Mining conveying belt tearing detection system, method, equipment and medium

Abstract

The invention provides a mining conveyor belt tearing detection system, method and device and a medium, the mining conveyor belt tearing detection system comprising a camera array, a gyroscope, a control module, an image processing module and a detection module is constructed, and sensors such as the camera array and the gyroscope are arranged at proper positions. The image processing module carries out gray level conversion, splicing fusion and edge enhancement on a plurality of shot images and carries out corresponding contour extraction and fusion, and then the detection module carries out calculation on the area, the included angle with the running direction of the conveying belt, the edge gradient mean value and the like according to the fused contour, so that when the calculation result meets the condition, the image processing module carries out detection. In this way, the specific tearing position of the conveying belt can be detected in real time in a fixed point mode, the requirements for real-time accurate detection and early warning are met, and then the production efficiency of a mine is improved in a disguised mode.

Description

Technical Field

[0001] The present application relates to the technical field of industrial vision inspection, and particularly to a tearing detection system, method, device and medium for a mine conveyor belt. Background Art

[0002] At present, belt conveyors are widely used for material transportation in mines. However, when a belt conveyor operates underground, the idlers are under the dual pressures of long-term material load and belt friction. If the idler surface is damaged or the bearing is stuck due to material fatigue or improper installation, its sharp edge will directly contact the belt carrying surface, forming local stress concentration. As the belt continues to run, this frictional effect will gradually expand the damage range, ultimately leading to longitudinal or transverse tearing.

[0003] It is difficult for tearing detection devices such as weighing and sensing detection devices to detect the specific tearing position, which cannot meet the requirements of real-time detection and early warning, resulting in a large amount of time spent on retrieving the fault point. Usually, due to the difficulty in finding the fault location, it will lead to long-term shutdowns, seriously affecting the production efficiency of the mine. Summary of the Invention

[0004] The present application provides a tearing detection system, method, device and medium for a mine conveyor belt, which can solve the problem of inaccurate and real-time detection of conveyor belt tearing in the background art.

[0005] To achieve the above object, the present application adopts the following technical solutions: In a first aspect, a tearing detection system for a mine conveyor belt is provided, the system comprising: A camera array disposed at intervals of idlers of a mine belt conveyor; A gyroscope for dynamically monitoring the real-time belt surface inclination angle of the mine belt conveyor; A control module for controlling the camera array to adjust to a target pitch angle for shooting based on the real-time belt surface inclination angle and the camera array pitch angle adjustment strategy; An image processing module for converting the original image captured by the camera array into a plurality of grayscale images, splicing and fusing the plurality of grayscale images to generate a panoramic image, performing edge enhancement processing on the panoramic image to obtain an enhanced image, extracting the closed contour of the connected domain from the enhanced image, and performing morphological closing operation fusion on adjacent closed contours that meet the first preset condition; and A detection module for calculating the area of the fused contour, the angle between the main axis direction of the fused contour and the running direction of the conveyor belt, and the average edge gradient of the fused contour, and determining that the conveyor belt is torn when the area of the fused contour, the angle and the average edge gradient meet the second condition.

[0006] In a possible design of the first aspect, the system further includes: a circularly polarized light source, and the installation position of the circularly polarized light source satisfies: , where is the position of the circularly polarized light source on the slide rail, is the belt width, and α is the light source divergence angle, to eliminate the reflection interference on the metal surface.

[0007] In a possible design of the first aspect, the pitch angle adjustment strategy of the camera array is: , where is the target inclination angle, is the real-time belt surface inclination angle, k is the proportionality coefficient, to ensure that the viewing angle covers the full width of the belt, is the current pitch angle of the camera array, is the target pitch angle.

[0008] In a possible design of the first aspect, the image processing module is specifically configured to: extract SIFT feature points of adjacent grayscale images and perform bidirectional matching, and use the RANSAC algorithm to eliminate mismatched points; solve the homography matrix after matching; and based on the homography matrix, apply a weighted fusion algorithm to eliminate the stitching gap and generate the panoramic image.

