-brightness monochromatic thin line, the surface characteristics of the conveying belt are more prominent, and the image identification processing is more facilitated.
Description
Technical Field
The invention relates to the technical field of machine vision and image processing, in particular to a conveyor belt longitudinal tearing detection system and method based on image matching.
Background
The conveyor is a key device for transporting coal, ore and the like, and the cost of the conveying belt is high and is about half of the total cost of the conveyor. In the long-term and high-load operation process of the belt conveyor, the conveying belt is in danger of being scratched and pierced by a carrier roller, roller friction, material clamping and pressing, blanking impact, foreign matters mixed with materials and the like, so that accidents such as tearing of the conveying belt are caused. The conveyer operation speed is fast, the distance is long, in case take place to tear the accident, must in time detect out and shut down, if can not solve effectively, will cause major incident such as vertical tear, can cause tens meters or even hundreds meters conveyer belt damage usually, and then lead to the shut down, cause the damage and the casualties and huge economic loss of equipment, seriously influence safety in production.
The existing mainstream methods for detecting abnormal conditions such as tearing of the conveying belt comprise three types: manual detection, contact detection, and non-contact detection. The manual detection mode has low efficiency, strong subjective factors and high labor cost; the contact detection method such as material leakage detection and pressure test has the advantages that although the structure is simple, the accuracy is low, the method is easy to damage, and the phenomena of leakage detection and error detection are easy to generate; non-contact detection methods such as embedding, ultrasound, X-ray detection, radio frequency, machine vision. Although the embedding method, the ultrasonic method and the frequency transmission method are reliable, the cost is high and the structure is complex; the X-ray detection has high precision, but has radiation effect on human bodies. In the running process of the conveyer belt, for the longitudinal tearing problem of the conveyer belt, a fault model is uncertain, time is random, the causal relationship is complex, fault diagnosis and detection theories are imperfect, the safety production of a transportation system and the service life of the transportation system are seriously influenced, and a new theoretical method is needed for the online detection of the longitudinal tearing of the conveyer belt. In the current related technology, the method for detecting the tearing fault is a novel method which adopts image processing to detect, but due to the complexity of the industrial field background and the possible interference of the conveying belt, the accuracy of the existing detection method is not high.
Disclosure of Invention
In view of the above drawbacks, the present invention aims to provide a system and a method for detecting longitudinal tearing of a conveyor belt based on image matching, which can accurately detect the longitudinal tearing problem of the conveyor belt, can provide tearing alarm information in time in industrial production, and reduce economic and equipment losses caused by tearing accidents.
In order to achieve the purpose, the technical scheme of the invention is as follows:
detection system is vertically torn to conveyer belt based on image matching, include the conveyer belt and tear detection system, its characterized in that, it includes laser generator, CCD camera, siren, computer processing software to tear detection system, the laser generator slope sets up the below of conveyer belt, the laser line perpendicular to that laser generator shines out the operation direction of conveyer belt, the CCD camera sets up the conveyer belt below is used for catching the laser line, CCD camera signal connection computer processing software, computer processing software signal connection the siren.
The method comprises the following steps: s1: a linear red laser generator is used for emitting a red laser line as an auxiliary laser source, and a high-speed CCD camera is combined with the red laser line to extract continuous videos of normal operation of a conveying belt through the CCD camera; s2: transmitting the continuous video extracted by the CCD camera to computer processing software for video processing, wherein when tearing does not occur, a red laser line is in a continuous curve, discontinuous breakpoints can appear on the red laser line at the tearing position, a current video with the initial state surface characteristics of the conveying belt is obtained after video processing, and the current video of the next section of the conveying belt is obtained after tearing judgment for dynamic updating; s3: comparing the current video with a template image formed when the conveyer belt rotates for one circle in the initial state after daily maintenance, after obtaining the template image of the surface characteristic of the conveyer belt and the current comparison image, accurately positioning the current video of the conveyer belt on the template image by using an image matching algorithm, analyzing the characteristic difference between the current video and the template image, and finally analyzing whether the difference is longitudinal tearing.
Wherein, the video processing in S2 comprises the following steps: the method comprises the following steps: extracting the section characteristic image of the single-frame conveying belt from the continuous video extracted by the CCD camera and running normally of the conveying belt, and then performing RGV (reduced graphics volume) graying processing on the section characteristic image of each frame of the conveying belt; step two: performing noise reduction treatment on the characteristic image of the section of the single-frame conveying belt subjected to RGV graying treatment in the step one, wherein the noise reduction treatment adopts a weighted fast median filtering algorithm; step three: the video processed in the second step is subjected to area segmentation on the characteristic curve area of the section of the conveying belt by adopting an automatic threshold value optimizing area segmentation algorithm; step four: removing the background image of the video in the third step; step five: after the background of the conveying belt is removed, carrying out subsection slope curve correction on the characteristic curve of the section of the conveying belt; step six: splicing and correcting the surface characteristic images of the conveying belt, wherein the splicing and correcting comprises jitter correction and size correction, and then splicing each frame of image by using an image splicing algorithm to obtain a current video with the surface characteristic of the conveying belt in an initial state; step seven: and acquiring the current video of the next section of the conveying belt for dynamic updating after tearing judgment.
After the conveyor belt system is maintained daily, positioning is carried out by using a coded disc, videos of the conveyor belt in an initial state of no load for one circle are collected, an RGV gray level algorithm, weighted rapid median filtering, an automatic threshold value optimizing region segmentation algorithm, background removal, jitter correction and size correction are carried out on each frame of image, and finally the videos are spliced into a template image with the surface characteristics of the conveyor belt in the initial state.
