Mine tank ear fault detection method and device based on video analysis

Abstract

This application proposes a method and apparatus for detecting mine car ear faults based on video analysis. The method includes: acquiring image information of a mine car ear; controlling an image segmentation algorithm to segment the image information to obtain the car ear's location information and contour map; comparing the contour maps of adjacent frames to obtain a differential contour image; and determining whether the car ear corresponding to the location information is faulty based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold. Thus, based on the image information of the mine car ear, the location information and contour map of the car ear are determined to accurately determine whether the car ear is faulty, achieving precise detection of mine car ear faults and ensuring the safe operation of mine car ears.

Description

Technical Field

[0001] This application relates to the field of mine canister detection technology, and in particular to a method and apparatus for detecting mine canister faults based on video analysis. Background Technology

[0002] Currently, mine hoisting systems play a vital role in coal mine production, responsible for transporting personnel and materials during mining operations. Hoisting systems commonly utilize roller lugs to guide, buffer, and stabilize the hoisting cage. However, during high-speed cage operation, the rubber wheels and buffer devices of the lugs are subjected to significant loads and resistance. After prolonged use, this can lead to lugs failing to rotate or even detaching. If multiple lugs simultaneously fail to rotate and are not detected promptly, continued operation can cause severe friction between the hoisting cage and the guide rail due to operational oscillations, damaging the shaft equipment. Therefore, a video analysis-based method for detecting mine lug failures is urgently needed to ensure safe mine production. Summary of the Invention

[0003] This application proposes a method and apparatus for detecting mine drum ear faults based on video analysis.

[0004] The first aspect of this application proposes a method for detecting mine car ear faults based on video analysis. The method includes: acquiring image information of a mine car ear; controlling an image segmentation algorithm to segment the image information to obtain the location information and contour map of the car ear; comparing the differences between the contour maps of two adjacent frames to obtain a differential contour image; and determining whether the car ear corresponding to the location information is faulty based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold.

[0005] In one embodiment of this application, the control image segmentation algorithm segments the image information to obtain the location information and contour map of the can ear, including: performing feature segmentation on the can ear image in the image information to obtain multiple feature image blocks; clustering and locating the multiple feature image blocks according to the similarity measure of the multiple feature image blocks to obtain an initial can ear contour map and location information; and denoising the initial can ear contour map to obtain a can ear contour map of the can ear contour map.

[0006] In one embodiment of this application, the step of denoising the initial can ear contour image to obtain a can ear contour image of the can ear contour image includes: scanning the initial can ear contour image to obtain each pixel of the initial can ear contour image; using the weighted average gray value of adjacent pixels as the denoised pixel of the initial can ear contour image; and generating a can ear contour image of the can ear based on the denoised pixel.

[0007] In one embodiment of this application, determining whether the can ear corresponding to the location information is faulty based on the difference between the pixel value in the differential contour image and a preset pixel threshold includes: binarizing the differential contour image to obtain a binarized difference image corresponding to the differential contour image; taking pixels in the binarized difference image whose pixel value is greater than the preset pixel threshold as foreground moving pixels; constructing a pixel motion region corresponding to the foreground moving pixels based on the foreground moving pixels; and determining whether the can ear under the location information is faulty based on the pixel motion region.

[0008] This application proposes a video analysis-based method for detecting mine car ear faults. The method acquires image information of the mine car ear, uses an image segmentation algorithm to segment the image information to obtain the car ear's location information and contour map, compares the contour maps of adjacent frames to obtain a differential contour image, and determines whether the car ear corresponding to the location information is faulty based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold. Thus, based on the image information of the mine car ear, the method determines the car ear's location information and contour map, accurately determining whether the car ear is faulty, achieving precise detection of mine car ear faults, and ensuring the safe operation of mine car ears.

[0009] A second aspect of this application proposes a mine shaft auger fault detection device based on video analysis. The device includes: an acquisition module for acquiring image information of a mine shaft auger; a segmentation module for controlling an image segmentation algorithm to segment the image information to obtain the location information and contour map of the auger; a comparison module for comparing the contour maps of the auger in two adjacent frames to obtain a differential contour image; and a determination module for determining whether the auger corresponding to the location information is faulty based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold.

