Operation state monitoring method of conveying belt and electronic equipment

Abstract

The invention belongs to the technical field of computers and image information processing, and particularly discloses a method for monitoring the running state of a conveying belt and electronic equipment, and the method comprises the steps: obtaining a target running state image of a target conveying belt; based on the target running state image, whether the target conveying belt is in an abnormal running state or not is determined through a pre-trained belt monitoring model, and the abnormal running state comprises one or more of a belt tearing state, a belt deviation state and a foreign matter mixing state; if the target conveying belt is in the abnormal operation state, abnormal state parameters of the target conveying belt are recognized, and the abnormal state parameters are used for representing the abnormal degree of the target conveying belt in the corresponding abnormal operation state. Through the technical scheme provided by the invention, the accuracy of monitoring the running state of the conveying belt can be improved.

Description

Technical Field

[0001] The present application belongs to the field of computer and image information processing technology, and specifically relates to a method and electronic equipment for monitoring the operating status of a conveyor belt. Background Art

[0002] At present, in traditional methods, to monitor whether a conveyor belt is torn, a pressure sensor is usually installed under the conveyor belt. When material on the conveyor belt leaks from the belt crack and falls on the pressure sensor, the pressure of the pressure sensor changes and triggers an alarm. To detect whether the conveyor belt is deviated, a belt deviation switch is often used. When the belt deviates, the edge of the belt touches the switch roller and deflects the roller. According to the deflection angle of the switch roller, different levels of belt deviation alarm signals are triggered. To determine whether there is foreign matter mixed on the conveyor belt, manual monitoring is often used. In the above methods, there are often problems such as detection hysteresis and a large number of false alarm triggering, which leads to low accuracy in monitoring the running status of the conveyor belt. Based on this, how to improve the accuracy of monitoring the running status of the conveyor belt is a technical problem that needs to be solved urgently. Summary of the Invention

[0003] The embodiments of the present application provide a method, device, computer program product or computer program, computer-readable storage medium, and electronic device for monitoring the operating status of a conveyor belt, thereby improving the accuracy of monitoring the operating status of the conveyor belt at least to a certain extent.

[0004] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by practice of the present application.

[0005] According to a first aspect of an embodiment of the present application, a method for monitoring the operating status of a conveyor belt is provided, characterized in that the method includes: obtaining a target operating status image of a target conveyor belt; based on the target operating status image, determining whether the target conveyor belt is in an abnormal operating state through a pre-trained belt monitoring model, the abnormal operating state including one or more of a belt tearing state, a belt offset state, and a foreign matter mixing state; if the target conveyor belt is in an abnormal operating state, identifying abnormal state parameters of the target conveyor belt, the abnormal state parameters being used to characterize the degree of abnormality of the target conveyor belt in the corresponding abnormal operating state.

[0006] In some embodiments of the present application, based on the aforementioned scheme, obtaining the target operating status image of the target conveyor belt includes: obtaining the operating image of the target conveyor belt acquired by an image acquisition device; and preprocessing the operating image of the target conveyor belt to obtain the target operating status image of the target conveyor belt.

[0007] In some embodiments of the present application, based on the aforementioned scheme, the belt monitoring model is trained through the following steps: obtaining multiple reference operating status images, wherein the reference operating status images are images of the conveyor belt when it is in an abnormal operating state; obtaining abnormal status labels for each reference operating status image, wherein the abnormal status labels include belt tear labels, belt offset labels, or foreign matter mixing labels; based on each reference operating status image and the abnormal status labels of each reference operating status image, a pre-constructed machine learning model framework is supervisedly trained to obtain a belt monitoring model.

[0008] In some embodiments of the present application, based on the aforementioned scheme, if the target conveyor belt is in a belt-torn state, the identifying of the abnormal state parameters of the target conveyor belt includes: performing inverse projection transformation processing on the target running state image to obtain a projection correction image of the target conveyor belt, and the projection correction image is the running state image of the target conveyor belt after eliminating the perspective error; extracting the belt-torn area in the projection correction image, and binarizing the belt-torn area to obtain a binarized image of the belt-torn area, and the binarized image includes pixel distribution characteristics of the belt crack; based on the pixel distribution characteristics of the belt crack in the binarized image, determining the crack length and crack width of the belt crack as the abnormal state parameters of the target conveyor belt.

[0009] In some embodiments of the present application, based on the aforementioned scheme, if the target conveyor belt is in a belt offset state, the identifying of the abnormal state parameters of the target conveyor belt includes: performing inverse projection transformation processing on the target operating state image to obtain a projection correction image of the target conveyor belt; based on the projection correction image, determining the exposed area of the rollers on both sides of the target conveyor belt, and based on the exposed area, determining the belt offset direction and belt offset percentage of the belt as the abnormal state parameters of the target conveyor belt.

[0010] In some embodiments of the present application, based on the aforementioned solution, determining the belt deflection direction and the belt deflection percentage of the belt based on the exposed area includes: comparing the exposed areas of the rollers on both sides of the target conveyor belt; if the exposed area of the first roller is larger than the exposed area of the second roller, determining that the target conveyor belt is deflected toward the side where the second roller is located, and the first roller and the second roller are respectively located on both sides of the target conveyor belt; if the exposed area of the first roller is smaller than the exposed area of the second roller, determining that the target conveyor belt is deflected toward the side where the first roller is located;

[0011] When the target conveyor belt deviates toward the side where the second roller is located, the belt deviation percentage is determined by the following first formula:

[0012]

[0013] When the target conveyor belt deviates toward the side where the first roller is located, the belt deviation percentage is determined by the following second formula:

[0014]

[0015] Wherein, μ represents the belt deviation percentage when the target conveyor belt deviates toward the side where the second roller is located, μ ′ It represents the belt deviation percentage when the target conveyor belt deviates toward the side where the first roller is located, D1 represents the exposed area of the first roller, and D2 represents the exposed area of the second roller.