The process of comparing the current video with the template image comprises the following steps: carrying out image matching on the current video in a template image, finding the surface characteristic condition of the initial state of the conveying belt corresponding to the current video, judging which characteristic regions in the current video are newly added regions, carrying out area filtering on the newly added regions, excluding the regions with smaller areas in the newly added regions, carrying out length filtering detection on the remaining newly added regions, judging that the conveying belt is longitudinally torn if the length of the newly added characteristic regions exceeds a tearing alarm threshold value set in advance, positioning the current video by using a code disc, and sending alarm information through an alarm.
Wherein the pattern matching adopts an improved normalized cross-correlation matching algorithm.
The area filtering implementation method comprises the following steps: 1. carrying out 8 neighborhood calibration on a binary image f (i, j) with the surface characteristics of the conveying belt by using a morphological marking algorithm, marking each characteristic region as different gray levels according to a positive integer sequence, and if the number of the marked regions is q, setting the maximum gray level of the marked region as q;2. calculating the number of each gray level to calculate the area S of each characteristic region k Size; 3. when the area S of the characteristic region k And when the area is smaller than the preset area critical threshold value Tm, clearing the corresponding gray level.
The method for realizing the length filtering detection comprises the following steps: 1. carrying out 8 neighborhood calibration on a binary image f (i, j) with the surface characteristics of the conveying belt by using a morphological marking algorithm, marking each characteristic region as different gray levels according to a positive integer sequence, and if the number of the marked regions is q, setting the maximum gray level of the marked region as q;2. searching the maximum and minimum points of each gray level in the longitudinal direction, wherein the difference b of the maximum and minimum points is the length of the characteristic region in the longitudinal direction; 3. and when the length b of the characteristic region is smaller than a preset length critical threshold Tl, clearing the corresponding gray scale level.
And judging whether the longitudinal tearing of the conveying belt occurs according to a standard increasing inclination angle, wherein after the inclination angle is judged to be increased to the length filtering detection, the inclination angle judging method comprises the following steps: 1. carrying out 8 neighborhood calibration on the image of the newly added surface features after area and length judgment by using a morphological marking algorithm, and marking each feature region as different gray levels; 2. respectively performing linear fitting on each gray level, solving a point-slope equation y = kx + b of each gray level, wherein k is the slope of a straight line, and then solving an inclination angle alpha through a formula alpha = arctan (k); 3. the characteristic of longitudinal tearing is kept when the inclination angle alpha is between 80 and 100, and the characteristic regions with the inclination angle a not in the range are removed.
The size correction adopts image scaling processing, and the scaling method adopts double-line interpolation.
After the technical scheme is adopted, the invention has the beneficial effects that:
1. the invention designs a conveyor belt longitudinal tearing detection system and method based on image matching, the system adopts a line red laser as an auxiliary laser source, a line laser emitter is arranged below a conveyor belt, and irradiated laser lines are perpendicular to the conveyor belt and form a certain angle with the running direction of the conveyor belt, so that the laser lines formed on the lower surface of the conveyor belt can scan all parts of the conveyor belt along with the running of a conveyor belt system when a high-speed CCD camera carries out shooting.
2. The region segmentation algorithm for automatic threshold value optimization solves the problem that the global threshold value is difficult to effectively segment the red laser line characteristics. Aiming at the problems of deformation when the conveying belt is loaded relative to the idle load and relative position change of characteristic points caused by left-right shaking and up-down vibration when the conveying belt runs, the invention provides a method for carrying out curve correction on a red laser line according to a segmented slope, and a method for correcting shaking of the conveying belt and correcting the size of an image.
In conclusion, the conveyor belt longitudinal tearing detection system and method based on image matching solve the technical problem that the conveyor belt longitudinal tearing detection method in the prior art is not high in accuracy.
Drawings
FIG. 1 is an effect diagram of an RGV gray scale image and three region segmentation algorithms;
FIG. 2 shows a simulation effect of the tear detection system;
FIG. 3 is a flow chart of the detection system;
Detailed Description
The invention is further illustrated below with reference to the figures and examples.
All the orientations referred to in the present specification are based on the orientations shown in the drawings, and represent relative positional relationships only, and do not represent absolute positional relationships.
The detection system is vertically torn to conveyer belt based on image matching, including conveyer belt and tearing detection system, it includes a word red laser generator, high-speed CCD camera, siren, computer processing software to tear detection system, a word red laser generator slope sets up in the below of conveyer belt, the running direction of laser line perpendicular to conveyer belt that a word red laser generator shines, high-speed CCD camera sets up and is used for catching the laser line below the conveyer belt, high-speed CCD camera signal connection computer processing software, computer processing software signal connection siren.
In the data acquisition stage, firstly, the tearing condition of the running conveyor belt is simulated and analyzed, when the conveyor belt system runs normally, if the longitudinal tearing condition happens suddenly, the tearing part is notched or overlapped in the direction vertical to the running direction of the conveyor belt, and according to the characteristic, a digital image processing mode is utilized, and a method of combining a high-speed CCD camera and a digital red laser source is used for carrying out transmission processing in the form of video. When the tearing does not occur, the red laser line is in a continuous curve, and discontinuous breakpoints can appear on the laser line at the tearing occurring position. Therefore, after processing of multi-frame continuous images in a video, the detection of the breakpoint position of the longitudinal tearing image of the conveying belt is compared with a template image formed when the conveying belt rotates for one circle in an initial state after daily maintenance, and whether the longitudinal tearing fault occurs on the conveying belt can be analyzed and judged. The experimental method can effectively eliminate the interference of patches, soil and the like, and improve the accuracy of judging whether longitudinal tearing occurs. The details are as follows:
a detection method of a conveyor belt longitudinal tearing detection system based on image matching comprises the following steps:
s1: a linear red laser generator is used for emitting a red laser line as an auxiliary laser source, and a high-speed CCD camera is combined with the red laser line to extract continuous videos of normal operation of a conveying belt through the CCD camera;
firstly, a mining conveying belt system is temporarily built in a production workshop of a certain company, and the system comprises a conveying belt, a three-phase motor, a coded disc, a metal bracket, a series of carrier rollers, a roller, a metal detection system and a tearing detection system. The three-phase motor drives the conveying belt to rotate, and the tearing detection system comprises a linear red laser generator, a high-speed CCD camera, an alarm and computer processing software. The red laser generator is placed below the conveying belt, irradiated laser lines are perpendicular to the conveying belt running direction and form a certain angle with the conveying belt running direction, and therefore the laser line coverage area is large. By means of the characteristics of good monochromaticity, high contrast, strong directivity and the like of the linear red laser line, the surface of the conveying belt is a highlight monochromatic thin line, the surface characteristics of the conveying belt are more prominent, and the image identification processing is more facilitated.