[0010] In one embodiment of this application, the segmentation module includes: a segmentation unit, configured to perform feature segmentation on the can ear image in the image information to obtain multiple feature image blocks; a generation unit, configured to cluster and locate the multiple feature image blocks according to a similarity metric, to obtain an initial can ear contour map and location information; and a denoising unit, configured to denoise the initial can ear contour map to obtain a can ear contour map of the can ear contour map.

[0011] In one embodiment of this application, the denoising unit is specifically configured to: scan the initial can ear contour map to obtain each pixel of the initial can ear contour map; use the weighted average gray value of adjacent pixels as the denoised pixel of the initial can ear contour map; and generate the can ear contour map of the can ear based on the denoised pixel.

[0012] In one embodiment of this application, the determining module is specifically used for: binarizing the differential contour image to obtain a binarized difference image corresponding to the differential contour image; taking pixels whose pixel values ​​in the binarized difference image are greater than the preset pixel threshold as foreground moving pixels; constructing a pixel motion region corresponding to the foreground moving pixels based on the foreground moving pixels; and determining whether the can ear under the position information is faulty based on the pixel motion region.

[0013] In one embodiment of this application, the determining module is specifically used for: binarizing the differential contour image to obtain a binarized difference image corresponding to the differential contour image; taking pixels whose pixel values ​​in the binarized difference image are greater than the preset pixel threshold as foreground moving pixels; constructing a pixel motion region corresponding to the foreground moving pixels based on the foreground moving pixels; and determining whether the can ear under the position information is faulty based on the pixel motion region.

[0014] This application proposes a mine shaft auger fault detection device based on video analysis. It acquires image information of the mine shaft auger, controls an image segmentation algorithm to segment the image information to obtain the location information and contour map of the auger. It compares the contour maps of adjacent frames to obtain a differential contour image. Based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold, it determines whether the auger corresponding to the location information is faulty. Thus, based on the image information of the mine shaft auger, the location information and contour map of the auger are determined to accurately determine whether the auger is faulty, achieving precise detection of mine shaft auger faults and ensuring the safe operation of mine shaft augers.

[0015] A third aspect of this application provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the video analysis-based mine drum ear fault detection method of this application.

[0016] The fourth aspect of this application provides a computer program product that, when executed by an instruction processor, implements the video analysis-based mine drum ear fault detection method of this application.

[0017] Other effects of the above-mentioned alternative methods will be described below in conjunction with specific embodiments. Attached Figure Description

[0018] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0019] Figure 1 This is a flowchart illustrating a video analysis-based method for detecting mine drum ear faults, as provided in an embodiment of this application.

[0020] Figure 2 This is a schematic diagram of the structure of a mine tank ear fault detection device based on video analysis provided in an embodiment of this application;

[0021] Figure 3 This is a flowchart illustrating another video analysis-based method for detecting mine drum ear faults provided in an embodiment of this application.

[0022] Figure 4 This is a flowchart illustrating another video analysis-based method for detecting mine drum ear faults provided in an embodiment of this application.

[0023] Figure 5 This is a flowchart illustrating another video analysis-based method for detecting mine drum ear faults provided in an embodiment of this application.

[0024] Figure 6 This is a flowchart illustrating another video analysis-based method for detecting mine drum ear faults provided in an embodiment of this application.

[0025] Figure 7 This is a flowchart of the algorithm module of a video analysis-based mine tank ear fault detection method provided in an embodiment of this application;

[0026] Figure 8 This is a schematic diagram of the structure of a mine tank ear fault detection device based on video analysis provided in an embodiment of this application;

[0027] Figure 9 This is a schematic diagram of another mine shaft ear fault detection device based on video analysis provided in the embodiments of this application. Detailed Implementation

[0028] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0029] The following describes a video analysis-based method and apparatus for detecting mine drum ear faults according to embodiments of this application, with reference to the accompanying drawings.

[0030] Figure 1 This is a flowchart illustrating a video analysis-based method for detecting mine canister malfunctions according to an embodiment of this application. It should be noted that the executing entity of this video analysis-based method for detecting mine canister malfunctions is a video analysis-based mine canister malfunction detection device. This device can be implemented in software and / or hardware. In this embodiment, the video analysis-based mine canister malfunction detection device can be configured in an electronic device. The electronic device in this embodiment may include a server, but this embodiment does not specifically limit the type of electronic device.