[0009] In a possible design of the first aspect, the image processing module is specifically configured to: perform gray mapping on the panoramic image to obtain a gray-level image; apply the Laplacian operator to perform second-order derivative edge detection on the gray-level image to obtain a second derivative; normalize the second derivative to obtain a normalized image; perform Gaussian smoothing on the normalized image to obtain a smoothed image; perform gray transformation on the smoothed image to enhance the edge region and suppress the non-edge region to obtain a transformed image; perform binarization on the transformed image to obtain a binary image; and based on connected component analysis of the binary image, extract the closed contour in the binary image.

[0010] In a possible design of the first aspect, the image processing module is further specifically configured to: calculate the closed contour area or the closed contour length-width ratio; and when the closed contour area or the closed contour length-width ratio meets the third preset condition, eliminate the noise contour.

[0011] In a possible design of the first aspect, the detection module is specifically configured to: Jointly calibrate the torn position of the conveyor belt through the encoder reference time, the real-time speed of the conveyor belt, and the image timestamp.

[0012] In a second aspect, a method for detecting a torn conveyor belt in a mine is provided. The method is based on the above system, and the method includes: Dynamically monitor the real-time belt surface inclination angle of the mine belt conveyor; based on the real-time belt surface inclination angle and the camera array pitch angle adjustment strategy, control the camera array to adjust to the target pitch angle for shooting; Convert the original images captured by the camera array into a number of grayscale images, splice and fuse the number of grayscale images to generate a panoramic image, perform edge enhancement processing on the panoramic image to obtain an enhanced image, extract the connected domain closed contours from the enhanced image, and perform morphological closing operation fusion on the adjacent closed contours whose distance meets the first preset condition; and Calculate the area of the fused contour, the angle between the main axis direction of the fused contour and the running direction of the conveyor belt, and the average edge gradient of the fused contour. When the area of the fused contour, the angle, and the average edge gradient meet the second condition, it is determined that the conveyor belt is torn.

[0013] In a third aspect, an electronic device is provided. The electronic device includes: a processor, and a memory coupled to the processor. The memory is used to store a computer program; the processor is used to execute the computer program stored in the memory so that the electronic device executes the method in any possible implementation manner of the second aspect.

[0014] In a fourth aspect, a computer-readable storage medium is provided, including a computer program or instruction. When the computer program or instruction runs on a computer, the computer is made to execute the method in any possible implementation manner of the second aspect.

[0015] In a fifth aspect, a computer program product is provided, including: a computer program or instruction. When the computer program or instruction runs on a computer, the computer is made to execute the method in any possible implementation manner of the second aspect.

[0016] Based on the above technical solution, a mine conveyor belt tear detection system including a camera array, a gyroscope, a control module, an image processing module, and a detection module is constructed. The camera array and sensors such as gyroscopes are arranged at appropriate positions. The image processing module performs grayscale conversion, stitching and fusion, and edge enhancement on a number of captured images, and performs corresponding contour extraction and fusion. Then, the detection module calculates the area, the angle with the running direction of the conveyor belt, the average edge gradient, etc. based on the fused contour. Thus, when the calculation results meet the conditions, it is determined that the conveyor belt is torn. In this way, the specific tear position of the conveyor belt can be detected in real time and at a fixed point, meeting the requirements of real-time and accurate detection and early warning, and thereby indirectly improving the production efficiency of the mine. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or related technical descriptions. Obviously, the following-described drawings are only some embodiments of the embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 is a flowchart of a visual detection method for a mine conveyor belt tear provided by an embodiment of the present application; Figure 2 is a schematic diagram of a mine conveyor device including a mine conveyor belt provided by an embodiment of the present application; Figure 3 is a schematic diagram of the relative position relationship between a mine conveyor belt, a roller, and a camera provided by an embodiment of the present application; Figure 4 is a schematic diagram of the relative position relationship between a roller and a camera provided by an embodiment of the present application; Figure 5 is a schematic diagram of the camera mounting structure provided by an embodiment of the present application; Figure 6 is a schematic diagram of the relative position relationship between a camera and a light source provided by an embodiment of the present application; Marking description: 1. Roller; 2. Belt; 3. Camera; 4. Roller frame; 5. Hinge; 6. Sliding base; 7. Camera slide rail; 8. Light source base; 9. Mobile light source. Detailed Embodiments

[0019] In order to make the objectives, technical solutions, and advantages of the present application clearer, the following further details the present application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] It should be noted that although the functional modules are divided in the schematic diagram of the device and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different module division in the device or a different order in the flowchart. The terms "first", "second", etc. in the specification, claims, and the above-mentioned drawings are used to distinguish similar objects and do not necessarily describe a specific order or sequence.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0022] An embodiment of this application provides a visual detection system and method for a torn mining conveyor belt, aiming to provide a solution that does not rely on large-scale training data, has strong adaptability, and high real-time performance.