[0016] In some embodiments of the present application, based on the aforementioned scheme, if the target conveyor belt is in a foreign matter mixed state, the identifying of the abnormal state parameters of the target conveyor belt further includes: performing inverse projection transformation processing on the target running state image to obtain a projection correction image of the target conveyor belt; determining the shape data and size data of the target foreign matter mixed on the target conveyor belt based on the projection correction image; retrieving a foreign matter type library, and matching a foreign matter type similar to the shape data and size data of the target foreign matter from the foreign matter type library as the foreign matter type of the target foreign matter; and using the foreign matter type, shape data and size data of the target foreign matter as the abnormal state parameters of the target conveyor belt.

[0017] In some embodiments of the present application, based on the aforementioned scheme, the method further includes: if there is no foreign body type with the same shape data and size data as the target foreign body in the foreign body type library, the shape data and size data of the target foreign body are written into the foreign body type library as the shape data and size data of the new foreign body type.

[0018] In some embodiments of the present application, based on the aforementioned scheme, the method further includes: if the target foreign object is a level one foreign object, controlling the target conveyor belt to stop running and issuing a level one alarm to the operation and maintenance platform, and the level one foreign object is a foreign object that can cause the conveyor belt to tear; if the target foreign object is a level two foreign object, issuing a level two alarm to the operation and maintenance platform, and the level two foreign object is a foreign object that will not cause the conveyor belt to tear, and the severity of the level one alarm is greater than the severity of the level two alarm.

[0019] According to a second aspect of an embodiment of the present application, a device for monitoring the operating status of a conveyor belt is provided, characterized in that the device includes: an acquisition unit for acquiring a target operating status image of a target conveyor belt; a determination unit for determining whether the target conveyor belt is in an abnormal operating status based on the target operating status image through a pre-trained belt monitoring model, wherein the abnormal operating status includes one or more of a belt tearing status, a belt offset status, and a foreign matter mixing status; and an identification unit for identifying abnormal status parameters of the target conveyor belt if the target conveyor belt is in an abnormal operating status, wherein the abnormal status parameters are used to characterize the degree of abnormality of the target conveyor belt in the corresponding abnormal operating status.

[0020] According to a third aspect of the embodiments of the present application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method described in any one of the embodiments of the first aspect above.

[0021] According to a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method described in any one of the embodiments of the first aspect above is implemented.

[0022] According to the fifth aspect of the embodiments of the present application, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, which, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the method described in any one of the embodiments of the first aspect above.

[0023] Based on the technical solution proposed in this application, firstly, by obtaining the target operating state image of the target conveyor belt, the physical characteristics, positional relationship and surrounding environment of the target belt surface can be intuitively presented. Compared with relying solely on sensor data, the information contained in the image is more comprehensive and rich, which can provide the belt monitoring model with richer original feature data, and help improve the accuracy of detecting abnormal states of the target conveyor belt. Secondly, by utilizing the feature extraction and classification capabilities of the belt monitoring model, it is possible to identify whether the target conveyor belt is in an abnormal operating state, which can improve the accuracy of monitoring the operating state of the conveyor belt. Finally, when the belt monitoring model determines that the target conveyor belt is in an abnormal operating state, it further identifies the abnormal state parameters, which can convert the originally fuzzy abnormal operating state into a specific numerical indicator, and can more accurately judge the severity of the abnormal operating state, providing a strong guarantee for the reliable operation of the conveyor belt and production safety.

[0024] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are incorporated into and constitute a part of the specification, illustrating embodiments consistent with the present application and, together with the specification, explaining the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can derive other drawings based on these drawings without inventive effort. In the drawings:

[0026] Figure 1 A flow chart showing a method for monitoring the running status of a conveyor belt in one embodiment of the present application is shown;

[0027] Figure 2 A flow chart of a belt monitoring model in one embodiment of the present application is shown;

[0028] Figure 3 A schematic diagram showing a conveyor belt in a torn state according to an embodiment of the present application is shown;

[0029] Figure 4 A schematic diagram of determining the width of a belt crack in one embodiment of the present application is shown;

[0030] Figure 5 A schematic diagram showing the distribution positions of image acquisition devices in one embodiment of the present application is shown;

[0031] Figure 6 A schematic diagram showing a conveyor belt in a belt deviation state according to an embodiment of the present application is shown;

[0032] Figure 7 A schematic diagram showing a conveyor belt in a state where foreign matter is mixed in according to one embodiment of the present application is shown;

[0033] Figure 8 A block diagram of a device for monitoring the running status of a conveyor belt in one embodiment of the present application is shown;

[0034] Figure 9 A schematic structural diagram of an electronic device in one embodiment of the present application is shown. DETAILED DESCRIPTION

[0035] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0036] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner.In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps etc. can be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0037] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0038] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0039] It should be noted that the term "plurality" used in this document refers to two or more. "And / or" describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. The character " / " generally indicates an "or" relationship between the associated objects.

[0040] It should be noted that the terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described.

[0041] At present, in traditional methods, to monitor whether a conveyor belt is torn, a pressure sensor is usually installed under the conveyor belt. When material on the conveyor belt leaks from the belt crack and falls on the pressure sensor, the pressure of the pressure sensor changes and triggers an alarm. To detect whether the conveyor belt is deviated, a belt deviation switch is often used. When the belt is deviated, the edge of the belt will touch the switch roller and deflect the roller. According to the deflection angle of the switch roller, different levels of belt deviation alarm signals are triggered. To determine whether there is foreign matter mixed on the conveyor belt, manual monitoring is often used. In the above methods, there are often problems such as detection hysteresis and a large number of false alarm triggering, which leads to low accuracy in monitoring the operating status of the conveyor belt. Based on this, the present application proposes a method for monitoring the operating status of a conveyor belt to improve the accuracy of monitoring the operating status of the conveyor belt.

[0042] Next, we will combine Figure 1 The method for monitoring the operating status of a conveyor belt proposed in this application is described in detail.

[0043] See also Figure 1 , shows a flow chart of a method for monitoring the running status of a conveyor belt in one embodiment of the present application. Specifically, the method can be executed by a device with a computing and processing function. Figure 1 As shown, the method for monitoring the running status of the conveyor belt may include at least steps 110 to 130, which are described in detail as follows:

[0044] In step 110 , a target operating state image of a target conveyor belt is acquired.