S2: transmitting the continuous video extracted by the CCD camera to computer processing software for video processing, wherein when tearing does not occur, a red laser line is in a continuous curve, discontinuous breakpoints can appear on the red laser line at the tearing position, a current video with the initial state surface characteristics of the conveying belt is obtained after video processing, and the current video of the next section of the conveying belt is obtained after tearing judgment for dynamic updating;
the video processing comprises the following steps:
the method comprises the following steps: extracting single-frame conveyor belt section characteristic images of a continuous video extracted by a CCD camera and used for normal operation of a conveyor belt, and then performing RGV (reduced graphics vector) graying processing on each frame of conveyor belt section characteristic image;
the algorithm utilizes three types of components, namely R (red component), G (green component) and V (brightness) in an HSV color space model, in an RGB color space model, and respectively gives corresponding weight values to perform graying of comprehensive calculation. Standard graying is not ideal for graying locations such as dirt, edges, etc. in the conveyor belt image. The gray image calculated by RGV graying is clearer in detail than the standard graying, and the edge contrast of the red laser line position is more obvious.
Standard graying formula:
I(i，j)＝0.2989*RGB(i，j，1)+0.5870*RGB(i，j，2)+... 0.1140*RGB(i，j，3) (1)
RGV composite graying formula:
I(i，j)＝0.358*RGB(i，j，1)+0.327*RGB(i，j，2)+... 0.314*HSV(i，j，3) (2)
step two: performing noise reduction treatment on the characteristic image of the section of the single-frame conveying belt subjected to RGV graying treatment in the step one, wherein the noise reduction treatment adopts a weighted fast median filtering algorithm;
because the mine field environment where the conveying belt system is located is complex, and the interference such as dust, illumination is large, the images collected by the CCD camera have interference information, and the images also generate noise in the transmission and storage processes, so that the images need to be subjected to noise reduction treatment. Standard median filtering requires a lot of rank comparison work, and this study is time consuming for the calculation of noise reduction for each frame of the video and the belt image. Therefore, speed-up optimization is carried out on the basis of standard median filtering, a weighted fast median filtering algorithm is adopted, the weighted fast median filtering algorithm firstly considers the influence of the gray values of the pixels in two columns shifted in and out on the median, and therefore the sequencing calculation amount of similar and even identical pixel gray level areas is reduced. The weighted fast median filtering algorithm not only obtains better denoising effect than the standard median filtering, but also better saves the edge information of the original image, and the experimental test shows that the weighted fast median filtering time is shortened by more than 70% compared with the standard median filtering time.
Step three: the video processed in the second step is subjected to area segmentation on the characteristic curve area of the section of the conveying belt by adopting an automatic threshold value optimizing area segmentation algorithm; by comparing the effect graphs of the Otsu algorithm and the iterative threshold segmentation algorithm, the algorithm can effectively improve the accuracy of the image segmentation of the conveying belt.
The region segmentation algorithm for automatic threshold optimization first uses an image rotation algorithm to keep the running direction of the conveyor belt in the image from top to bottom or from bottom to top so as to keep the red laser line on the image belt transverse in the overall direction. Most of the traditional conveying belts are dark, so that the collected video data can also display dark colors, and a red laser line is used as an auxiliary laser source and irradiated on the conveying belt to form a strip-shaped brightness area with higher brightness and element gray values than a normal conveying belt area. It can be observed that the gray value of the pixel in the laser line irradiation area is higher than that of other pixels in the same column. By utilizing the characteristic, the research divides the conveying belt into a plurality of areas with equal size according to the longitudinal direction, and integrates the advantages of iterative threshold segmentation and Otsu to perform automatic threshold value optimizing area segmentation binarization on the conveying belt image.
The area binarization computing method for automatic threshold value optimization comprises the following steps:
knowing that the pixel size of the original gray image is [ m, a x n ], dividing the image into a parts according to longitudinal average, and automatically searching for the optimal threshold value, wherein the pixel size of each part of the image is [ m, n ];
inquiring the pixel width of a red laser line in an image as u in advance;
setting I (I, j) as pixel gray values of image (I, j) positions, wherein 1-I-j-m, 1-j-n;
calculating the optimal threshold value Ta of each block area of the image by using the following calculation formula:
1. the gray level range of the original gray image is [0,1], a pixel value 0 is selected as an initial threshold value T0 of the whole image, the threshold value is set to be increased by &aftereach iteration, namely the precision level of the threshold value is &;
2. dividing the image into two regions by using an initial threshold value T0, wherein the number of pixel points with the gray level greater than T0 is N2, and the number of pixel points with the gray level less than or equal to T0 is N1;
determining whether the following inequality holds:
N 2 ＞＝u*n (3)
3. if the inequality is true, a new threshold value Ti +1 is calculated by using the following formula:
T i+1 ＝T i +& (4)
4. repeating the steps 2 and 3 until the inequality is not satisfied.