[0023] ‌In the first aspect‌, as Figure 1 shown, the visual detection system for a torn mining conveyor belt includes: a camera array disposed at intervals between the rollers of the mining belt conveyor; a gyroscope for dynamically monitoring the real-time belt surface inclination angle of the mining belt conveyor; a control module for controlling the camera array to adjust to a target pitch angle for shooting based on the real-time belt surface inclination angle and the pitch angle adjustment strategy of the camera array; an image processing module for converting the original images captured by the camera array into a number of grayscale images, stitching and fusing the number of grayscale images to generate a panoramic image, performing edge enhancement processing on the panoramic image to obtain an enhanced image, extracting the closed contours of the connected regions from the enhanced image, and performing morphological closing operation fusion on the adjacent closed contours whose distance meets the first preset condition; and a detection module for calculating the area of the fused contour, the angle between the main axis direction of the fused contour and the running direction of the conveyor belt, and the average value of the edge gradients of the fused contour, and determining that the conveyor belt is torn when the area of the fused contour, the angle, and the average value of the edge gradients meet the second condition.

[0024] Specifically, industrial cameras are used and connected into a multi-camera stitching array to shoot the lower bottom surface of the conveyor belt of the mining belt conveyor.

[0025] The camera array can be disposed at intervals between the rollers of the mining belt conveyor, as Figures 2-5 shown.

[0026] Deploy a multi-camera array with adjustable angles at intervals between the rollers of the belt conveyor, and dynamically adjust the pitch angle of the camera according to the belt inclination angle to ensure complete coverage of the lower bottom surface of the belt. As Figure 6As shown, the camera can be slidably arranged on the slide rail through a movable base. As Figure 5 shown, the slide rails can be hinged through hinges. The camera is movably assembled with the movable base through a rotating structure. These components can all be provided with driving parts to act in a controlled manner.

[0027] In this embodiment, a light source emitter is also designed, which adopts a circularly polarized light source, and its installation position is adjacent to the camera position.

[0028] In the second aspect, as Figure 1 shown, the visual detection method for tearing of the mine conveyor belt is based on the above system, and this method includes: Step S1: Obtain pictures during the operation of the belt conveyor.

[0029] Specifically, based on the belt running speed, the image acquisition frequency of the industrial camera can be dynamically adjusted to capture the bottom surface images of multiple sections of the belt.

[0030] Step S2: Stitch the images obtained by the camera array.

[0031] Specifically, the image stitching method includes: extracting SIFT feature points of adjacent images and performing bidirectional matching. The specific extraction of features is to detect extreme points through a Gaussian difference pyramid to generate a 128-dimensional feature vector containing position / scale / orientation information; the bidirectional matching is specifically to first find the nearest neighbor in image B based on the features of image A, and then verify in the reverse direction to ensure the uniqueness of the matching; the RANSAC algorithm is used to eliminate mismatched points and solve the homography matrix H. Specifically, randomly select 4 pairs of matching points to calculate the initial homography matrix H, calculate the reprojection error of all matching points after being transformed by H, retain the inliers with an error less than the threshold (3 pixels), iterate 2500 times, and select the homography matrix H with the most inliers as the final required homography matrix; based on the finally selected homography matrix through iteration, the weighted fusion algorithm is applied to eliminate the stitching gap. Specifically, a linear gradient weight is used to smoothly transition from the edge area to the center, and at the same time, multi-band fusion is used to fuse different frequency bands of the image pyramid respectively to avoid misalignment of low-frequency information and generate a seamless panoramic image.

[0032] Step S3: Adopt geometric correction technology to convert the obtained curved surface image into a flat surface.

[0033] Step S4: Convert the image into a grayscale image, take the second derivative of the grayscale image to obtain the details of the image contour.

[0034] Specifically, normalization and Gaussian smoothing: After mapping the image to the range of [0, 255], the conversion of the grayscale image is completed. The specific formula is: , where I is the pixel matrix of the original image, containing the grayscale values of all pixels, Grayscale is the grayscale conversion operation, The converted image has a value range of [0, 255]. The specific operation of the function is , where Min(I) is the minimum gray value of the image pixels (i.e., the minimum value in the matrix), and max(I) is the maximum gray value of the image pixels (i.e., the maximum value in the matrix). And Gaussian smoothing is performed, and the Gaussian kernel adopted is , where σ is the smoothing intensity, and x, y are the pixel positions in the digital image.

[0035] Second-order derivative edge detection: For the grayscale image after Gaussian filtering The following Laplacian operator is applied to obtain the second-order derivative contour details: , where i and j are the offsets of x and y, and K is the Laplacian kernel.

[0036] Step S5: Perform gray-scale transformation on the second-order derivative contour detail map, and use median filtering to smooth the contour details.

[0037] Specifically, gray-scale transformation enhancement: Piecewise linear transformation is adopted to enhance the edge contrast: , where k is the control slope, θ is the threshold, S is the second-order contour detail map, T(S) is the piecewise linear transformation function, and E is the image after gray-scale change. The sigmoid function is used to enhance the edge region (high values are retained) and suppress the non-edge region (low values are compressed).

[0038] Before gray-scale transformation enhancement, corresponding normalization and Gaussian filtering can also be performed.

[0039] Normalize the second-order derivative contour details: , map the second-order derivative to [0, 255] for subsequent operations, min( ) is the minimum gray value of the image pixels, max( ) is the maximum gray value of the image pixels, L is the second-order contour detail map, and N( ) is the result after normalization.

[0040] Gaussian kernel: .

[0041] Step S6: Perform binarization processing on the image and find the contours.

[0042] Specifically, C = FindCountours(B) extracts the total number of closed contours in the binary image based on connected component analysis , B is the digital image after binarization operation, FindContours() represents the operation of finding contours, and the specific operation is to adopt topological contour analysis using the Suzuki-Abe contour algorithm.

[0043] Step S7: Through contour analysis and region fusion, screen the contours that meet the tearing characteristics.

[0044] Specifically, the contour fusion method includes filtering small contours using the following formula: , where Area( ) represents the total set of all recognized contours, represents the contour area threshold, Φ represents the threshold screening mechanism, removing noise contours with an area smaller than the threshold or an aspect ratio greater than the threshold, and bounding all the remaining contours. Based on the morphological closing operation, merge the remaining bounded adjacent small contours with a distance less than to form a complete edge. The merging uses the following algorithm: . Among them, calculate the contour area and the aspect ratio , and eliminate or noise contours, where represents the distance threshold, represents the aspect ratio threshold.

[0045] Step S8: Check the constraint conditions: 1. Using HSL color space analysis, the lightness value (L) of the crack area should be lower than the reference value by 35% ± 5% (the reference value is the average of the surrounding normal areas); 2. Based on morphological feature detection, ensure that the continuous length of the contour ≥ 300 mm (filter out small scratches shorter than the idler spacing); 3. The color saturation (S) needs to be < 15%, conforming to the low saturation characteristics caused by ore ash deposition; 4. The standard deviation of the main axis angle < 8° (ensuring the linear extension feature); 5. The fractal dimension < 1.25 (excluding stains with dendritic diffusion).

[0046] Step S9: Determine whether there are conforming contours. If so, execute Step S10; otherwise, execute Step S1.

[0047] Specifically, the conditions for triggering an alarm are that all of the following are satisfied simultaneously: Condition 1: The area of the fused contour ; where is the area of the fused contour, is the pre-set contour area threshold.

[0048] Condition 2: The angle between the main axis direction of the contour and the running direction of the belt ; where is the minimum threshold of the set angle, is the maximum threshold.

[0049] Condition 3: The average value of the contour edge gradient ; where is the actual contour edge gradient, is the set gradient threshold.

[0050] Step S10: Alarm and record the tearing position, and stop the belt conveyor.

[0051] Specifically, when recording the tearing position, the tearing position is jointly calibrated by the encoder and the image timestamp: , where is the real-time speed of the belt, is the encoder reference time, is the alarm trigger time.

[0052] In specific applications, the specific method for dynamically adjusting the camera pitch angle includes: real-time monitoring of the belt conveyor inclination angle by the gyroscope, generating an angle compensation signal to drive the hinge motor, and adjusting the pitch angle of the camera array to the target angle , satisfying the following formula constraint: . Among them, is the target coverage angle, is the actually measured inclination angle by the gyroscope, k is the proportionality coefficient, ensuring that the viewing angle covers the full width of the belt, is the current pitch angle of the camera.

[0053] The light source emitter uses a circularly polarized light source, and its installation position is adjacent to the camera position, and it can be automatically moved to a position that satisfies: Among them, is the position of the circularly polarized light source on the slide rail, is the belt width, α is the light source divergence angle, to eliminate the reflective interference on the metal surface.

[0054] The advantages of the embodiments of the present application are mainly as follows: 1. Environmental robustness: Suppress the reflective interference on the metal surface through the circularly polarized light source, and combine the dynamic threshold algorithm to automatically adapt to complex working conditions such as oil stains and dust on the belt surface, effectively eliminate image noise, improve the stability and accuracy of tearing detection, and still maintain a detection rate of more than 95% under extreme conditions where the stain coverage rate reaches 30%.

[0055] 2. Inclination universality: Adopt the hinge-type camera array and the gyroscope inclination feedback technology to adjust the camera pitch angle in real time (covering the inclination range of -30° to +45°), and can adapt to the complex belt conveyor layout of underground inclination and undulation without manual intervention, reducing the equipment installation and commissioning time by more than 80%.

[0056] 3. Real-time performance: Through the multi-level image processing pipeline and the FPGA hardware acceleration technology, compress the full process delay of image acquisition, stitching, and analysis to within 200 ms, which is 5 times faster than the traditional CPU solution, ensuring that an alarm can be triggered in real time when the belt runs at high speed (5 m / s), and avoiding production accidents caused by the expansion of tears.

[0057] 4. Precise positioning: Based on the encoder pulse synchronization and image timestamp matching technology, combined with the belt speed integration algorithm, the calibration error of the tear position is less than ±5 cm, and the accuracy is improved by 60% compared with the traditional vision positioning method, significantly shortening the troubleshooting time of maintenance personnel.

[0058] The embodiment of the present application also provides a method for detecting the tear of a mine conveyor belt. The method is based on the above system, and the method includes: Dynamically monitoring the real-time belt surface inclination angle of the mine belt conveyor; based on the real-time belt surface inclination angle and the camera array pitch angle adjustment strategy, controlling the camera array to adjust to the target pitch angle for shooting; Converting the original images taken by the camera array into a number of grayscale images, splicing and fusing the number of grayscale images to generate a panoramic image, performing edge enhancement processing on the panoramic image to obtain an enhanced image, extracting the connected domain closed contour from the enhanced image, and performing morphological closing operation fusion on the adjacent closed contours whose distance meets the first preset condition; and Calculating the area of the fused contour, the angle between the main axis direction of the fused contour and the running direction of the conveyor belt, and the average value of the edge gradients of the fused contour. When the area of the fused contour, the angle and the average value of the edge gradients meet the second condition, it is determined that the conveyor belt is torn.

[0059] The embodiment of the present application also provides an electronic device, including: a processor, and a memory coupled to the processor. The memory is used to store a computer program; the processor is used to execute the computer program stored in the memory so that the electronic device executes the method described in any one of the above embodiments.

[0060] The electronic device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The electronic device may include, but is not limited to, a processor and a memory.

[0061] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects all parts of the entire device using various interfaces and lines.

[0062] The memory can be used to store the computer program. By running or executing the computer program stored in the memory and calling the data stored in the memory, the processor realizes various functions of the electronic device.

[0063] The memory may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as hard disks, memory, plug-in hard disks, Smart Media Cards (SMCs), Secure Digital (SD) cards, Flash Cards, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0064] The embodiments of the present application also provide a storage medium. The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0065] The embodiment of the present application also provides a computer program product, including: a computer program or instruction, when the computer program or instruction runs on a computer, enabling the computer to execute the method according to any one of the above possible implementation manners.

[0066] The above is the preferred implementation manner of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present application.

Claims

1. A mine conveyor belt tear detection system, characterized in that, The system includes: A camera array disposed at intervals of the idlers of the mine belt conveyor; A gyroscope for dynamically monitoring the real-time belt surface inclination angle of the mine belt conveyor; A control module for controlling the camera array to adjust to a target pitch angle for shooting based on the real-time belt surface inclination angle and the camera array pitch angle adjustment strategy; An image processing module for converting the original images captured by the camera array into a plurality of grayscale images, mosaicking and fusing the plurality of grayscale images to generate a panoramic image, performing edge enhancement processing on the panoramic image to obtain an enhanced image, extracting the closed contours of the connected regions from the enhanced image, and performing morphological closing operation fusion on the adjacent closed contours whose distance meets the first preset condition; and A detection module for calculating the area of the fused contour, the angle between the main axis direction of the fused contour and the running direction of the conveyor belt, and the average value of the edge gradients of the fused contour, and determining that the conveyor belt is torn when the area of the fused contour, the angle and the average value of the edge gradients meet the second condition.

2. The system according to claim 1, wherein The system further includes: a circularly polarized light source, and the installation position of the circularly polarized light source satisfies: , where is the position of the circularly polarized light source on the slide rail, is the belt width, and α is the light source divergence angle, so as to eliminate the reflection interference on the metal surface.

3. The system according to claim 1, characterized in that, The pitching angle adjustment strategy of the camera array is as follows: , where is the target inclination angle, is the real-time belt surface inclination angle, k is the proportionality coefficient to ensure that the viewing angle covers the full width of the belt, is the current pitching angle of the camera array, is the target pitching angle.

4. The system according to claim 1, wherein Specifically, the image processing module is used for: extracting SIFT feature points of adjacent grayscale images and performing bidirectional matching, and using the RANSAC algorithm to eliminate mis-matched points; Solving the homography matrix after matching; And Based on the homography matrix, applying a weighted fusion algorithm to eliminate the stitching gap and generate the panoramic image.

5. The system according to claim 1, wherein Specifically, the image processing module is used for: Performing grayscale mapping on the panoramic image to obtain a grayscale level image; Applying a Laplace operator to perform second-order derivative edge detection on the grayscale level image to obtain a second derivative; Normalizing the second derivative to obtain a normalized image; Performing Gaussian smoothing on the normalized image to obtain a smoothed image; Performing grayscale transformation on the smoothed image to enhance the edge region and suppress the non-edge region to obtain a transformed image; Performing binarization on the transformed image to obtain a binarized image; And Based on connected region analysis of the binarized image, extracting the closed contours in the binarized image.

6. The system according to claim 5, wherein Specifically, the image processing module is further used for: Calculating the area of the closed contour or the aspect ratio of the length and width of the closed contour; and When the area of the closed contour or the aspect ratio of the length and width of the closed contour meets the third preset condition, eliminating the noise contours.

7. The system according to claim 1, wherein Specifically, the detection module is used for: Jointly calibrating the tearing position of the conveyor belt through the encoder reference time, the real-time speed of the conveyor belt and the image timestamp.

8. A method for detecting the tearing of a mine conveyor belt, characterized in that, The method is based on the system according to any one of claims 1-7, and the method includes: Dynamically monitoring the real-time belt surface inclination angle of the mine belt conveyor; based on the real-time belt surface inclination angle and the camera array pitch angle adjustment strategy, controlling the camera array to adjust to a target pitch angle for shooting; Converting the original images captured by the camera array into a plurality of grayscale images, mosaicking and fusing the plurality of grayscale images to generate a panoramic image, performing edge enhancement processing on the panoramic image to obtain an enhanced image, extracting the closed contours of the connected regions from the enhanced image, and performing morphological closing operation fusion on the adjacent closed contours whose distance meets the first preset condition; and Calculate the area of the fused contour, the angle between the major axis direction of the fused contour and the running direction of the conveyor belt, and the average edge gradient of the fused contour. When the area of the fused contour, the angle, and the average edge gradient meet the second condition, it is determined that the conveyor belt is torn.

9. An electronic device, characterized in that, The electronic device includes: a processor, and a memory coupled to the processor, The memory is used to store computer programs; and The processor is configured to execute the computer program stored in the memory, so that the electronic device executes the method as claimed in claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes computer programs or instructions. When the computer programs or instructions are run on a computer, the computer is caused to execute the method as claimed in claim 8.

Images