[0031] Figure 1 This is a flowchart illustrating a video analysis-based method for detecting mine drum ear faults, as provided in an embodiment of this application.

[0032] like Figure 1 As shown, the video analysis-based method for detecting mine canister ear faults may include:

[0033] Step 101: Obtain image information of the ear of the mine.

[0034] In some embodiments, the image information of the mine shaft can be image information collected by a monitoring device installed near the mine shaft, wherein the monitoring device can be a camera, but not limited thereto, and this embodiment does not specifically limit it.

[0035] It is understandable that, in order to improve the accuracy and coverage of image information, the monitoring equipment near the cage ear in the mine can be set on the top of the cage corresponding to the cage ear, at a distance of 1 meter above the cage ear and at an angle of 60 degrees, so as to facilitate the monitoring equipment to monitor and collect image information of the cage.

[0036] In other embodiments, to improve the clarity of image information, two sets of DC mine explosion-proof floodlights can be installed 2 meters above the tank ear at an angle of 60 degrees to provide supplementary lighting for the camera, thereby ensuring the image quality of the image information collected by the monitoring equipment.

[0037] Step 102: Control the image segmentation algorithm to segment the image information to obtain the location information and outline of the jar ear.

[0038] In some embodiments, the image segmentation algorithm may be a deep learning segmentation algorithm, but it is not limited thereto, and this embodiment does not specifically limit it.

[0039] In other embodiments, the image information may include an image of the can ear, the location information of the can ear, but is not limited to this.

[0040] One approach is to process the image of the can ear using a deep learning segmentation algorithm to segment out the can ear contour map. Specifically, the image segmentation algorithm can adopt a two-stage structure. First, a first-order network is used to find a Region Proposal Network (RPN). Then, each can ear contour RoI found by the RPN is classified, located, and a binary mask is found to obtain the location information and contour map of the can ear.

[0041] The location information of the can ear is used to mark the location of the can ear in the mine, which facilitates the management of the can ear in the mine.

[0042] Step 103: Compare the differences between the ear contour images of two adjacent frames to obtain a differential contour image.

[0043] In some embodiments, comparing the differences between the ear contour images of two adjacent frames to obtain a differential contour image can be implemented by taking the frame difference between the previous frame and the current frame of the ear image and retaining the difference between the two frames to use the difference between the two frames as a differential contour image, but it is not limited to this.

[0044] To further ensure the accuracy of canister ear fault detection, multi-frame difference comparison can be performed. The number of frames can be set randomly, and the threshold for frame difference can also be adjusted. It is advisable to adjust it to the optimal value according to the actual business scenario.

[0045] Step 104: Based on the difference between the pixel values ​​in the differential contour image and the preset pixel threshold, determine whether the can ear corresponding to the location information is faulty.

[0046] In some embodiments, the difference between the pixel values ​​in the differential contour image and a preset pixel threshold can be used to determine whether the differential contour image belongs to a motion region. Thus, based on the motion region corresponding to the differential contour images of multiple frame differences, it can be determined whether the can ear corresponding to the location information is faulty, thereby achieving accurate detection of the can ear.

[0047] Specifically, if the differential contour image does not belong to the moving region, it can be considered as background pixels. If the differential contour image belongs to the moving region, the movement of the can ear can be determined based on the moving regions corresponding to multiple differential contour images. If the can ear is moving, it is rotating and working without fault. Conversely, the can ear is in a faulty state.

[0048] This application proposes a video analysis-based method for detecting mine car ear faults. The method acquires image information of the mine car ear, uses an image segmentation algorithm to segment the image information to obtain the car ear's location information and contour map, compares the contour maps of adjacent frames to obtain a differential contour image, and determines whether the car ear corresponding to the location information is faulty based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold. Thus, based on the image information of the mine car ear, the method determines the car ear's location information and contour map, accurately determining whether the car ear is faulty, achieving precise detection of mine car ear faults, and ensuring the safe operation of mine car ears.