Step four: removing the background image of the video in the third step;
removing the background of the high-exposure conveying belt: after the binarization by the automatic threshold value optimizing region segmentation algorithm, a large white background can be displayed at the background, and the problem of large white background elements caused by high exposure after the binarization in the conveying belt image is solved by means of the comprehensive application of partial morphological algorithm in the digital image processing.
Removing the background of the low-exposure and normal-exposure conveyer belt: in order to eliminate partial interference points caused by low exposure except the section characteristic curve and accurately segment and position the section characteristic curve of the conveying belt in the image, curve fitting needs to be carried out on the image after binaryzation of the conveying belt, and after comparison and analysis, an iterative and low-order nonlinear least square curve fitting algorithm is selected to fit the laser line in the image. The advantage of the low order versus the high order is that it does not fit all points of the known data into the curve, but rather, while smoothing the curve, approximates as many data points as possible, thus eliminating individual noise points in the image after one fit, while after multiple low order fits, the curve fit will maximally approximate the red laser line while eliminating interference points.
Step five: after the background of the conveying belt is removed, carrying out subsection slope curve correction on the characteristic curve of the section of the conveying belt; the problem that the aspect ratio of the surface characteristic position of the conveying belt is influenced by the increase of the curvature of the laser line when the conveying belt is loaded can be effectively solved by correcting the sectional slope curve, so that the accuracy rate of image matching is increased, and the system can detect the tearing problem of the conveying belt under the conditions of no load and load.
The red laser line irradiation area in the image of the conveying belt can present a linear or low-curvature curve when the conveying belt is in no load, and when the conveying belt is loaded, the curvature of the red laser line irradiation area can be obviously increased due to the deformation of the conveying belt, and the pixel width of the conveying belt in the image is reduced due to the inward contraction of the edge position of the conveying belt. Therefore, the straight line correction of the laser line can not be directly carried out by using a conventional projection method, and the algorithm for correcting the sectional curve is designed according to the change rule of the curvature of the red laser line in the conveying belt.
The specific content of the algorithm is as follows:
1. dividing the image of the conveying belt into smaller areas longitudinally, wherein the laser line in each area is close to a straight line;
2. performing straight line fitting on the laser line in each area by adopting a least square method, wherein the fitting formula is (5);
3. according to the fitted linear coefficients a and b, the positions of the laser lines at the two ends of each area are obtained, and the area image is cut according to the obtained position information to obtain an image G;
4. according to the slope a of the fitted straight line, calculating the inclination angle alpha of the straight line by using a formula (6);
α＝arctan(a) (6)
5. rotating the image G around a central point according to the inclination angle alpha, and performing neighborhood interpolation by using a nearest method, wherein the formula is (7) to ensure that a complete rotation image H is generated;
where x, y are both non-negative integers and f (x, y) represents the pixel value at the source image (x, y).
6. The length of the rotated image is kept unchanged around the central point, and the image is cut by twice the width of the laser line pixel so as to keep the widths of all the areas of the images of the conveying belt uniform;
7. and transversely splicing the laser line images obtained in the steps in all areas of the conveyor belt image to obtain an image with the corrected conveyor belt section curve.
After the slope correction is carried out on the section characteristic image of the conveying belt, the length of the section characteristic is corrected and is linear, but the edge position of the section characteristic is uneven, so that the subsequent splicing work is not facilitated, the section characteristic needs to be smoothly corrected, and the edge can be smoothly corrected by directly adopting a projection method. The projection method is to project an image in a specified direction and to realize the purpose of image processing and analysis according to the characteristics of the projection.
Assuming that the size of the image is m × n, i.e., m rows and n columns, and f (x, y) is a gray value at (x, y), after an automatic threshold-optimization region segmentation algorithm, the image becomes a binary image, wherein a black (gray value is 0) region represents a conveyor belt background or a fracture, and a white (gray value is 1) region represents a conveyor belt cross-section feature. And (3) accumulating the gray values of the images f (x, y) along the longitudinal direction, taking an average value, and then performing projection in the horizontal direction to obtain the conveyor belt section linear characteristic T (y) with smooth edges. The calculation formula is as follows:
step six: splicing and correcting the surface characteristic images of the conveying belt, wherein the splicing and correcting comprises jitter correction and size correction, and then splicing each frame of image by using an image splicing algorithm to obtain a current video with the surface characteristic of the conveying belt in an initial state;
in the process of image acquisition, the conveying belt shakes left and right and shakes up and down, so that in the conveying belt video recorded by the CCD camera, the distances from the left edge and the right edge of the conveying belt to the left boundary and the right boundary of the image in each frame of image have certain difference, and left and right deviation can occur at the tearing position detected by each frame of image, so that the edge of the conveying belt needs to be corrected. The method carries out shake correction on the characteristic image of the section of the conveying belt by detecting the initial positions of laser lines on the edges of the two sides of the conveying belt in the image.
For the left edge of the laser line, the conveyor belt section feature image is scanned from left to right, and the position of the sharp rising boundary of the white pixel is recorded, so that the left edge of the conveyor belt section feature image is determined. For the right edge of the characteristic image of the section of the conveying belt, the image is scanned from right to left, and the position of a boundary where a white pixel rises sharply is recorded, so that the right edge of the characteristic image of the section of the conveying belt is determined.
And (4) correcting the size of the image, wherein in order to ensure that the number of pixels of each line of the characteristic image of the section of the conveying belt obtained by each frame of image of the video data through the steps is the same and ensure that the following splicing algorithm is successful, the size of the breakpoint image of the laser line needs to be corrected, so that the number of pixels of each line is kept consistent.
The image zooming processing is mainly used for changing the size of an image, after the image is zoomed, the sizes of the line number and the column number of the image are changed correspondingly, when the line number or the column number of the image is increased, the pixels of the image are higher, and the image is blurred; as the number of rows or columns of the image decreases, smoothness and sharpness increase.