[0045] In the present application, the target conveyor belt may refer to a conveyor belt that needs to be monitored, such as a conveyor belt in a steel plant, a port, or the like.

[0046] In the present application, the target operating status image can be obtained through an image acquisition device, which can reflect the current operating status of the target conveyor belt. The image can include the surface condition, position, and whether there are foreign objects of the target conveyor belt.

[0047] Furthermore, the acquisition of the target operating state image of the target conveyor belt may be performed according to the following steps 111 to 112:

[0048] Step 110: Acquire a running image of the target conveyor belt captured by an image capture device.

[0049] Step 112 : pre-processing the target conveyor belt operation image to obtain a target operation state image of the target conveyor belt.

[0050] In the present application, the image acquisition device can be a camera with high-definition and high-frame rate camera functions, for example, it can be an industrial camera or a high-speed camera. The specific one can be determined according to actual needs, and this application does not make any specific restrictions on this.

[0051] In the present application, the image acquisition device may acquire the running image of the target conveyor belt in real time, or may acquire the running image of the target conveyor belt once every certain period of time. This application does not make any specific limitation on this.

[0052] In this application, the shooting area of the image acquisition device needs to ensure sufficient lighting and small light changes to avoid false alarms. The lens of the image acquisition device needs to be maintained regularly to avoid dust accumulation and affect the imaging quality of the running image.

[0053] In the present application, the running image of the target conveyor belt is preprocessed, specifically, the running image of the target conveyor belt is filtered, denoised and contrast enhanced to improve the image quality and ensure the accuracy of subsequent analysis.

[0054] In the present application, on the one hand, the image acquisition device can select different models according to different working conditions to ensure the quality of the acquired image and enhance the robustness of the device. On the other hand, by preprocessing the running image of the target conveyor belt, noise can be effectively eliminated, the clarity of the image can be improved, and the key features of the abnormal running state (such as cracks caused by belt tearing, mixed foreign matter, etc.) can be enhanced, effectively improving the accuracy of subsequent analysis results.

[0055] Continue to refer to Figure 1 In step 120, based on the target operating status image, a pre-trained belt monitoring model is used to determine whether the target conveyor belt is in an abnormal operating state, wherein the abnormal operating state includes one or more of a belt tearing state, a belt offset state, and a foreign matter mixing state.

[0056] In the present application, the belt monitoring model may be a model constructed based on a machine learning algorithm, which has previously learned a large number of images of conveyor belts in abnormal operating states and has the ability to identify whether the conveyor belt is operating abnormally.

[0057] In this application, the abnormal operating state refers to a situation in which the conveyor belt appears during operation that is inconsistent with the normal state, which may affect the normal operation of the conveyor belt and even cause equipment damage or production accidents. Specifically, it may include at least one of a belt tearing state (referring to a situation in which cracks or damage appear on the surface of the conveyor belt), a belt offset state (referring to a situation in which the conveyor belt deviates from the normal position during operation, which may cause material spillage or equipment wear), and foreign matter mixing (referring to objects that should not be present mixed into the conveyor belt, and these foreign objects may cause damage to the conveyor belt or subsequent equipment).

[0058] In this application, the target operating status image of the target conveyor belt is analyzed through a pre-trained belt monitoring model, which can directly capture visual information such as the surface texture of the conveyor belt, the belt edge position, and the characteristics of foreign objects. It can break through the limitations of physical contact detection in traditional methods, and thus determine the abnormal state of the conveyor belt in the early stage of the abnormality (such as when the crack has not been penetrated by material, and before the belt offset touches the switch), thereby solving the problems of strong hysteresis and high misjudgment rate in traditional technologies, and effectively improving the accuracy of conveyor belt operating status monitoring.

[0059] Next, in order to enable those skilled in the art to better understand, this application will provide a detailed description of the belt monitoring model.

[0060] Reference Figure 2 , shows a flow chart of a belt monitoring model in one embodiment of the present application. Specifically, training the belt monitoring model may include steps 210 to 230:

[0061] In step 210 , a plurality of reference operating state images are acquired, wherein the reference operating state images are images when the conveyor belt is in an abnormal operating state.

[0062] In the present application, the acquisition of multiple reference operating status images can be collecting images of the conveyor belt in different abnormal operating states (such as belt tearing state, belt offset state and foreign matter mixing state), ensuring that there are sufficient samples for each type of abnormal operating state.

[0063] Continue to refer to Figure 2 In step 220, the abnormal state labels of each reference running state image are obtained, and the abnormal state labels include belt tearing labels, belt deviation labels, or foreign matter mixed labels.

[0064] In the present application, the abnormal state label can mark the abnormal state type of the conveyor belt in each reference operating state image, so that the model can accurately learn the correspondence between the abnormal operating state and the abnormal state label, which helps the trained belt monitoring model to more accurately judge the abnormal operating state of the conveyor belt in subsequent monitoring tasks.

[0065] In this application, by acquiring image data with multiple abnormal state labels, the model can be exposed to more diverse abnormal operation scenarios, thereby enhancing the generalization ability of the belt monitoring model, improving the belt monitoring model's ability to recognize unknown abnormal situations, and thereby improving the accuracy of conveyor belt operation status monitoring.

[0066] Continue to refer to Figure 2 In step 230, based on each reference running state image and the abnormal state label of each reference running state image, the pre-built machine learning model framework is supervised trained to obtain a belt monitoring model.

[0067] In this application, by training the machine learning model framework through supervised training, the model can learn the mapping relationship between the reference operating status image and the abnormal status label, so that the model can analyze the new conveyor belt operating status image, determine whether the conveyor belt is in an abnormal operating state, and identify the specific abnormality type.

[0068] In a specific embodiment of the present application, the belt monitoring model can specifically include a CBS module (convolution + normalization + activation function), an ELAN module (extended potential attention network) and an MP module (maximum pooling). The CBS module is designed with three different convolution kernel sizes and step sizes, so that it can refer to abnormal state features in the running state image at different scales (such as belt tearing features, belt offset features, and foreign matter mixing features). The ELAN module can control the shortest gradient path and the longest gradient path in the network to determine the most suitable parameters that can be fitted while avoiding overfitting. The MP module consists of two branches and performs a maximum pooling operation to reduce the image resolution of the reference running state image, thereby saving computing resources. The belt monitoring model can also include four modules, spatial pyramid pooling (SPPCSPC), feature pyramid network () (FPN), re-parameterized structure (RepConv) and detection head. The SPC module obtains different receptive fields through maximum pooling to adapt to reference operating status images of different resolutions. FPN can enhance the belt monitoring model's ability to integrate different features. RepConv reparameterizes the branch parameters to the main branch during inference, thereby reducing computational complexity and memory consumption.

[0069] In a specific embodiment of the present application, the belt monitoring model uses a positioning loss function (Wise-IoUv3) to enable the model to more accurately determine the position and bounding box of foreign body features. For example, if the width of a belt crack is only 3-5 mm, the position and boundary of the crack can be located more accurately; the belt monitoring model automatically improves the weight of the sample through cross-entropy loss calculation, for example, automatically reducing the weight of the reference operating state image corresponding to abnormal state features that are larger in number and better learned, and increasing the weight of the reference operating state image corresponding to abnormal state features that are smaller in number and poorer learned.

[0070] In this application, by using reference operating status images with abnormal status labels to conduct supervised training on the machine learning model framework, the accuracy, adaptability and robustness of the belt monitoring model can be effectively improved, so that it can accurately identify whether the conveyor belt is in an abnormal operating state, thereby realizing real-time monitoring and early warning, thereby reducing manual intervention, reducing operation and maintenance costs, and ensuring production safety and efficiency.

[0071] Continue to refer to Figure 1 In step 130, if the target conveyor belt is in an abnormal operating state, an abnormal state parameter of the target conveyor belt is identified, where the abnormal state parameter is used to characterize the degree of abnormality of the target conveyor belt in the corresponding abnormal operating state.

[0072] In this application, by identifying the abnormal state parameters of the target conveyor belt under abnormal operating conditions, the severity of the abnormal state can be quantified more accurately, and more specific fault information can be provided to operation and maintenance personnel. This not only helps to quickly locate the cause of the abnormal operating state, but also can formulate corresponding maintenance strategies based on the degree of abnormality. In addition, the abnormal state parameters can also be used to evaluate the remaining service life of the target conveyor belt, providing data support for preventive maintenance of the equipment, thereby effectively extending the service life of the equipment and reducing maintenance costs.

[0073] In the present application, firstly, by acquiring the target operating state image of the target conveyor belt, the physical characteristics, positional relationship and surrounding environment of the target belt surface can be intuitively presented. Compared with relying solely on sensor data, the information contained in the image is more comprehensive and rich, which can provide the belt monitoring model with richer original feature data, and help improve the accuracy of abnormal state detection of the target conveyor belt. Secondly, by utilizing the feature extraction and classification capabilities of the belt monitoring model, it is possible to identify whether the target conveyor belt is in an abnormal operating state, which can improve the accuracy of monitoring the operating state of the conveyor belt. Finally, when the belt monitoring model determines that the target conveyor belt is in an abnormal operating state, it further identifies the abnormal state parameters, which can convert the originally fuzzy abnormal operating state into a specific numerical indicator, and can more accurately judge the severity of the abnormal operating state, providing a strong guarantee for the reliable operation and production safety of the conveyor belt.

[0074] Furthermore, in the above step 130, if the target conveyor belt is in a belt-torn state, the identifying of abnormal state parameters of the target conveyor belt may be performed in accordance with the following steps 131 to 133:

[0075] Step 131 : performing inverse projection transformation processing on the target running state image to obtain a projection correction image of the target conveyor belt. The projection correction image is the running state image of the target conveyor belt after eliminating perspective error.

[0076] Step 132 : extracting the belt tear region in the projected correction image, and performing binarization processing on the belt tear region to obtain a binarized image of the belt tear region, wherein the binarized image includes pixel distribution characteristics of the belt crack.

[0077] Step 133 : determining the crack length and crack width of the belt crack based on the pixel distribution characteristics of the belt crack in the binary image, as abnormal state parameters of the target conveyor belt.

[0078] In this application, please refer to Figure 3 , shows a schematic diagram of a conveyor belt in a torn state in one embodiment of the present application. As shown in the figure, a longitudinal crack appears in the conveyor belt during operation. The crack length and crack width of the crack can be determined by the method proposed in the present application.

[0079] In this application, during the training process, the belt monitoring model can also divide the transverse tears and longitudinal tears in the belt into different types of labels for annotation, so that the belt monitoring model can distinguish the two different tear categories of transverse tears and longitudinal tears, providing important decision-making basis for subsequent processing, repair and other work.

[0080] In the present application, it should be noted that under perspective transformation, the shape of the photographed object will be distorted. For example, although in reality, the two edges of the column are parallel, in the image obtained by oblique photography, the two edges of the column may not be parallel and will intersect at a certain position; therefore, by solving the inverse change of the perspective projection transformation and converting the obliquely photographed image into a front view of the belt plane through the inverse projection transformation, the crack length and width measurement errors caused by perspective deformation can be eliminated, making subsequent image analysis more accurate and able to accurately measure the crack length and width of the belt crack.

[0081] In this application, in order to establish a correspondence between the pixel coordinates in the image and the real coordinates, a calibration plate (such as a chessboard) can be used to take multiple images, and the correspondence between the corner points on the calibration plate in the image coordinate system and the world coordinate system can be calculated. The internal and external parameters of the image acquisition device are solved using a specific algorithm (such as the Zhang Zhengyou calibration method), and then the internal parameters of the image acquisition device are used to convert the coordinates in the image coordinate system into coordinates in the camera coordinate system. The coordinates in the camera coordinate system are then converted into coordinates in the world coordinate system in combination with the external parameters. In this way, the pixel coordinates in the image can be compared with the world coordinates in the real world, and accordingly, the crack length and crack width can be determined according to the pixel length and pixel width corresponding to the belt crack.

[0082] The specific formulas are shown in the following formulas (1) to (8):

[0083]

[0084] Where J represents pixel resolution; O and O' represent the actual physical size and its corresponding pixel size respectively; L represents the distance between the object and the camera; f represents the focal length of the camera; α represents the number of pixels contained in the physical size of the camera's photosensitive chip; f and α are both fixed parameters of the camera, T x and T y represents the component without perspective error, T represents the true width of the crack, and H is the reversible perspective transformation matrix; X and X' are the homogeneous coordinates of any point in the plane and the homogeneous coordinates of its corresponding point after perspective transformation, respectively.

[0085] In this application, the binarization process of the projected correction image is as follows: Figure 4As shown, a schematic diagram of crack width judgment of a belt crack in one embodiment of the present application is shown. As shown in the figure, the belt crack is the white area in the binary image. The binary image of the belt crack may include the pixel distribution characteristics of the belt crack. The crack length and crack width of the belt crack are determined based on the pixel distribution characteristics of the belt crack in the binary image. Specifically, a strip template can be set. The strip template can be composed of a series of fixed-length straight lines with a length of in the horizontal or vertical direction. The angle between the strip and the horizontal direction is, which is evenly distributed in the range of 0° to 180°, as shown in FIG. Figure 4 The angle increase δ used in this paper is θ When the tearing angle is in the range of 0° to 45° or 135° to 175°, it indicates the vertical width of the strip. When the strip angle is in the range of 50° to 130°, w s Represents the horizontal width of the strip. To obtain the width of any crack location, a pixel on the crack skeleton line is used as the center point. A local area with a length and width of W, including the belt crack, can be extracted from the binary image. This area is then template-convolved with the strip template at each angle. Template convolution actually obtains the number of pixels in the overlapping area between the strip template and the crack, that is, the area of the overlapping area. When the area of the overlapping area is minimized, the strip template is considered perpendicular to the crack at this location. The convolution result is the product of the strip width and the crack width. The crack width at this location can be obtained by the strip width. Then, starting from either end of the crack skeleton line, the above operation can be repeated along the crack skeleton line to obtain the crack width at any location.

[0086] The specific calculation formulas are as follows: Formula (9) to Formula (11)

[0087]

[0088] Where n is the total number of pixels on the crack skeleton line; p is the maximum crack width at the pth pixel along the crack skeleton line; w i is the crack width at the i-th pixel along the crack skeleton line; m is the range of the average value required to prevent the influence of abnormal values of the crack width

[0089] In the present application, by performing inverse projection transformation on the target operating status image to eliminate perspective errors, and extracting the belt tear area for binarization processing, the pixel distribution characteristics of the belt crack can be effectively improved, so that the crack length and crack width of the belt crack can be accurately determined based on the pixel distribution characteristics of the belt crack in the binary image, thereby improving the accuracy and reliability of the detection of the conveyor belt operating status.

[0090] In the above step 130, if the target conveyor belt is in the belt deviation state, the identification of the abnormal state parameters of the target conveyor belt may be performed in accordance with the following steps 134 to 135:

[0091] Step 134 , performing inverse projection transformation processing on the target operating state image to obtain a projection correction image of the target conveyor belt.

[0092] Step 135 , based on the projected correction image, determining the exposed areas of the rollers on both sides of the target conveyor belt, and based on the exposed areas, determining the belt offset direction and belt offset percentage of the belt as abnormal state parameters of the target conveyor belt.

[0093] In the present application, the image acquisition device for acquiring the image corresponding to the belt deviation state can be set at the tail or head of the belt conveyor, or can be set near the counterweight wheel of the belt conveyor, specifically Figure 5 As shown, a schematic diagram of the distribution positions of the image acquisition devices in one embodiment of the present application is shown. As shown in the figure, the first image acquisition device 501 is arranged at the tail of the belt conveyor and can be used to acquire images of belt deviation. The second image acquisition device 502 is arranged in the middle part of the belt conveyor. The third image acquisition device 503 is arranged in the middle position of the upper conveyor belt and the lower conveyor belt. The second image acquisition device 502 and the third image acquisition device 503 can both be used to acquire images of belt tearing. The fourth image acquisition device 504 is arranged at the head of the belt conveyor and can be used to acquire images of foreign matter mixing.

[0094] In the present application, by performing inverse projection transformation processing on the target operating status image, the perspective error of the exposed area of the rollers on both sides of the conveyor belt can be eliminated, the accuracy of the exposed area detection of the rollers on both sides of the conveyor belt can be improved, and the accuracy of the subsequent analysis process can be improved.

[0095] In this application, please refer to Figure 6 , shows a schematic diagram of a conveyor belt in an embodiment of the present application in a belt deviation state. As shown in the figure, the exposed areas of the rollers on both sides of the conveyor belt are different, indicating that the conveyor belt has deviated during operation. The deviation percentage of the conveyor belt is calculated based on the difference in the exposed areas of the rollers on both sides, which can intuitively reflect the severity of the conveyor belt deviation.

[0096] Furthermore, in the above step 135, the belt offset direction and the belt offset percentage of the belt are determined based on the exposed area, which can be specifically performed according to the following steps 1351 to 1354:

[0097] Step 1351, compare the exposed areas of the rollers on both sides of the target conveyor belt. If the exposed area of the first roller is larger than the exposed area of the second roller, it is determined that the target conveyor belt is offset toward the side where the second roller is located, and the first roller and the second roller are respectively located on both sides of the target conveyor belt.

[0098] Step 1352: If the exposed area of the first roller is smaller than the exposed area of the second roller, it is determined that the target conveyor belt is offset to the side where the first roller is located.

[0099] Step 1353: When the target conveyor belt deviates toward the side where the second roller is located, the belt deviation percentage is determined by the following formula (12):

[0100]

[0101] Step 1354: When the target conveyor belt deviates toward the side where the first roller is located, the belt deviation percentage is determined by the following formula (13):

[0102]

[0103] Wherein, μ represents the belt deviation percentage when the target conveyor belt deviates toward the side where the second roller is located, μ ′ It represents the belt deviation percentage when the target conveyor belt deviates toward the side where the first roller is located, D1 represents the exposed area of the first roller, and D2 represents the exposed area of the second roller.

[0104] Throughout this application, please continue to refer to Figure 6 Specifically, the first roller and the second roller can be located on both sides of the conveyor belt respectively.

[0105] In the present application, the belt deviation percentage can be used to determine the degree of belt deviation. Specifically, when the belt deviation percentage is less than 30%, the conveyor belt can be considered to be in a normal operating state. When the belt deviation percentage is greater than or equal to 30% and less than 60%, the conveyor belt can be considered to be in a slightly deviated state. When the belt deviation percentage is greater than or equal to 60%, the conveyor belt can be considered to be in a severely deviated state.

[0106] In this application, by comparing the exposed areas of the rollers on both sides of the conveyor belt to determine the offset direction of the conveyor belt, and using a formula to quantify the belt offset percentage, the accuracy and reliability of offset status detection can be significantly improved, thereby providing intuitive and specific offset information to operation and maintenance personnel, helping to adjust the position of the conveyor belt in a timely manner, reducing the risk of production interruption and equipment damage caused by belt offset, and improving the safety of the transportation process.

[0107] In the above step 130, if the target conveyor belt is in a foreign matter mixed state, the identification of the abnormal state parameters of the target conveyor belt may be performed in accordance with the following steps 136 to 139:

[0108] Step 136 : Perform inverse projection transformation processing on the target operating state image to obtain a projection correction image of the target conveyor belt.

[0109] Step 137 : determining shape data and size data of the target foreign matter mixed on the target conveyor belt based on the projected correction image.

[0110] Step 138 : searching a foreign object type library, and matching a foreign object type from the foreign object type library with a shape data and a size data similar to that of the target foreign object as the foreign object type of the target foreign object.

[0111] Step 139: Using the foreign object type, shape data, and size data of the target foreign object as abnormal state parameters of the target conveyor belt.

[0112] In this application, please refer to Figure 7 , shows a schematic diagram of a conveyor belt in a state of foreign matter mixed in one embodiment of the present application. As shown in the figure, there are relatively obvious foreign matter on the conveyor belt, and the foreign matter type of the foreign matter can be determined by searching the foreign matter type library.

[0113] In the present application, by performing inverse projection transformation processing on the target operating status image, the perspective error of the shape data and size data of the target foreign object can be eliminated, the accuracy of the detection of the shape data and size data of the target foreign object can be improved, and the accuracy of the subsequent analysis process can be improved.

[0114] In the present application, the foreign object type library can be constructed based on common items in the belt transportation process. For example, when the conveyor belt is transporting materials, foreign objects such as steel bars, steel pipes, wood, and lining boards may be mixed in. Based on this, the shape data and size data of the above foreign objects can be written into the foreign object type library.

[0115] In the present application, firstly, the perspective error in the target running state image is eliminated through inverse projection transformation processing, which can provide a more accurate image basis for subsequent foreign object identification, thereby improving the accuracy of foreign object identification and avoiding misjudgment or missed judgment due to image perspective distortion. Secondly, the shape data and size data of the target foreign object are matched using the foreign object type library, which can quickly and accurately determine the foreign object type of the target foreign object, help to evaluate the potential hazards of the target foreign object, and improve the safety of the transportation process.

[0116] Furthermore, based on the above solution, the method for monitoring the running status of the conveyor belt may further perform the following step 301:

[0117] Step 301: If there is no foreign body type with the same shape data and size data as the target foreign body in the foreign body type library, the shape data and size data of the target foreign body are written into the foreign body type library as the shape data and size data of a new foreign body type.

[0118] In the present application, by writing the shape data and size data of the target foreign object that has not been recorded in the foreign object type library into the foreign object type library, the foreign object type data resources in the foreign object type library can be enriched, thereby improving the accuracy and comprehensiveness of subsequent foreign object identification. In addition, the belt monitoring model can also have the ability of self-learning and continuous optimization, so that it can better adapt to more complex foreign object mixing situations, thereby effectively improving the accuracy of transmission belt operation status monitoring.

[0119] Furthermore, based on the above solution, the method for monitoring the running status of the conveyor belt may further perform the following steps 302 to 303:

[0120] Step 302: If the target foreign object is a level one foreign object, the target conveyor belt is controlled to stop running and a level one alarm is issued to the operation and maintenance platform. The level one foreign object is a foreign object that can cause the conveyor belt to tear.

[0121] Step 303: If the target foreign object is a level 2 foreign object, a level 2 alarm is issued to the operation and maintenance platform. The level 2 foreign object is a foreign object that will not cause the conveyor belt to tear. The severity of the level 1 alarm is greater than the severity of the level 2 alarm.

[0122] In the present application, the first-level foreign objects are foreign objects that can cause the conveyor belt to tear, specifically foreign objects with high hardness, sharp shape or heavy weight such as long iron rods, scrapers, lining plates, etc. The second-level foreign objects are foreign objects that will not cause the conveyor belt to tear, such as foam blocks, rubber blocks, cardboard and other soft foreign objects with no sharp edges or light weight.

[0123] In this application, the shutdown operation is automatically triggered according to the type of the target foreign object, which can quickly block the physical damage of the first-level foreign object to the conveyor belt and subsequent equipment, reducing maintenance costs. At the same time, the alarm function transmits the information of the target foreign object to the operation and maintenance platform, which can ensure that the problem of foreign object mixing can be solved in time and improve the efficiency of material transportation.

[0124] Based on the technical solution proposed in this application, firstly, by obtaining the target operating state image of the target conveyor belt, the physical characteristics, positional relationship and surrounding environment of the target belt surface can be intuitively presented. Compared with relying solely on sensor data, the information contained in the image is more comprehensive and rich, which can provide the belt monitoring model with richer original feature data, and help improve the accuracy of detecting abnormal states of the target conveyor belt. Secondly, by utilizing the feature extraction and classification capabilities of the belt monitoring model, it is possible to identify whether the target conveyor belt is in an abnormal operating state, which can improve the accuracy of monitoring the operating state of the conveyor belt. Finally, when the belt monitoring model determines that the target conveyor belt is in an abnormal operating state, it further identifies the abnormal state parameters, which can convert the originally fuzzy abnormal operating state into a specific numerical indicator, and can more accurately judge the severity of the abnormal operating state, providing a strong guarantee for the reliable operation of the conveyor belt and production safety.

[0125] The following describes an embodiment of the device of the present application, which can be used to implement the method for monitoring the running status of the conveyor belt in the above-mentioned embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method for monitoring the running status of the conveyor belt in the above-mentioned embodiment of the present application.

[0126] Figure 8 A block diagram of a device for monitoring the operating status of a conveyor belt in one embodiment of the present application is shown.

[0127] Reference Figure 4 As shown, a device 800 for monitoring the running status of a conveyor belt according to an embodiment of the present application includes: an acquisition unit 801 , a determination unit 802 and an identification unit 803 .

[0128] Among them, the acquisition unit 801 acquires the target operating state image of the target conveyor belt; the determination unit 802 is used to determine whether the target conveyor belt is in an abnormal operating state based on the target operating state image through a pre-trained belt monitoring model, and the abnormal operating state includes one or more of a belt tearing state, a belt offset state, and a foreign matter mixing state; the identification unit 803 identifies the abnormal state parameters of the target conveyor belt if the target conveyor belt is in an abnormal operating state, and the abnormal state parameters are used to characterize the degree of abnormality of the target conveyor belt in the corresponding abnormal operating state.

[0129] In some embodiments of the present application, based on the aforementioned scheme, the acquisition unit 801 is configured to: acquire the running image of the target conveyor belt acquired by the image acquisition device; pre-process the running image of the target conveyor belt to obtain the target running state image of the target conveyor belt.

[0130] In some embodiments of the present application, based on the aforementioned scheme, the device also includes a training unit, which is configured to: obtain multiple reference operating status images, where the reference operating status images are images of the conveyor belt when it is in an abnormal operating state; obtain abnormal status labels for each reference operating status image, where the abnormal status labels include belt tear labels, belt offset labels, or foreign matter mixed labels; based on each reference operating status image and the abnormal status labels of each reference operating status image, perform supervised training on a pre-constructed machine learning model framework to obtain a belt monitoring model.

[0131] In some embodiments of the present application, based on the aforementioned scheme, the identification unit 803 is configured to: perform inverse projection transformation processing on the target running status image to obtain a projection correction image of the target conveyor belt, and the projection correction image is the running status image of the target conveyor belt after eliminating the perspective error; extract the belt tearing area in the projection correction image, and perform binarization processing on the belt tearing area to obtain a binarized image of the belt tearing area, and the binarized image includes pixel distribution characteristics of the belt crack; based on the pixel distribution characteristics of the belt crack in the binarized image, determine the crack length and crack width of the belt crack as the abnormal state parameters of the target conveyor belt.

[0132] In some embodiments of the present application, based on the aforementioned scheme, the identification unit 803 is configured to: perform inverse projection transformation processing on the target operating status image to obtain a projection correction image of the target conveyor belt; based on the projection correction image, determine the exposed area of the rollers on both sides of the target conveyor belt, and based on the exposed area, determine the belt offset direction and belt offset percentage of the belt as abnormal state parameters of the target conveyor belt.

[0133] In some embodiments of the present application, based on the aforementioned solution, the identification unit 803 is configured to: compare the exposed areas of the rollers on both sides of the target conveyor belt; if the exposed area of the first roller is larger than the exposed area of the second roller, then determine that the target conveyor belt is offset toward the side where the second roller is located, and the first roller and the second roller are located on both sides of the target conveyor belt respectively; if the exposed area of the first roller is smaller than the exposed area of the second roller, then determine that the target conveyor belt is offset toward the side where the first roller is located;

[0134] When the target conveyor belt deviates toward the side where the second roller is located, the belt deviation percentage is determined by the following first formula:

[0135]

[0136] When the target conveyor belt deviates toward the side where the first roller is located, the belt deviation percentage is determined by the following second formula:

[0137]

[0138] Wherein, μ represents the belt deviation percentage when the target conveyor belt deviates toward the side where the second roller is located, μ ′ It represents the belt deviation percentage when the target conveyor belt deviates toward the side where the first roller is located, D1 represents the exposed area of the first roller, and D2 represents the exposed area of the second roller.

[0139] In some embodiments of the present application, based on the aforementioned scheme, the identification unit 803 is configured to: perform inverse projection transformation processing on the target operating state image to obtain a projection correction image of the target conveyor belt; determine the shape data and size data of the target foreign object mixed on the target conveyor belt based on the projection correction image; retrieve a foreign object type library, and match a foreign object type similar to the shape data and size data of the target foreign object from the foreign object type library as the foreign object type of the target foreign object; and use the foreign object type, shape data and size data of the target foreign object as the abnormal state parameters of the target conveyor belt.

[0140] In some embodiments of the present application, based on the aforementioned scheme, the identification unit 803 is configured as follows: if there is no foreign body type with the same shape data and size data as the target foreign body in the foreign body type library, the shape data and size data of the target foreign body are written into the foreign body type library as the shape data and size data of the new foreign body type.

[0141] In some embodiments of the present application, based on the aforementioned scheme, the identification unit 803 is configured as follows: if the target foreign object is a level one foreign object, the target conveyor belt is controlled to stop running, and a level one alarm is issued to the operation and maintenance platform, and the level one foreign object is a foreign object that can cause the conveyor belt to tear; if the target foreign object is a level two foreign object, a level two alarm is issued to the operation and maintenance platform, and the level two foreign object is a foreign object that will not cause the conveyor belt to tear, and the severity of the level one alarm is greater than the severity of the level two alarm.

[0142] As another embodiment of the present application, a computer program product or computer program is further provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method described in the above embodiment.

[0143] As another embodiment of the present application, a computer-readable storage medium is provided. The computer-readable storage medium may be included in the electronic device described in the above embodiments, or may exist independently without being incorporated into the electronic device. The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device implements the method described in the above embodiments.

[0144] Based on the same inventive concept, the embodiment of the present application also provides an electronic device. Figure 9 , which shows a schematic diagram of the structure of an electronic device in one embodiment of the present application. The electronic device includes one or more memories 904, one or more processors 902, and at least one computer program (program code) stored in the memories 904 and executable on the processors 902. When the processors 902 execute the computer program, the aforementioned method is implemented.

[0145] Among them, Figure 9 In the embodiment of the present invention, a bus architecture (represented by bus 900) is shown. Bus 900 may include any number of interconnected buses and bridges, and bus 900 links together various circuits including one or more processors represented by processor 902 and memory represented by memory 904. Bus 900 may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 905 provides an interface between bus 900 and receiver 901 and transmitter 903. Receiver 901 and transmitter 903 may be the same component, namely a transceiver, which provides a unit for communicating with various other devices over a transmission medium. Processor 902 is responsible for managing bus 900 and general processing, while memory 904 may be used to store data used by processor 902 when performing operations.

[0146] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of this application and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hardwiring, or a combination of any of these. Furthermore, the functional units may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit.

[0147] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0148] The units described as separate components may or may not be physically separate, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0149] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0150] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for monitoring the running status of a conveyor belt, characterized in that: The method comprises: Acquire a target operating state image of a target conveyor belt; Based on the target operating state image, determining whether the target conveyor belt is in an abnormal operating state using a pre-trained belt monitoring model, wherein the abnormal operating state includes one or more of a belt tearing state, a belt deviation state, and a foreign matter mixing state; If the target conveyor belt is in an abnormal operating state, an abnormal state parameter of the target conveyor belt is identified, where the abnormal state parameter is used to characterize the degree of abnormality of the target conveyor belt in the corresponding abnormal operating state.

2. The method according to claim 1, characterized in that The step of obtaining a target operating state image of a target conveyor belt comprises: Acquiring a running image of the target conveyor belt captured by an image acquisition device; The target conveyor belt operation image is preprocessed to obtain a target operation state image of the target conveyor belt.

3. The method according to claim 1, characterized in that The belt monitoring model is trained through the following steps: Acquire a plurality of reference operating state images, wherein the reference operating state images are images when the conveyor belt is in an abnormal operating state; Acquire an abnormal state label of each reference running state image, wherein the abnormal state label includes a belt tear label, a belt deviation label, or a foreign matter mixed label; Based on each reference operating state image and the abnormal state label of each reference operating state image, a pre-built machine learning model framework is supervised trained to obtain a belt monitoring model.

4. The method according to claim 1, wherein If the target conveyor belt is in a belt-torn state, the identifying of abnormal state parameters of the target conveyor belt includes: Performing inverse projection transformation processing on the target running state image to obtain a projection correction image of the target conveyor belt, wherein the projection correction image is the running state image of the target conveyor belt after eliminating perspective error; Extracting the belt tear region in the projected correction image and performing binarization processing on the belt tear region to obtain a binarized image of the belt tear region, wherein the binarized image includes pixel distribution characteristics of the belt crack; Based on the pixel distribution characteristics of the belt crack in the binary image, the crack length and crack width of the belt crack are determined as abnormal state parameters of the target conveyor belt.

5. The method according to claim 1, wherein If the target conveyor belt is in a belt deviation state, the identifying of abnormal state parameters of the target conveyor belt includes: Performing inverse projection transformation processing on the target running state image to obtain a projection correction image of the target conveyor belt; Based on the projected correction image, the exposed areas of the rollers on both sides of the target conveyor belt are determined, and based on the exposed areas, the belt offset direction and belt offset percentage of the belt are determined as abnormal state parameters of the target conveyor belt.

6. The method according to claim 5, characterized in that The determining of the belt deviation direction and the belt deviation percentage of the belt based on the exposed area includes: comparing exposed areas of rollers on both sides of the target conveyor belt, and if the exposed area of the first roller is greater than the exposed area of the second roller, determining that the target conveyor belt is offset toward the side where the second roller is located, the first roller and the second roller being located on both sides of the target conveyor belt, respectively; If the exposed area of the first roller is smaller than the exposed area of the second roller, it is determined that the target conveyor belt is offset toward the side where the first roller is located; When the target conveyor belt deviates toward the side where the second roller is located, the belt deviation percentage is determined by the following first formula: When the target conveyor belt deviates toward the side where the first roller is located, the belt deviation percentage is determined by the following second formula: Wherein, μ represents the belt deviation percentage when the target conveyor belt deviates toward the side where the second roller is located, μ ′ It represents the belt deviation percentage when the target conveyor belt deviates toward the side where the first roller is located, D1 represents the exposed area of the first roller, and D2 represents the exposed area of the second roller.

7. The method according to claim 1, characterized in that If the target conveyor belt is in a foreign matter mixed state, the identifying of abnormal state parameters of the target conveyor belt further includes: Performing inverse projection transformation processing on the target running state image to obtain a projection correction image of the target conveyor belt; determining shape data and size data of a target foreign object mixed on the target conveyor belt based on the projected corrected image; Retrieving a foreign object type library, and matching a foreign object type similar to the shape data and size data of the target foreign object from the foreign object type library as the foreign object type of the target foreign object; The foreign object type, shape data and size data of the target foreign object are used as abnormal state parameters of the target conveyor belt.

8. The method according to claim 7, characterized in that The method further comprises: If the foreign body type library does not contain the same foreign body type as the target foreign body in shape and size data, the shape and size data of the target foreign body are written into the foreign body type library as the shape and size data of the new foreign body type.

9. The method according to claim 7, characterized in that The method further comprises: If the target foreign object is a level one foreign object, the target conveyor belt is controlled to stop running and a level one alarm is issued to the operation and maintenance platform. The level one foreign object is a foreign object that can cause the conveyor belt to tear; If the target foreign object is a secondary foreign object, a secondary alarm is issued to the operation and maintenance platform. The secondary foreign object is a foreign object that will not cause the conveyor belt to tear. The severity of the primary alarm is greater than the severity of the secondary alarm.

10. An electronic device, characterized in that: The electronic device includes one or more processors and one or more memories, wherein at least one program code is stored in the one or more memories, and the at least one program code is loaded and executed by the one or more processors to implement the method according to any one of claims 1 to 9.

Images