[0049] Furthermore, it is understood that, in order to clearly describe the video analytics-based mine canister ear fault detection method, this method can be applied to the industrial control computer in the mine, such as... Figure 2 As shown in the figure, this application embodiment also provides a structural schematic diagram of a mine can ear fault detection device based on video analysis. Specifically, the mine can ear fault detection device based on video analysis may include can ear 1, can ear 2, can guideway 3, monitoring equipment 4, base station 5, and industrial control computer 6. During the normal operation of the mine can ear, can ear 1 and 2 are continuously attached and rotated along the can guideway 3. The monitoring equipment 4 continuously records the operation of can ear 1 and 2. The monitoring equipment 4 can be a pan-tilt camera, and the pan-tilt angle can be adjusted remotely to enable the recording of other groups of can ears, thereby achieving all-round monitoring of the mine can ears.

[0050] On the other hand, such as Figure 2 As shown, when the image information of the mine shaft canister is detected, the PTZ camera 4 connects to the industrial control computer 6 via the base station 5 and wireless communication Wi-Fi. The PTZ camera 4 then transmits the image information of the canisters 1 and 2 captured by the camera to the algorithm module of the industrial control computer 6 as a video stream for fault detection and analysis to determine whether the mine shaft canister is faulty.

[0051] To clearly understand this application, the following will be combined with... Figure 3 The processing procedure of the video analysis-based mine shaft ear fault detection method is described exemplarily, wherein this embodiment is a further refinement or extension of the above embodiment.

[0052] like Figure 3 As shown, the video analysis-based method for detecting mine canister ear faults may include:

[0053] Step 301: Obtain image information of the ear of the mine shaft.

[0054] It should be noted that the specific implementation of step 301 can be found in the relevant description in the above embodiments.

[0055] Step 302: Perform feature segmentation on the ear image in the image information to obtain multiple feature image blocks.

[0056] In some embodiments, the number of can ear images can be at least several hundred to ensure the accuracy of the can ear fault detection method.

[0057] In other embodiments, feature segmentation can be performed based on whether the region of the original image corresponding to each pixel in the can ear image contains the can ear outline, thereby obtaining multiple feature image blocks corresponding to the can ear outline.

[0058] Step 303: Based on the similarity measure of multiple feature image blocks, cluster and locate multiple feature image blocks to obtain the initial outline map and location information of the ear of the jar.

[0059] Step 304: Denoise the initial jar ear outline to obtain the jar ear outline of the jar ear outline.

[0060] In some embodiments, one implementation of denoising the initial can ear contour image to obtain a can ear contour image of another can ear contour image is to use Gaussian filtering to denoise the initial can ear contour image. Specifically, the initial can ear contour image is scanned to obtain each pixel of the initial can ear contour image, and the weighted average gray value of adjacent pixels is used as the denoised pixel of the initial can ear contour image. Based on the denoised pixel, the can ear contour image of another can ear is generated.

[0061] For example, a template can be used to scan each pixel of the initial jar ear outline image, and the value of the center pixel of the template can be replaced by the weighted average gray value of the pixels in the neighborhood determined by the template, so as to serve as the noise-reducing pixel, thereby generating the jar ear outline image.

[0062] The template mentioned above can be a convolution or a mask, but is not limited to these. This embodiment does not specifically limit it.

[0063] Step 305: Compare the differences between the ear contour images of two adjacent frames to obtain a differential contour image.

[0064] Step 306: Based on the difference between the pixel values ​​in the differential contour image and the preset pixel threshold, determine whether the can ear corresponding to the location information is faulty.

[0065] This application proposes a video analysis-based method for detecting mine car ear faults. The method acquires image information of mine car ears, performs feature segmentation on the car ear images to obtain multiple feature image blocks, clusters and locates these blocks based on similarity metrics to obtain an initial car ear contour map and location information, denoises the initial car ear contour map to obtain a car ear contour map of the car ear contour map, compares the differences between adjacent frames of car ear contour maps to obtain a differential contour image, and determines whether the car ear corresponding to the location information is faulty based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold. Therefore, by clustering and locating the image information of mine car ears, the method accurately determines the location information and contour map of the car ear, thus improving the speed of car ear fault location, reducing car ear fault repair time, and increasing the working efficiency of car ears.

[0066] Figure 4 This is a schematic diagram of the structure of a mine shaft ear fault detection device based on video analysis provided in an embodiment of this application.