The condition that the corresponding pixel point cannot be found exists in the original image pixel after the zooming, so that the approximation processing must be carried out according to the pixel value of the neighborhood. The general processing method is to perform interpolation processing, and common interpolation processing includes nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation.
After the characteristic area of the surface of the conveying belt is extracted, the spliced data volume is large, so that the scaling method adopts double-line interpolation, the calculation speed can be improved, and a better processing result can be obtained.
After the cross section characteristics of the conveying belt are extracted, subjected to shake correction and size correction through the steps and then spliced together in the same sequence, video data are converted into binary images with the surface characteristics of the conveying belt. The width of the characteristic image is the image width set when the characteristic image of the section of the conveying belt is corrected in size, and the pixel number on the image height is the number of frames of the video multiplied by the pixel width of the laser line on the image of the conveying belt.
Step seven: and acquiring the current video of the next section of the conveying belt for dynamic updating after tearing judgment.
S3: comparing the current video with a template image formed when the conveyer belt rotates for one circle in the initial state after daily maintenance, obtaining the template image of the surface characteristics of the conveyer belt and the current contrast image, accurately positioning the current video of the conveyer belt on the template image by using an image matching algorithm, analyzing the characteristic difference between the current video and the initial no-load state of the conveyer belt, and finally analyzing whether the difference is longitudinal tearing. Whether the longitudinal tearing fault occurs to the conveying belt can be analyzed and judged.
The method comprises the steps of dynamically generating a current video of a conveying belt, calculating and processing each frame image of the current video dynamically acquired in the normal operation process of the conveying belt according to the steps of RGV graying, weighted rapid median filtering, segmentation automatic optimization area segmentation algorithm binarization, morphology background removal, curve fitting, jitter correction, size correction and the like introduced above, finally splicing the video into the current video with the surface characteristics of the initial state of the conveying belt, and acquiring the current video of the next section of the conveying belt for dynamic updating after tearing judgment.
The characteristic region matching and tearing result judging method comprises the steps of carrying out matching positioning on a present video in a template image, finding the surface characteristic condition of the initial state of a conveying belt corresponding to the present video, judging which characteristic regions in the present video are newly increased regions, carrying out area filtering on the newly increased regions, excluding the regions with smaller areas in the newly increased regions, carrying out length filtering detection on the remaining newly increased regions, judging that the conveying belt is longitudinally torn if the length of the newly increased characteristic regions exceeds a preset tearing alarm threshold value, positioning the present video by using a code disc, sending alarm information by using an alarm, carrying out the above algorithm to obtain a template image and a present comparison image of the surface characteristic of the conveying belt, carrying out accurate positioning on the present video of the conveying belt on the template image by using an image matching algorithm, analyzing the characteristic difference of the initial no-load state of the present video and the conveying belt, and finally analyzing whether the difference is longitudinal tearing.
And performing image matching by adopting an improved normalized cross-correlation matching algorithm, assuming that S (x, y) is the current video with the surface characteristics of the conveyer belt with the size of M multiplied by N, and T (x, y) is an M multiplied by N conveyer belt template image with the surface characteristics of the initial state of the conveyer belt, wherein M is more than 0 and less than or equal to M, and N is more than 0 and less than or equal to N. The area of the current video S moving on the template image T of the conveyor belt covered by the current video is called a template image sub-picture T (x, y). And (i, j) is the coordinate of the pixel point at the upper left corner of the template image subgraph in the template image T, i is more than or equal to 0 and less than or equal to M-M +1, and j is more than or equal to 0 and less than or equal to N-N +1, and the coordinate is called as a reference point. In the template image T (x, y), taking (i, j) as the upper left corner, taking a template image subgraph of the size of m multiplied by n, calculating the gray value similarity of the template image subgraph and the current video, traversing the current video through the whole template image, and finding the template subgraph most similar to the current video as a final matching result [56]. In general, the similarity matching algorithm for T (x, y) and S (x, y) can be achieved by the following algorithm:
normalized cross-correlation algorithm (NCC), formula:
wherein E (Ti, j) and E (S) respectively represent the average gray scale of the current sub-image and the current video at (i, j)
Because the template image and the current video in the research are binary images obtained by extracting and splicing the cross section characteristics of the conveying belt, the black problem characteristic area (the pixel value is 1) is less, and the white normal conveying belt area (the pixel value is 0) is more, so the mean value difference of the template image and the current video is very slight, an improved NCC algorithm is adopted, and the gray level correlation of the two images is obtained by a formula (11):
because the template image in the research and the current video are corrected by adopting the same size when being subjected to multi-image splicing, the two images have the same width, and when a matching point is searched, the template image is matched from top to bottom within a certain range only according to the rough positioning of a code disc, so that a matching result can be obtained.
After the image matching is successful, the program feeds back position information, wherein the position information is the position information corresponding to the first element on the upper left corner of the current video of the conveyer belt on the template image. And using the position information to frame the corresponding position of the current video on the template image of the conveying belt by using a red frame according to the size of the current video of the conveying belt.
Because the initial surface state of the conveyor belt has a certain probability of being a characteristic region formed by objects which are easy to fall off, such as soil, dirt and the like, the characteristic region can fall off along with the normal operation of the conveyor belt system, so that the characteristic region has a characteristic region on the template image of the conveyor belt and is not provided in the current video.
Therefore, the processing of the current video and the corresponding feature position of the template image only needs to reserve the added feature area of the current video and does not reserve other feature areas. When the current video of the conveyer belt is compared with the template image, the required effect cannot be achieved by simple AND operation.
In the research, a method that the current situation video and the corresponding position of the template image are subtracted, the [0,1] part of the subtraction result is reserved, and other results are abandoned is adopted. The method can effectively reserve the newly added conveying belt surface characteristic area in the current video.
