A conveyor belt breakage detection method

Abstract

The application discloses a kind of machine belt breakage detection methods, comprising the following steps: S1: construct initial feature library;S2: during the operation of belt conveyor, frame rate R is collected in real time with the image data of belt by patrol machine;S3: the feature vector f of belt image is obtained by step S1, then the distance of belt image feature vector and normal belt feature clustering center is calculated, whether the belt in belt image is damaged according to the result is judged;S4: dynamic updating is carried out to feature vector library.The beneficial effects of the present application are: by real-time image acquisition and unsupervised learning analysis, without a large number of labeled data, the damage condition of the belt can be quickly identified, potential damage risks can be found in time, the burden of manual inspection can be reduced, and the monitoring efficiency can be improved.Meanwhile, the feature vector library is dynamically updated, so that the influence caused by the gradual change of the belt appearance during long-term operation can be effectively overcome, and the robustness of the method in practical application can be ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of belt conveyors, in particular to a belt conveyor belt damage detection method. BACKGROUND

[0002] Belt conveyors are widely used in mining, coal, steel, power, chemical and other industries. As a key material conveying equipment, it undertakes continuous and stable transportation tasks. However, due to various factors, the belt may crack, tear, and have various damage forms such as flanging during long-term operation. These damages not only lead to reduced transportation efficiency and increased maintenance costs, but also may cause equipment downtime, material loss, and even safety accidents. Therefore, real-time and accurate monitoring of the health status of the belt is the key to ensuring production safety and improving equipment reliability.

[0003] Traditional belt damage detection methods rely on manual inspection and rule-based monitoring systems. Manual inspection can detect obvious damage, but it is time-consuming and prone to miss small damage, which cannot meet the needs of efficient, all-weather, and uninterrupted monitoring. Rule-based monitoring systems usually require manual setting of fault modes and alarm thresholds, but this method has poor flexibility and is difficult to adapt to complex changes in different working environments or operating states, and cannot effectively identify unforeseen damage patterns. In addition, existing computer vision-based detection methods mostly rely on supervised learning and require a large amount of labeled data for training. However, it is difficult to obtain labeled data for belt damage, especially in actual production, so it is difficult to be widely applied in actual production environment. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art and provide a belt conveyor belt damage detection method.

[0005] The purpose of the present application is achieved by the following technical solution: a belt conveyor belt damage detection method, comprising the following steps:

[0006] S1: constructing an initial feature library;

[0007] S2: during the operation of the belt conveyor, the inspection machine acquires image data of the belt at a frame rate R in real time;

[0008] S3: obtaining the feature vector f of the belt image through step S1, then calculating the distance between the belt image feature vector and the normal belt feature clustering center, and judging whether the belt in the belt image is damaged according to the result;

[0009] S4: dynamically updating the feature vector library.

[0010] Preferably, step S1 further comprises the following steps:

[0011] S11: Collecting N normal belt image data through the camera of the inspection machine;

[0012] S12: Filtering the image content of the non-belt area in the image;

[0013] S13: Extracting features from the filtered image;

[0014] S14: Constructing a feature vector library of the normal belt image set and calculating the statistical features of the normal belt image feature set.

[0015] Preferably, step S12 further comprises the following steps:

[0016] S12.1: The image is segmented into a belt area by a belt segmentation algorithm,

[0017] B(x, y) = S(I(x, y));

[0018] Where (x, y) is the image before filtering; S is the belt segmentation algorithm; B(x, y) is the mask of the belt area;

[0019] S12.2: The non-belt area in the image is set to zero to obtain an image containing only the belt part,

[0020] I filtered (x, y) = I(x, y)·B(x, y);

[0021] Where I filtered (x, y) is the image after filtering.

[0022] Preferably, in step S13, the features of the filtered image are extracted by an image feature extraction model.

[0023] Preferably, in step S14, calculating the statistical features of the normal belt image feature set further comprises the following steps:

[0024] S14.1: Calculating the cluster center of the normal class,

[0025]

[0026] S14.2: Calculating the average distance of the sample to the cluster center,

[0027]

[0028] S14.3: Calculating the standard deviation of the sample to the cluster center,

[0029]

[0030] Preferably, in step S3, the formula for calculating the distance between the belt image feature vector and the normal belt feature cluster center is:

[0031] d = ||f - f c ||.

[0032] Preferably, in step S3, the standard for judging whether the belt in the belt image is damaged is:

[0033]

[0034] where 0 represents a normal belt and 1 represents an abnormal belt.

[0035] Preferably, in step S4, if the belt is judged to be in a normal state by step S3, the earliest one of the normal belt features is deleted from the feature library representing the normal belt in chronological order, and then the feature vector f calculated by step S3 is added to the feature library representing the normal belt.

[0036] The present application has the following advantages: the present application can quickly identify the damage condition of the belt, discover potential damage risks in a timely manner, reduce the burden of manual inspection, improve monitoring efficiency, and dynamically update the feature vector library, thereby effectively overcoming the influence of gradual changes in the appearance of the belt during long-term operation, and further ensuring that the method has strong robustness in actual application. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 FIG. 1 is a structural schematic diagram of a belt damage detection method flowchart;

[0038] Figure 2 FIG. 3 is a structural schematic diagram of an initial feature library construction method. DETAILED DESCRIPTION

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0040] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the scope of protection of the present application.

[0041] It should be noted that the embodiments and features in the present application can be combined with each other in the case of no conflict.

[0042] It should be noted that similar reference numerals and letters refer to like items in the following drawings, and thus, once an item is defined in one drawing, it is not necessary to further define and explain it in the subsequent drawings.

[0043] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship commonly understood by those skilled in the art, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second" and the like are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0044] In the description of the present application, it should also be noted that unless otherwise specifically defined and limited, the terms "arrangement", "installation", "connection", "connection" should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0045] In the present embodiment, as shown in Figure 1 A belt conveyor damage detection method, comprising the following steps:

[0046] S1: Construct an initial feature library;

[0047] S2: During the operation of the belt conveyor, the inspection machine acquires image data of the belt at a frame rate R in real time;

[0048] S3: Obtain the feature vector f of the belt image through step S1, then calculate the distance between the belt image feature vector and the normal belt feature clustering center, and judge whether the belt in the belt image is damaged according to the result;

[0049] S4: Dynamically updating the feature vector library. Through real-time image acquisition and unsupervised learning analysis, without a large amount of labeled data, the damage condition of the belt can be quickly identified, potential damage risks can be found in time, the burden of manual inspection can be reduced, the monitoring efficiency can be improved, and the feature vector library can be dynamically updated, thereby effectively overcoming the influence of the gradual change of the belt appearance in the long-term operation process, and further ensuring that the method has strong robustness in actual application.

[0050] Further, as shown in Figure 2 step S1, the following steps are further included:

[0051] S11: Collecting N normal belt image data through the camera of the inspection machine;

[0052] S12: Filtering the image content of the non-belt area in the image; specifically, the following steps are further included:

[0053] S12.1: The image is segmented into a belt area by a belt segmentation algorithm,

[0054] B(x, y) = S(I(x, y));

[0055] where (x, y) is the image before filtering; S is the belt segmentation algorithm; B(x, y) is the mask of the belt area;

[0056] S12.2: The non-belt area in the image is set to zero to obtain an image containing only the belt part,

[0057] I filtered (x, y) = I(x, y)·B(x, y);

[0058] where I filtered (x, y) is the image after filtering, that is, only the pixel values of the belt area are retained, and the pixel values of the remaining areas are set to zero.

[0059] S13: Extracting features from the filtered image;

[0060] S14: Constructing a feature vector library of the normal belt image set and calculating the statistical features of the normal belt image feature set.

[0061] Further, in step S13, the features of the filtered image are extracted by an image feature extraction model. Specifically, a feature extraction algorithm is used to extract features from the filtered image, and the specific feature extraction algorithm is a general-purpose image feature extraction model with strong generalization ability and light weight,

[0062] f i = F(I i ).

[0063] In the embodiment, in step S14, the calculation of the statistical features of the normal belt image feature set further comprises the following steps:

[0064] S14.1: calculating the cluster centers of the normal category,

[0065]

[0066] S14.2: calculating the average distance of the sample to the cluster center,

[0067]

[0068] S14.3: calculating the standard deviation of the sample to the cluster center,

[0069]

[0070] In the embodiment, in step S3, the formula for calculating the distance between the belt image feature vector and the normal belt feature cluster center is:

[0071] d = ||f - f c ||.

[0072] Further, in step S3, the criterion for judging whether the belt in the belt image is damaged is:

[0073]

[0074] wherein 0 represents a normal belt and 1 represents an abnormal belt.

[0075] Still further, in step S4, if the belt is judged to be in a normal state by step S3, the earliest one of the normal belt features is deleted from the feature library representing the normal belt in time sequence, and then the feature vector f calculated by step S3 is added to the feature library representing the normal belt. Specifically, since the condition of the belt will change slowly as the belt runs for a long time, a probability p is set as a hyperparameter to update the queue of the feature library with a probability p each time, and preferably, p is set to 0.2, and the parameter can be adjusted according to the frame rate of the camera and the running speed of the belt. When the belt is judged to be in a normal state by step S3, the earliest one of the normal belt features is deleted from the feature library representing the normal belt in time sequence, and then the feature vector f calculated by step S3 is added to the feature library representing the normal belt.

[0076] Although the present application has been described in detail with reference to the foregoing embodiments, the technical solutions recorded in the foregoing embodiments can be modified, or some of the technical features can be replaced by equivalent features, by those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting damage to conveyor belts, characterized in that: Includes the following steps: S1: Construct the initial feature library; S2: During the operation of the belt conveyor, the inspection machine is monitored at a frame rate. Real-time acquisition of image data of the belt; S3: Obtain the feature vector of the belt image through step S1. Then calculate the distance between the feature vector of the belt image and the cluster center of the normal belt features, and determine whether the belt in the belt image is damaged based on the result; S4: Dynamically update the feature vector library; In step S4, dynamic update specifically involves setting a probability. As a hyperparameter, it allows each time to be based on probability The feature library queue is updated, and hyperparameters are adjusted based on the camera's frame rate and belt speed. When the belt is determined to be in a normal state in step S3, the earliest feature in the normal belt features is first deleted from the feature library representing normal belts in chronological order. Then, the feature vector calculated in step S3 is updated. Add it to the feature library representing a normal belt.

2. The method for detecting belt breakage according to claim 1, characterized in that: Step S1 further includes the following steps: S11: Data collected via cameras on the inspection machine. Zhang's normal belt image data; S12: Filter the image content in non-belt areas of the image; S13: Extract features from the filtered image; S14: Construct a feature vector library for a set of normal belt images and calculate the statistical features of the normal belt image feature set.

3. The method for detecting belt breakage according to claim 2, characterized in that: Step S12 further includes the following steps: S12.1: The image is segmented into belt regions using a belt segmentation algorithm. ； in, The image is before filtering; S represents the belt segmentation algorithm. For the mask of the belt area; S12.2: Set the non-belt regions in the image to zero to obtain an image containing only the belt portion. ； in, This is the filtered image.

4. The method for detecting belt breakage according to claim 3, characterized in that: In step S13, features are extracted from the filtered image using an image feature extraction model.

5. The method for detecting belt breakage according to claim 4, characterized in that: In step S14, calculating the statistical features of the normal belt image feature set further includes the following steps: S14.1: Calculate the cluster centers of the normal categories. ； S14.2: Calculate the average distance of a sample from its cluster center. ； S14.3: Calculate the standard deviation of the sample distance from the cluster center. 。 6. The method for detecting belt breakage according to claim 5, characterized in that: In step S3, the formula for calculating the distance between the feature vector of the belt image and the cluster center of the normal belt features is as follows: 。 7. The method for detecting belt breakage according to claim 6, characterized in that: In step S3, the criterion for determining whether the belt in the belt image is damaged is as follows: ； In this system, 0 represents a normal belt and 1 represents a faulty belt.

Images