[0067] Step 401: Obtain image information of the ear of the mine.

[0068] Step 402: Control the image segmentation algorithm to segment the image information to obtain the location information and outline of the jar ear.

[0069] Step 403: Compare the differences between the ear contour images of two adjacent frames to obtain a differential contour image.

[0070] It should be noted that the specific implementation methods of steps 401 to 403 can be found in the relevant descriptions in the above embodiments.

[0071] Step 404: Binarize the difference contour image to obtain the binarized difference image corresponding to the difference contour image.

[0072] In some embodiments, by setting the grayscale value of the pixels on the differential contour image to 0 or 255, the entire differential contour image presents a visual effect of only black and white, serving as the binarized difference image corresponding to the differential contour image.

[0073] Step 405: Pixels whose pixel values ​​in the binarized difference image are greater than a preset pixel threshold are taken as foreground moving pixels.

[0074] In some embodiments, when the light source corresponding to the binarized difference image is constant and the position of the rubber ring contour is constant, if the pixel value of a pixel in the binarized difference image is less than or equal to a preset pixel threshold, it is considered to be a background pixel; if the pixel value of a pixel in the binarized difference image is greater than the preset pixel threshold, it is considered to be a pixel caused by a moving object and is used as a foreground moving pixel.

[0075] The preset pixel threshold can be set to 180, but it is not limited to this. The specific preset pixel threshold can be set by relevant technical personnel according to actual business needs. This embodiment does not make specific limitations on this.

[0076] Step 406: Based on the foreground moving pixels, construct the pixel motion region corresponding to the foreground moving pixels.

[0077] In some embodiments, a pixel motion region is constructed based on the foreground moving pixels and calibrated to determine whether the can ear is rotating based on the calibrated pixel motion region, thereby determining whether the can ear is faulty.

[0078] Step 407: Determine whether the ear of the can is faulty based on the pixel motion area and the position information.

[0079] In some embodiments, if the pixel motion area changes significantly, exceeding a certain change threshold, it indicates that the ear is rotating normally, and the ear under that position information is operating normally. If the pixel motion area changes slightly, and is less than a certain change threshold, it indicates that the ear is rotating abnormally, and the ear under that position information is faulty.

[0080] This application proposes a video analysis-based method for detecting mine car ear faults. The method acquires image information of the mine car ear, uses an image segmentation algorithm to segment the image information to obtain the car ear's location information and contour map, compares the contour maps of adjacent frames to obtain a differential contour image, binarizes the differential contour image to obtain a corresponding binarized differential image, and identifies pixels in the binarized differential image whose pixel values ​​are greater than a preset pixel threshold as foreground moving pixels. Based on the foreground moving pixels, a pixel motion region is constructed corresponding to each foreground moving pixel. Based on the pixel motion region, the method determines whether the car ear is faulty under the given location information. Thus, based on the image information of the mine car ear, a differential contour map of the car ear is obtained, and based on the pixel motion region corresponding to the differential contour image, the method accurately determines whether the car ear is faulty, achieving precise detection of mine car ear faults and improving the service life of the car ear.

[0081] Figure 5 This is a flowchart illustrating another video analysis-based method for detecting mine drum ear faults, as provided in an embodiment of this application.

[0082] Step 501: Obtain image information of the ear of the mine shaft.

[0083] Step 502: Control the image segmentation algorithm to segment the image information to obtain the position information and outline of the jar ear.

[0084] Step 503: Compare the differences between the ear contour images of two adjacent frames to obtain a differential contour image.

[0085] Step 504: Based on the difference between the pixel values ​​in the differential contour image and the preset pixel threshold, determine whether the can ear corresponding to the location information is faulty.

[0086] It should be noted that the specific implementation methods of steps 501 to 504 can be found in the relevant descriptions in the above embodiments.

[0087] Step 505: If there is a faulty can ear under the location information, save the fault video corresponding to the faulty can ear to generate alarm information corresponding to the fault video.

[0088] In some embodiments, the fault video corresponding to the faulty tank ear can be obtained from the monitoring equipment monitoring the mine tank ear, but it is not limited to this.