The area judgment is that when the conveyor belt system normally operates, objects such as soil, dirt, conveyor belt joints, small patches and the like may be attached to the surface of the conveyor belt, which affects the continuity of red laser lines, and causes spots, small-area characteristic lines and the like in the current video of the conveyor belt.
The area filtering implementation method comprises the following steps:
1. carrying out 8 neighborhood calibration on a binary image f (i, j) with the surface characteristics of the conveying belt by using a morphological marking algorithm, marking each characteristic region as different gray levels according to a positive integer sequence, and if the number of the marked regions is q, setting the maximum gray level of the marked region as q;
2. calculating the number of each gray level to calculate the area S of each characteristic region k Size;
3. when the area S of the characteristic region k And when the area is smaller than the preset area critical threshold value Tm, clearing the corresponding gray level.
Therefore, the area filtering effect can be achieved, and the area filtering can effectively filter the target with a small area in the binary image, so that the algorithm also has a good filtering effect on the current video representing the surface characteristics of the conveyor belt.
And (5) judging the length. After area filtering, the feature region with a small newly added area in the current video is eliminated, and at this time, a newly added feature region with a sufficiently large area remains, but the feature region may have situations such as a conveyor belt joint, a transverse patch, and the like.
The method of length filtering is similar to the area filtering method except that the area calculation of the middle portion is changed to length calculation. The area filtered threshold Tm is changed to the length filtered threshold Tl.
The implementation method of the length filtering comprises the following steps:
1, carrying out 8 neighborhood calibration on a binary image f (i, j) with the surface characteristics of a conveyor belt by using a morphological marking algorithm, marking each characteristic region as different gray levels according to a positive integer sequence, and if the number of the marked regions is q, setting the maximum gray level of the marked region as q;
2. searching the maximum and minimum points of each gray level in the longitudinal direction, wherein the difference b of the maximum and minimum points is the length of the characteristic region in the longitudinal direction;
3. and when the length b of the characteristic region is smaller than a preset length critical threshold Tl, clearing the corresponding gray scale level.
After the length filtering is finished, the remaining feature regions in the newly added feature region map satisfy the tearing conditions in terms of area and length, but in order to make the tearing determination more accurate, the determination of the inclination angle of the image is also required.
And judging the inclination angle. According to the experience of several skilled maintenance workers in the actual application field of the mine conveyor belt system, when the longitudinal tearing fault occurs to the conveyor belt, the tearing angle is between 80 and 100 degrees. After the area and length are determined, a region with the area and the length in the newly added feature image according with the longitudinal tearing feature is reserved, at this time, the detection and determination of the inclination angle of the remaining feature region are still needed, and if the slope of the feature region after straight line fitting is between [5.67, + ∞ ] and [ -5.67, - ∞ ], namely the inclination angle alpha is between 80 and 100, the longitudinal tearing is determined to occur.
The method for determining the inclination angle is as follows:
1. carrying out 8 neighborhood calibration on the image of the newly added surface features after area and length judgment by using a morphological marking algorithm, and marking each feature region as different gray levels;
2. respectively performing linear fitting on each gray level, solving a point-slope equation y = kx + b of each gray level, wherein k is the slope of the straight line, and then solving an inclination angle alpha through a formula (12);
α＝arctan(k) (12)
3. the characteristic of longitudinal tearing is kept when the inclination angle alpha is between [80,100], and the characteristic regions with the inclination angle alpha not in the range are removed.
After comprehensive application is carried out through three determination methods of area, length and inclination angle, the three characteristic conditions are simultaneously met, the tearing characteristic is determined to be the tearing characteristic, longitudinal tearing can be determined to occur, and alarm information is sent out through an alarm.
And (3) actual measurement data verification: in order to verify the feasibility of the scheme, simulated experimental data are collected on a conveyor belt system temporarily built in a production plant of a certain company, and the algorithm is verified on corresponding mathematical software in a laboratory.
The experimental result shows that the template image and the current video obtained by performing image matching tear detection analysis on a group of conveyor belt videos are shown in fig. 2, the tear detection result is shown in fig. 2 (b), and the red frame is an area determined as a tear position. The characteristic region shown in fig. 2 (a) is a characteristic region obtained by extracting characteristics after interference of soil, patches and the like on laser lines is in a blue frame, a characteristic region caused by cracks is in a red frame, a characteristic position caused by patches of a conveyer belt is in a green frame, the characteristic region is matched and compared with a template image, the original characteristic region of the initial state of the conveyer belt is removed, and after the judgment methods of area, length, inclination angle and the like are carried out, the interference caused by newly added small blocks of soil, stains and the like is removed, the remaining characteristic region is a part with longitudinal tearing.
The conveying belt longitudinal tearing detection system and method based on image matching, which are described by the invention, have the advantages of improving the algorithm and the effectiveness of the method through specific experimental data, better solve the problem of interference of conveying belt patches, joints and soil on single image tearing fault identification, effectively improve the accuracy of the conveying belt tearing detection result, and have great significance in the fields of machine vision and image processing.
The present invention is not limited to the above-described embodiments, and those skilled in the art will be able to make various modifications without creative efforts from the above-described conception, and fall within the scope of the present invention.
Claims (10)
1. Vertical tear detecting system of conveyer belt based on image matching, including conveyer belt and tear detecting system, its characterized in that, tear detecting system and include laser generator, CCD camera, siren, computer processing software, laser generator slope sets up the below of conveyer belt, the laser line perpendicular to that laser generator shines the running direction of conveyer belt, the CCD camera sets up the conveyer belt below is used for catching the laser line, CCD camera signal connection computer processing software, computer processing software signal connection the siren.
2. The detection method of the conveyor belt longitudinal tearing detection system based on image matching is characterized by comprising the following steps of:
s1: a linear red laser generator is used for emitting a red laser line as an auxiliary laser source, and a high-speed CCD camera is combined with the red laser line to extract continuous videos of normal operation of a conveying belt through the CCD camera;
s2: transmitting the continuous video extracted by the CCD camera to computer processing software for video processing, wherein when tearing does not occur, the red laser line is in a continuous curve, discontinuous breakpoints can appear on the red laser line at the tearing position, a current video with the initial state surface characteristics of the conveying belt is obtained after video processing, and the current video of the next section of the conveying belt is obtained after tearing judgment for dynamic updating;
s3: comparing the current video with a template image formed when the conveyer belt rotates for one circle in the initial state after daily maintenance, after obtaining the template image of the surface characteristic of the conveyer belt and the current comparison image, accurately positioning the current video of the conveyer belt on the template image by using an image matching algorithm, analyzing the characteristic difference between the current video and the template image, and finally analyzing whether the difference is longitudinal tearing.
3. The detection method for the conveyor belt longitudinal tearing detection system based on image matching as claimed in claim 2, wherein the video processing in S2 comprises the following steps:
the method comprises the following steps: extracting single-frame conveyor belt section characteristic images of a continuous video extracted by a CCD camera and used for normal operation of a conveyor belt, and then performing RGV (reduced graphics vector) graying processing on each frame of conveyor belt section characteristic image;
step two: performing noise reduction treatment on the characteristic image of the section of the single-frame conveying belt subjected to RGV graying treatment in the step one, wherein the noise reduction treatment adopts a weighted fast median filtering algorithm;
step three: the video processed in the second step is subjected to area segmentation on the characteristic curve area of the section of the conveying belt by adopting an automatic threshold value optimizing area segmentation algorithm;
step four: removing the background image of the video in the third step;
step five: after removing the background of the conveying belt, carrying out sectional slope curve correction on the characteristic curve of the section of the conveying belt;
step six: splicing and correcting the surface characteristic images of the conveying belt, wherein the splicing and correcting comprises jitter correction and size correction, and then splicing each frame of image by using an image splicing algorithm to obtain a current video with the surface characteristic of the conveying belt in an initial state;
step seven: and acquiring the current video of the next section of the conveying belt for dynamic updating after tearing judgment.
4. The detection method of the conveyor belt longitudinal tearing detection system based on image matching as claimed in claim 2, wherein the template image generation step includes positioning by using a code wheel after daily maintenance of the conveyor belt system, collecting the video when the conveyor belt is unloaded for one turn at the initial state, and performing RGV graying, weighted fast median filtering, automatic threshold value optimizing region segmentation algorithm, background removal, jitter correction and size correction on each frame image, and finally splicing the video into the template image with the surface characteristics of the conveyor belt at the initial state.
5. The detection method of the conveyor belt longitudinal tearing detection system based on image matching as claimed in claim 2, wherein the process of comparing the current video with the template image comprises the following steps: the method comprises the steps of carrying out image matching on a current video in a template image, finding the surface characteristic condition of an initial state of a conveying belt corresponding to the current video, judging which characteristic regions in the current video are newly added regions, carrying out area filtering on the newly added regions, removing regions with smaller areas in the newly added regions, carrying out length filtering detection on the remaining newly added regions, judging that the conveying belt is longitudinally torn if the length of the newly added characteristic regions exceeds a preset tearing alarm threshold value, positioning the current video by using a code disc, and sending alarm information through an alarm.
6. The detection method for the conveyor belt longitudinal tearing detection system based on image matching as claimed in claim 5, wherein the image matching adopts a modified normalized cross-correlation matching algorithm.
7. The detection method of the conveyor belt longitudinal tear detection system based on image matching as claimed in claim 5, wherein the implementation method of the area filtering is as follows: 1. using morphological marking algorithms on binary images with conveyor surface features
Performing 8 neighborhood calibration, marking each characteristic region as different gray levels according to the sequence of positive integers, and if the number of the marked regions is q, then the maximum gray level of the marked region is q;2. the number of each gray level is calculated, thereby calculating eachArea of the characteristic region
Size; 3. area of characteristic region
And when the area is smaller than the preset area critical threshold value Tm, clearing the corresponding gray level.
8. The detection method of the conveyor belt longitudinal tearing detection system based on image matching as claimed in claim 5, characterized in that the implementation method of the length filtering detection comprises the following steps: 1. using morphological marking algorithms on binary images with conveyor surface features
Performing 8 neighborhood calibration, marking each characteristic region as different gray levels according to the sequence of positive integers, and if the number of the marked regions is q, then the maximum gray level of the marked region is q;2. searching the maximum and minimum points of each gray level in the longitudinal direction, wherein the difference b of the maximum and minimum points is the length of the characteristic region in the longitudinal direction; 3. and when the length b of the characteristic region is smaller than a preset length critical threshold Tl, clearing the corresponding gray scale level.
9. The detection method of the conveyor belt longitudinal tearing detection system based on the image matching as claimed in claim 5, characterized in that the criterion for judging the occurrence of the longitudinal tearing of the conveyor belt is added with a slope angle judgment, and after the slope angle judgment is added to the length filtering detection, the slope angle judgment method is as follows: 1. carrying out 8 neighborhood calibration on the image of the newly added surface features after area and length judgment by using a morphological marking algorithm, and marking each feature region as different gray levels; 2. performing linear fitting on each gray level to obtain their point-slope equations
Where k is the slope of the line, and then by the formula
Determining the angle of inclination
(ii) a 3. Angle of inclination
In [80,100]]The characteristic of longitudinal tearing is kept, and the inclination angle is adjusted
Feature regions not within this range are culled.