[0089] Understandably, if there is a faulty can ear under the location information, the fault video corresponding to the faulty can ear is saved. Different fault videos correspond to different alarm information. For example, if the fault video shows that there are some oil stains and dirt on the surface of the can ear rubber ring causing the can ear to malfunction, the corresponding dirt information is generated to facilitate maintenance by the management personnel.

[0090] In other embodiments, if there is no faulty tank ear under the location information, the tank ear operates normally and there is no need to save the monitoring information of the tank ear.

[0091] Step 506: Generate a control signal to regulate the alarm information based on the alarm information.

[0092] In some embodiments, the number of tank ear failures in the alarm information affects the control signal of the alarm information. For example, if there is only one tank ear failure in the alarm information, the control signal corresponding to the alarm information is to save the fault video of the faulty tank ear. If there are two or more tank ear failures in the alarm information, the control signal corresponding to the alarm information is a shutdown maintenance command, and the fault videos of all faulty tank ears are saved for maintenance of the mine tank ear.

[0093] Step 507: Send the control signal to the system control platform corresponding to the can ear, and regulate the can ear through the system control platform.

[0094] Understandably, the system control platform adjusts the can ears differently based on different control signals. Specifically, when multiple can ears malfunction simultaneously, the system control platform stops the can ears and puts them into maintenance mode until maintenance is completed, then restarts them and tests the can ears after maintenance until no faults are detected.

[0095] This application proposes a video analysis-based method for detecting mine drum ear faults. The method acquires image information of the drum ear in the mine, uses an image segmentation algorithm to segment the image information to obtain the location information and contour map of the drum ear, compares the contour maps of adjacent frames to obtain a differential contour image, and determines whether the drum ear corresponding to the location information is faulty based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold. If a faulty drum ear is found at the location information, the corresponding fault video is saved to generate alarm information. Based on the alarm information, a control signal for regulating the alarm information is generated and sent to the system control platform corresponding to the drum ear. The system control platform then regulates the drum ear. Therefore, based on the location information and contour map of the drum ear in the mine, when a fault is determined, the system control platform corresponding to the drum ear regulates the drum ear, achieving safe operation of the mine drum ear and improving mine production efficiency.

[0096] In summary, to better understand the video analytics-based method for detecting mine canister ear faults, this method can be applied to industrial control computers in mines, such as... Figure 6 As shown, Figure 6 This is a flowchart illustrating another video analysis-based fault detection method for mine shaft canisters provided in this application embodiment. Specifically, multi-frame image information of the canister is collected by a PTZ camera corresponding to the canister and sent to the algorithm module of the industrial control computer for canister detection algorithm analysis and processing. When the canister rotates normally, continuous detection is performed. If the canister rotates abnormally, an alarm is triggered and an alarm message is generated. At the same time, the fault video corresponding to the alarm message is recorded. Based on the fault video, it is determined whether multiple canisters rotate abnormally. If there are no multiple canisters rotating abnormally, continuous detection is performed. If multiple canisters rotate abnormally, the system control platform is notified and maintenance is carried out to ensure the safe operation of the mine shaft canisters.

[0097] The process of processing the image information of the can ear by the algorithm module of the aforementioned industrial control computer can be as follows: Figure 7 As shown, Figure 7This is a flowchart of the algorithm module for a mine shaft auger fault detection method based on video analysis, provided in an embodiment of this application. Specifically, multiple frames of image information are input into the algorithm module to locate the position information and initial outline of the auger. Gaussian filtering is applied to the initial outline of the auger to denoise it, resulting in a more accurate outline. Frame difference comparison is performed between adjacent frames of the auger outline to obtain the frame difference difference portion. The pixel values ​​of the frame difference difference portion are used to determine whether the auger is abnormal. If the auger is normal, the next image information is input for processing. If the auger is abnormal, an alarm message is generated, and the fault video corresponding to the alarm message is recorded to generate the control signal corresponding to the fault video. Finally, the next image information is used for detection. Thus, based on multiple frames of image information of the mine shaft auger, accurate detection of auger faults is achieved, saving maintenance costs.

[0098] Figure 8 This is a schematic diagram of the structure of a mine drum ear fault detection device based on video analysis provided in an embodiment of this application.

[0099] like Figure 8 As shown, the video analysis-based mine canister ear fault detection device 800 includes: an acquisition module 801, a segmentation module 802, a comparison module 803, and a determination module 804, wherein:

[0100] The acquisition module 801 is used to acquire image information of the ear of the mine.

[0101] The segmentation module 802 is used to control the image segmentation algorithm to segment the image information to obtain the position information and outline of the ear of the jar.

[0102] The comparison module 803 is used to compare the differences between the ear contour images of two adjacent frames to obtain a differential contour image.

[0103] The determination module 804 is used to determine whether the ear corresponding to the location information is faulty based on the difference between the pixel value in the differential contour image and the preset pixel threshold.

[0104] This application proposes a mine shaft auger fault detection device based on video analysis. It acquires image information of the mine shaft auger, controls an image segmentation algorithm to segment the image information to obtain the location information and contour map of the auger. It compares the contour maps of adjacent frames to obtain a differential contour image. Based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold, it determines whether the auger corresponding to the location information is faulty. Thus, based on the image information of the mine shaft auger, the location information and contour map of the auger are determined to accurately determine whether the auger is faulty, achieving precise detection of mine shaft auger faults and ensuring the safe operation of mine shaft augers.

[0105] In one embodiment of this application, Figure 9 This is a schematic diagram of another video analysis-based mine shaft ear fault detection device provided in the embodiments of this application, as shown below. Figure 9 As shown, the mine shaft ear fault detection device 900 based on video analysis may further include: an acquisition module 901, a segmentation module 902, a comparison module 903, a determination module 904, a storage module 905, a generation module 906, and a control module 907, wherein the segmentation module 902 includes a segmentation unit 9021, a generation unit 9022, and a noise reduction unit 9023.

[0106] For detailed descriptions of the acquisition module 901, segmentation module 902, comparison module 903, and determination module 904, please refer to [link / reference]. Figure 8 The descriptions of the acquisition module 801, segmentation module 802, comparison module 803, and determination module 804 in the illustrated embodiment are not repeated here.

[0107] In one embodiment of this application, such as Figure 9 As shown, the segmentation module 902 includes:

[0108] The segmentation unit 9021 is used to perform feature segmentation on the ear image in the image information to obtain multiple feature image blocks.

[0109] The generation unit 9022 is used to cluster and locate multiple feature image blocks based on the similarity measure of multiple feature image blocks to obtain the initial outline map and location information of the ear of the jar.

[0110] The denoising unit 9023 is used to denoise the initial jar ear contour map to obtain the jar ear contour map of the jar ear contour map.

[0111] In one embodiment of this application, such as Figure 9 As shown, the noise reduction unit 9023 is specifically used for:

[0112] The initial jar ear outline is scanned to obtain each pixel of the initial jar ear outline.

[0113] The weighted average gray value of adjacent pixels is used as the denoised pixel in the initial ear contour map.

[0114] Based on the denoised pixels, generate the outline of the jar ear.

[0115] In one embodiment of this application, such as Figure 9 As shown, module 904 is specifically used for:

[0116] The difference contour image is binarized to obtain the corresponding binarized difference image.

[0117] Pixels whose pixel values ​​in the binarized difference image are greater than a preset pixel threshold are considered as foreground moving pixels.

[0118] Based on the moving pixels in the foreground, construct the pixel motion region corresponding to the moving pixels in the foreground.

[0119] Based on the pixel motion area, determine whether the can ear is faulty according to the location information.

[0120] In one embodiment of this application, such as Figure 9 As shown, the device also includes:

[0121] The storage module 905 is used to save the fault video corresponding to the faulty can ear when there is a faulty can ear under the location information, so as to generate alarm information corresponding to the fault video.

[0122] The generation module 906 is used to generate control signals for regulating alarm information based on alarm information.

[0123] The control module 907 is used to send control signals to the system control platform corresponding to the can ear, and to control the can ear through the system control platform.

[0124] This application proposes a mine shaft auger fault detection device based on video analysis. It acquires image information of the mine shaft auger, controls an image segmentation algorithm to segment the image information to obtain the location information and contour map of the auger. It compares the contour maps of adjacent frames to obtain a differential contour image. Based on the difference between the pixel values ​​in the differential contour image and a preset pixel threshold, it determines whether the auger corresponding to the location information is faulty. Thus, based on the image information of the mine shaft auger, the location information and contour map of the auger are determined to accurately determine whether the auger is faulty, achieving precise detection of mine shaft auger faults and ensuring the safe operation of mine shaft augers.

[0125] According to embodiments of this application, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the video analysis-based mine drum ear fault detection method of this application embodiments.

[0126] This application also proposes a computer program product that, when executed by an instruction processor, implements the video analysis-based mine drum ear fault detection method of the embodiments of this application.

[0127] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0128] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0129] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for detecting mine canister ear faults based on video analysis, characterized in that, The method includes: Acquire image information of the drum in the mine; The image segmentation algorithm is controlled to perform feature segmentation on the ear-shaped image in the image information to obtain multiple feature image blocks; Based on the similarity metric of the multiple feature image blocks, the multiple feature image blocks are clustered and located to obtain the initial outline map and location information of the ear of the jar. The initial jar ear contour image is denoised to obtain the jar ear contour image of the jar ear contour image; The differences between the outline images of the can ear in two adjacent frames are compared to obtain a differential outline image. Based on the difference between the pixel values ​​in the differential contour image and the preset pixel threshold, it is determined whether the can ear corresponding to the location information is faulty; The step of determining whether the can ear corresponding to the location information is faulty based on the difference between the pixel value in the differential contour image and a preset pixel threshold includes: The differential contour image is binarized to obtain the corresponding binarized difference image; Pixels whose pixel values ​​in the binarized difference image are greater than the preset pixel threshold are taken as foreground moving pixels; Based on the foreground moving pixels, construct the pixel motion region corresponding to the foreground moving pixels; Based on the pixel motion region, determine whether the can ear under the position information is faulty; The method further includes: If a faulty can ear exists under the location information, save the fault video corresponding to the faulty can ear to generate alarm information corresponding to the fault video; Based on the alarm information, a control signal is generated to regulate the alarm information; The control signal is sent to the system control platform corresponding to the can ear, and the can ear is regulated by the system control platform.

2. The method according to claim 1, characterized in that, The step of denoising the initial ear-shaped outline to obtain the ear-shaped outline of the original ear-shaped outline includes: The initial can ear outline is scanned to obtain each pixel of the initial can ear outline; The weighted average gray value of adjacent pixels is used as the denoised pixel of the initial ear contour map; Based on the denoised pixels, a contour map of the jar ear is generated.

3. A mine shaft ear fault detection device based on video analysis, characterized in that, The device includes: The acquisition module is used to acquire image information of the drum in the mine. The segmentation module is used to control the image segmentation algorithm to segment the image information to obtain the position information and outline of the ear of the can; The comparison module is used to compare the differences between the contour images of the can ear in two adjacent frames to obtain a differential contour image. The determination module is used to determine whether the can ear corresponding to the location information is faulty based on the difference between the pixel value in the differential contour image and the preset pixel threshold. The segmentation module includes: The segmentation unit is used to perform feature segmentation on the can ear image in the image information to obtain multiple feature image blocks; The generation unit is used to cluster and locate the multiple feature image blocks according to the similarity metric of the multiple feature image blocks, so as to obtain the initial outline map and location information of the ear of the can; A denoising unit is used to denoise the initial ear contour map to obtain the ear contour map of the ear contour map. The determining module is specifically used for: The differential contour image is binarized to obtain the corresponding binarized difference image; Pixels whose pixel values ​​in the binarized difference image are greater than the preset pixel threshold are taken as foreground moving pixels; Based on the foreground moving pixels, construct the pixel motion region corresponding to the foreground moving pixels; Based on the pixel motion region, determine whether the can ear under the position information is faulty; The device further includes: The storage module is used to save the fault video corresponding to the faulty can ear when there is a faulty can ear under the location information, so as to generate alarm information corresponding to the fault video; The generation module is used to generate a control signal for regulating the alarm information based on the alarm information. The control module is used to send the control signal to the system control platform corresponding to the can ear, and to control the can ear through the system control platform.

4. The apparatus according to claim 3, characterized in that, The noise reduction unit is specifically used for: The initial can ear outline is scanned to obtain each pixel of the initial can ear outline; The weighted average gray value of adjacent pixels is used as the denoised pixel of the initial ear contour map; Based on the denoised pixels, a contour map of the jar ear is generated.

Images