10. The detection method for the conveyor belt longitudinal tearing detection system based on the image matching as claimed in claim 3, wherein the size correction adopts image scaling processing, and the scaling method adopts two-line interpolation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211092351.0A CN115535525B (en) | 2022-09-08 | System and method for detecting longitudinal tearing of conveyor belt based on image matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211092351.0A CN115535525B (en) | 2022-09-08 | System and method for detecting longitudinal tearing of conveyor belt based on image matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115535525A true CN115535525A (en) | 2022-12-30 |
| CN115535525B CN115535525B (en) | 2024-10-29 |
Family
ID=
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116142727A (en) * | 2023-04-14 | 2023-05-23 | 合肥金星智控科技股份有限公司 | Conveyor belt tearing detection method and system based on laser stripe defect identification |
| CN117585399A (en) * | 2024-01-19 | 2024-02-23 | 汶上义桥煤矿有限责任公司 | Coal mine conveyor belt tearing detection method based on image processing |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102951426A (en) * | 2012-11-07 | 2013-03-06 | 太原理工大学 | Visual detection method of longitudinal tear of conveying belt |
| CN104792796A (en) * | 2015-04-13 | 2015-07-22 | 太原理工大学 | On-line monitoring system for mining adhesive tape operation condition based on machine vision |
| CN111591716A (en) * | 2020-06-08 | 2020-08-28 | 西安电子科技大学 | Photoelectric integrated intelligent detection device for belt damage condition of conveyor |
| CN111862041A (en) * | 2020-07-21 | 2020-10-30 | 华夏天信（北京）智能低碳技术研究院有限公司 | Bottom shaft tail rope fracture detection method based on template matching |
| CN114890090A (en) * | 2022-06-08 | 2022-08-12 | 扎鲁特旗扎哈淖尔煤业有限公司 | Belt longitudinal tearing detection method and device |
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102951426A (en) * | 2012-11-07 | 2013-03-06 | 太原理工大学 | Visual detection method of longitudinal tear of conveying belt |
| CN104792796A (en) * | 2015-04-13 | 2015-07-22 | 太原理工大学 | On-line monitoring system for mining adhesive tape operation condition based on machine vision |
| CN111591716A (en) * | 2020-06-08 | 2020-08-28 | 西安电子科技大学 | Photoelectric integrated intelligent detection device for belt damage condition of conveyor |
| CN111862041A (en) * | 2020-07-21 | 2020-10-30 | 华夏天信（北京）智能低碳技术研究院有限公司 | Bottom shaft tail rope fracture detection method based on template matching |
| CN114890090A (en) * | 2022-06-08 | 2022-08-12 | 扎鲁特旗扎哈淖尔煤业有限公司 | Belt longitudinal tearing detection method and device |
Non-Patent Citations (1)
| Title |
|---|
| 王以娜: "基于视觉检测的皮带纵向撕裂检测关键技术研究", 基于视觉检测的皮带纵向撕裂检测关键技术研究, no. 2021, 16 January 2021 (2021-01-16), pages 1 - 59 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116142727A (en) * | 2023-04-14 | 2023-05-23 | 合肥金星智控科技股份有限公司 | Conveyor belt tearing detection method and system based on laser stripe defect identification |
| CN116142727B (en) * | 2023-04-14 | 2023-09-19 | 合肥金星智控科技股份有限公司 | Conveyor belt tearing detection method and system based on laser stripe defect identification |
| CN117585399A (en) * | 2024-01-19 | 2024-02-23 | 汶上义桥煤矿有限责任公司 | Coal mine conveyor belt tearing detection method based on image processing |
| CN117585399B (en) * | 2024-01-19 | 2024-04-09 | 汶上义桥煤矿有限责任公司 | Coal mine conveyor belt tearing detection method based on image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11688057B2 (en) | Method and system for quickly matching image features applied to mine machine vision | |
| CN114937055B (en) | Image self-adaptive segmentation method and system based on artificial intelligence | |
| CN111667462B (en) | Automatic detection method and system for small and weak targets in welding defects | |
| CN115239704B (en) | Accurate detection and repair method for wood surface defects | |
| CN103051872B (en) | Based on the belt deflection detection method of Edge extraction | |
| US20080232715A1 (en) | Image processing apparatus | |
| CN103543394A (en) | Discharge ultraviolet imaging quantization parameter extraction method of high-voltage electric equipment | |
| US20060067569A1 (en) | Image inspection device, image inspection method, and image inspection program | |
| CN113284154B (en) | Steel coil end face image segmentation method and device and electronic equipment | |
| US10109045B2 (en) | Defect inspection apparatus for inspecting sheet-like inspection object, computer-implemented method for inspecting sheet-like inspection object, and defect inspection system for inspecting sheet-like inspection object | |
| CN118037722B (en) | Copper pipe production defect detection method and system | |
| JP5705711B2 (en) | Crack detection method | |
| CN115690089A (en) | Image enhancement preprocessing method and system for weak defect detection | |
| CN113222992A (en) | Crack characteristic characterization method and system based on multi-fractal spectrum | |
| EP2973223B1 (en) | Enhanced analysis for image-based serpentine belt wear evaluation | |
| CN109767426B (en) | Shield tunnel water leakage detection method based on image feature recognition | |
| US11493453B2 (en) | Belt inspection system, belt inspection method, and recording medium for belt inspection program | |
| CN117689651A (en) | Coal mine belt deviation detection method based on semantic segmentation and target detection | |
| CN109592342B (en) | Visual cylindrical material conveying method | |
| CN115535525B (en) | System and method for detecting longitudinal tearing of conveyor belt based on image matching | |
| CN115535525A (en) | Conveyor belt longitudinal tearing detection system and method based on image matching | |
| JP2021064215A (en) | Surface property inspection device and surface property inspection method | |
| US11880969B2 (en) | Belt examination system and computer-readable non-transitory recording medium having stored belt examination program | |
| CN115100224A (en) | Method and system for extracting coal mine tunnel tunneling head-on cross fracture | |
| JP5190619B2 (en) | Inspection method for inspection object and inspection apparatus therefor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |