Conveyor belt internal damage identification method based on X-ray detection

Abstract

The invention relates to the technical field of image processing, in particular to a conveyor belt internal damage identification method based on X-ray detection, which comprises the following steps: acquiring a preprocessed image of an X-ray image in a conveyor belt in real time; constructing a gray level histogram of the preprocessed image, removing background pixel points to obtain a target image only containing steel wire rope pixel points, and marking the steel wire rope pixel points as main pixel points; constructing a rectangular coordinate system of the target image; calculating the abnormal probability of the main pixel point; determining a bending area; the bending abnormal degree of the bending area is calculated; determining a fracture abnormity degree; taking a normalization result of the sum of the bending abnormal degree and the fracture abnormal degree as the abnormal degree of the preprocessed image; and according to the abnormal degree, identification of the internal damage of the conveying belt based on X-ray detection is realized. According to the method, the tiny damage can be accurately identified, the accuracy of identifying the internal damage of the conveying belt is improved, and the equipment fault risk can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and more specifically, to a method for identifying internal damage of a conveyor belt based on X-ray detection. Background Art

[0002] Conveyor belts are an indispensable part of the coal mining industry's conveying system. As conveyor belts need to continuously transport coal and other materials, long-term friction and material impact can cause damage to the steel wire rope inside the conveyor belt, or even bend and break. Therefore, it is necessary to identify whether there is damage inside the conveyor belt. X-ray detection technology has the characteristics of real-time scanning imaging and non-contact, and is widely used in the detection of internal damage to conveyor belts. This technology can provide detailed images of the internal structure of the conveyor belt and help identify minor damage, such as broken ends of the steel wire rope, joint conditions, surface damage, etc.

[0003] The patent document with announcement number CN117110330B discloses a conveyor belt flaw detection method, device, equipment and storage medium, the method comprising acquiring a first image transmitted by an X-ray device; acquiring a reference image; judging whether there is a color difference area on the first image based on comparison between the reference image and the first image; if the color difference area exists on the first image, marking the first image, acquiring a first distance value between the color difference area and the initial position, and controlling a display device to display the first distance value and the color difference area.

[0004] However, the above patent document uses color difference for judgment, which cannot solve the problem that the steel wire rope will gradually lose its original straightness, bend or even break due to long-term heavy pressure and use. Since the steel wire ropes are densely and complexly intertwined inside the conveyor belt, it is difficult to detect the degree of damage inside the conveyor belt; and it is difficult to accurately identify these subtle damages using traditional detection methods, resulting in inaccurate identification of the degree of damage inside the conveyor belt, thus ignoring the potential safety hazards of the conveyor belt. Summary of the invention

[0005] In order to solve the problem that it is difficult to accurately identify these subtle damages using traditional detection methods, resulting in inaccurate identification of the damage degree inside the conveyor belt, thereby ignoring the potential safety hazards of the conveyor belt, the present invention proposes a conveyor belt internal damage identification method based on X-ray detection, the method comprising the following steps: The preprocessed image of the X-ray image inside the conveyor belt is obtained in real time; the grayscale histogram of the preprocessed image is constructed, and the background pixels are removed to obtain a target image containing only the wire rope pixels, and the wire rope pixels are recorded as the main pixels; the rectangular coordinate system of the target image is constructed to obtain the angles between the main pixels; the abnormal probability of the main pixel is calculated based on the running direction angle of the conveyor belt in the target image and the sum of the angles between any main pixel and the adjacent pixel pair, wherein the adjacent pixel pair is two pixels that do not belong to the same wire rope as the main pixel and are adjacent to the main pixel; the main pixel whose abnormal probability is greater than the abnormal threshold is recorded as a curved pixel; according to the curved image The curved pixels are clustered according to the gradient values ​​and abnormal probabilities of the pixels to obtain several curved areas; the degree of the curved abnormality of the curved area is calculated based on the proportion of curved pixels in all main pixels in any curved area and the mean value of the abnormal probability of the curved pixels; the main pixels in the grayscale histogram whose grayscale values ​​are within the fracture range are recorded as fracture pixels, and the ratio of the number of fracture pixels to the number of background pixels in the preprocessed image is taken as the degree of fracture abnormality; the normalized result of the sum of the bending abnormality degree and the fracture abnormality degree is taken as the degree of abnormality of the preprocessed image; according to the size of the abnormality degree, the internal damage recognition of the conveyor belt based on X-ray detection is realized.

[0006] By calculating the abnormal probability of the main pixel point, possible abnormal situations can be identified in time. By comparing the abnormal probability with the set threshold, it is possible to more effectively determine which pixels are bent pixels, thereby discovering potential safety hazards in advance; by clustering the bent pixels, the damaged area can be identified centrally, reducing noise interference and improving analysis efficiency; by quantifying the degree of fracture abnormality, the damage assessment is made more scientific and operational, providing data support for decision-making; by early detection and identification of conveyor belt damage, downtime and maintenance costs can be reduced, thereby reducing overall maintenance costs and improving production efficiency; the X-ray-based detection method is suitable for various types of conveyor belts and can be flexibly applied to different industrial environments, improving the versatility and adaptability of the method.

[0007] Furthermore, the pre-processed image is obtained by filtering and denoising the X-ray image using a mean filtering algorithm, and then equalizing the denoised X-ray image using a histogram equalization algorithm to obtain the pre-processed image.

[0008] The preprocessed image can improve the accuracy and efficiency of subsequent damage identification through mean filtering and histogram equalization.

[0009] Furthermore, constructing the rectangular coordinate system of the target image includes: taking any of the four corners of the target image as the coordinate origin of the rectangular coordinate system, and extending them along the boundary of the target image as the horizontal and vertical coordinates of the rectangular coordinate system, respectively, and the horizontal and vertical coordinates increase successively.

[0010] Furthermore, the abnormal probability satisfies the following relationship: ; In the formula, The target image The abnormal probability of the main pixel point is is the running direction angle of the conveyor belt in the target image, For the The sum of the angles between the main pixel and the adjacent pixel pairs, For the The main pixel and the adjacent pixel pair The angle between the main pixels, For the The main pixel and the adjacent pixel pair The angle between the main pixels, is the normalization function, is the absolute value symbol.

[0011] By introducing angle information, the angle information of adjacent pixels provides geometric features about the image content, thereby improving the ability to identify abnormal situations; by calculating the angle between adjacent main pixels, the morphological changes of the wire rope during operation can be effectively captured, which helps to timely discover potential defects or abnormal situations.

[0012] Furthermore, the method of clustering the curved pixels according to the gradient values ​​and abnormal probabilities of the curved pixels to obtain a plurality of curved regions includes: calculating the probability that any two curved pixels are in the same curved region. , and Respectively and The gradient value of the curved pixel, and Respectively and The abnormal probability of curved pixels is is the natural exponential function, is the absolute value symbol; in response to If the value is greater than or equal to the bending threshold, the two bending pixels are considered to belong to the same bending region. The probability that all the bending pixels belong to the same bending region is obtained, and they are connected by region growing to obtain several bending regions.

[0013] By combining the gradient value and anomaly probability of curved pixels, the characteristics of the pixels can be more comprehensively characterized, thereby improving the accuracy of curved area recognition and helping to reduce the probability of misjudging unrelated or dissimilar pixels as the same curved area. By using the region growing algorithm, pixels identified as the same curved area can be efficiently connected, significantly reducing calculation time and complexity.

[0014] Furthermore, the degree of bending abnormality of the bending area satisfies the following relationship: ; In the formula, The target image The degree of curvature anomaly in the curvature region is the number of bending regions, The target image The number of curved pixels in the curved area, is the number of all primary pixels in the target image, The target image The mean abnormal probability of curved pixels in the curved area, is the normalization function.

[0015] By calculating the degree of bending anomaly of each curved area, we can effectively identify distinctive or abnormal areas in the image and improve the accuracy of anomaly detection. The analysis of curved areas can provide more detailed anomaly detection information, and the degree of bending anomaly of the curved area not only considers the number of curved pixels, but also combines the mean of the anomaly probability, so that more accurate judgments can be made for specific areas.

[0016] Furthermore, the method for obtaining the fracture range is: The gray value at the peak farthest from the coordinate origin in the gray histogram is recorded as the fracture gray value, and the preset value range centered on the fracture gray value is the fracture range.

[0017] Furthermore, the method of realizing the identification of internal damage of the conveyor belt based on X-ray detection includes: in response to the abnormality degree being not less than the warning threshold, determining that the conveyor belt area corresponding to the current X-ray image is abnormal, and issuing a warning prompt, thereby completing the identification of internal damage of the conveyor belt based on X-ray detection.

[0018] The present invention has the following beneficial effects: The present invention is through Preprocessing of X-ray images can effectively eliminate background noise and focus only on the main pixel points on the wire rope, thereby improving the accuracy of identifying subtle damage and reducing the limitations of traditional detection methods; by calculating the probability of abnormalities and the degree of abnormal fractures, the actual safety status of the conveyor belt can be more accurately evaluated and corresponding maintenance decisions can be made; by analyzing X-ray images at different time points, real-time dynamic monitoring can be carried out to track the health status of the conveyor belt, and possible damage can be discovered and handled in a timely manner; by timely and accurate identification of damage inside the conveyor belt, safety accidents caused by equipment failure can be effectively prevented, thereby improving the safety of the overall working environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] By reading the detailed description below with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals are the same or corresponding parts, wherein: Figure 1 It is a flowchart of the steps of a conveyor belt internal damage identification method based on X-ray detection according to an embodiment of the present invention.

[0020] Figure 2 It is a schematic diagram of the installation of an X-ray detection system and a conveyor belt in a method for identifying internal damage of a conveyor belt based on X-ray detection according to an embodiment of the present invention.

[0021] Figure 3 It is a module schematic diagram of an X-ray detection system of a conveyor belt internal damage identification method based on X-ray detection according to an embodiment of the present invention.

[0022] Figure 4 It is a schematic diagram of the internal steel wire rope of the conveyor belt of the conveyor belt internal damage identification method based on X-ray detection in an embodiment of the present invention; in the figure, a is the bending area and b is the broken area. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0024] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0025] See also Figure 1 , which shows an embodiment of the present invention provides a A flow chart of the steps of the method for identifying internal damage of a conveyor belt by radiographic detection, the method comprising the following steps: S1: A pre-processed image of an X-ray image of the interior of the conveyor belt is acquired in real time.

[0026] like Figure 2 As shown, the X-ray image is obtained by an X-ray detection system installed on the conveyor belt, and the X-ray detection system has functions such as real-time scanning and real-time imaging; Figure 3 As shown, the X-ray detection system is mainly composed of an X-ray light source, a high-speed X-ray detection system and upper-level software. The specific functions of the system are existing technologies and will not be described in detail here.

[0027] It should be noted that due to the different density and thickness of the steel wire rope inside the conveyor belt, and different materials have different absorption degrees for X-rays, the thicker part will absorb more X-rays, causing these areas on the image to appear darker, while the thinner part is relatively brighter. This brightness difference will interfere with the detection and identification of minor damage inside the conveyor belt. In addition, the conveyor belt is long, and the electronic components photographed during the real-time detection process work for a long time, so it is easy to introduce noise into the collected image. Therefore, the collected X-ray image needs to be preprocessed to improve the image quality and facilitate subsequent processing.

[0028] Specifically, the preprocessed image is to filter and denoise the X-ray image through a mean filtering algorithm, and then use a histogram equalization algorithm to equalize the denoised X-ray image to improve the contrast of the denoised X-ray image, make the brightness distribution in the denoised X-ray image more uniform, and make the details clearer, so as to obtain a preprocessed image.

[0029] S2: construct a grayscale histogram of the preprocessed image and remove background pixels to obtain a target image containing only wire rope pixels, and record the wire rope pixels as main pixels.

[0030] It should be noted that if Figure 4 As shown, the steel wire rope is thicker and absorbs more X-rays, so it appears black. The background pixels correspond to the rubber area of ​​the conveyor belt, which is thinner and absorbs less X-rays, so it appears gray. X-rays can pass through the broken area, so the broken area appears white.

[0031] Traverse all the pixels in the preprocessed image, obtain the grayscale value of each pixel, and construct a grayscale histogram. The horizontal axis of the grayscale histogram is the grayscale value, and the vertical axis is the number of pixels corresponding to each grayscale value. In the grayscale histogram, the black pixel represents the wire rope, which presents a higher peak value closest to the origin in the grayscale histogram. The gray pixel represents the background area, which presents a higher peak value farther from the origin in the grayscale image. The horizontal axis grayscale value corresponding to this high peak value is recorded as the background grayscale value. , then the gray value is The pixels within the range are recorded as background pixels. According to the above operation, the background pixels are removed, and the pixels that are not removed are wire rope pixels, which are recorded as main mark pixels.

[0032] Implementers can set the , for example, 5.

[0033] S3: Construct the rectangular coordinate system of the target image.

[0034] Construct the rectangular coordinate system of the target image and obtain the angle between each main pixel point.

[0035] Specifically, the rectangular coordinate system of the target image is constructed, including: Any of the four corners of the target image is used as the coordinate origin of the rectangular coordinate system, and is extended along the boundary of the target image as the horizontal and vertical coordinates of the rectangular coordinate system, and the horizontal and vertical coordinates increase in sequence.

[0036] S4: Calculate the abnormal probability of the main pixel.

[0037] It should be noted that the main pixel point is the pixel point on the wire rope. Since the wire rope is laid along the direction of the conveyor belt, the theoretical direction of the wire rope is horizontal. Due to long-term heavy pressure and use, the wire rope will gradually lose its original straightness and bend, so the main texture direction can be obtained. At the pixel point where there is a bend, there will be a certain angle with the direction of the conveyor belt movement, and the abnormal probability of the main pixel point is calculated based on this angle.

[0038] The abnormal probability of the main pixel is calculated based on the running direction angle of the conveyor belt in the target image and the sum of the angles between any main pixel and an adjacent pixel pair, where the adjacent pixel pair is two pixel points that do not belong to the same wire rope as the main pixel and are adjacent to the main pixel.

[0039] The running direction angle of the conveyor belt is the angle between the conveyor belt and the horizontal plane. For example, when the conveyor belt is set horizontally, the angle is 0°, and when the conveyor belt is set vertically, the angle is 90°.

[0040] Specifically, the abnormal probability satisfies the following relationship: ; In the formula, The target image The abnormal probability of the main pixel point is is the running direction angle of the conveyor belt in the target image, For the The sum of the angles between the main pixel and the adjacent pixel pairs, For the The main pixel and the adjacent pixel pair The angle between the main pixels, For the The main pixel and the adjacent pixel pair The angle between the main pixels, is the normalization function, is the absolute value symbol.

[0041] in, Indicates The main pixel and the adjacent pixel pair The angle between the main pixels, Indicates The main pixel and the adjacent pixel pair The angle between the main pixels, It means taking the average of the above sums and taking this average as the first The angle between the main pixel and the horizontal direction; when the difference between the conveyor belt running direction angle in the target image and the angle between the main pixel and the horizontal direction The larger the value, the The greater the abnormal probability of the number of main pixels, the greater the abnormal probability of the The larger the number of main pixels, the greater the probability that they belong to the curved area, and vice versa.

[0042] S5: Determine the bending area.

[0043] The main pixel points whose abnormal probability is greater than the abnormal threshold are recorded as curved pixels; the curved pixels are clustered according to the gradient values ​​and abnormal probabilities of the curved pixels to obtain several curved areas.

[0044] Implementers can set the abnormal threshold according to specific implementation conditions, for example, 0.6.

[0045] It should be noted that the curved pixels in the same texture and with smaller differences in corresponding abnormal probabilities are divided into the same curved area. Whether they are in the same texture can be reflected by the difference in gradient values ​​between the two curved pixels. The degree of curvature between the two curved pixels in the same curved area, i.e., the abnormal probability value, should be as close as possible.

[0046] Specifically, calculate the probability that any two curved pixels belong to the same curved region , and Respectively and The gradient value of the curved pixel, and Respectively and The abnormal probability of curved pixels is is the natural exponential function, is the absolute value symbol; In response to If the value is greater than or equal to the bending threshold, the two bending pixels are considered to belong to the same bending area. The probability that all curved pixels belong to the same curved region is obtained, and they are connected by region growing to obtain several curved regions.

[0047] The method for obtaining the bending threshold is as follows: calculate the angle between the first main pixel point and any other main pixel points on the same wire rope to satisfy: , For the The first The main pixel and the The angle between the main pixels, and Respectively The first The ordinate and abscissa of the main pixel and Respectively The first The ordinate and abscissa of the main pixel point, is an inverse trigonometric function; the average of the angles between the first pixel and all other pixels on the same wire rope is taken as the texture direction of the wire rope, and the third quartile of the texture directions of all wire ropes is taken as the main texture direction of the preprocessed image; the bending threshold is set according to the difference between the main texture direction and the running direction of the conveyor belt: the greater the difference between the two, the greater the bending threshold; the smaller the difference between the two, the smaller the threshold. For example, when the difference between the main texture direction of the target image and the running direction of the conveyor belt is When the bending threshold ; When the main texture direction of the target image differs from the running direction of the conveyor belt When the bending threshold , where 15 is an exemplary value, which can be set by implementers according to specific implementation conditions.

[0048] S6: Calculate the degree of bending abnormality of the bending area.

[0049] It should be noted that the degree of bending abnormality in the bending area can be obtained through the abnormal probability of each bending pixel point in the bending area.

[0050] Based on the proportion of curved pixels in any curved area to all main pixels and the mean value of the abnormal probability of curved pixels, the degree of curvature abnormality of the curved area is calculated.

[0051] Specifically, the degree of bending abnormality of the bending area satisfies the following relationship: ; In the formula, The target image The degree of curvature anomaly in the curvature region is the number of bending regions, The target image The number of curved pixels in the curved area, is the number of all primary pixels in the target image, The target image The mean abnormal probability of curved pixels in the curved area, is the normalization function.

[0052] Among them, when The larger the value of The larger the value, the The larger the bending area, the greater the bending anomaly, and vice versa.

[0053] S7: Determine the degree of fracture abnormality.

[0054] The main pixel points in the grayscale histogram whose grayscale values ​​are within the fracture range are recorded as fracture pixels. The fracture area is obtained by using the connected domain analysis algorithm for the fracture pixels. Figure 4 , the ratio of the number of fracture pixels to the number of background pixels in the preprocessed image is taken as the fracture abnormality degree, that is, the fracture abnormality degree of the fracture area.

[0055] It should be noted that the steel wire rope is thicker and absorbs more X-rays, which is black. The background pixel corresponds to the rubber area of ​​the conveyor belt, which is thinner and absorbs less X-rays, which is gray. X-rays can pass through the broken area, so the broken area is white. Therefore, in the grayscale histogram, the grayscale value corresponding to the peak farthest from the origin is the grayscale value of the broken pixel.

[0056] Specifically, the method for obtaining the fracture range is: The gray value at the peak farthest from the coordinate origin in the gray histogram is recorded as the fracture gray value , then the preset numerical range centered on the fracture gray value is the fracture range.

[0057] Among them, the fracture range is recorded as the fracture gray value ,but is the fracture range, then the gray value is The main pixel point within the range is the broken pixel point.

[0058] Implementers can set the , for example, 5.

[0059] S8: taking the normalized result of the sum of the bending abnormality degree and the fracture abnormality degree as the abnormality degree of the preprocessed image.

[0060] S9: Based on the degree of abnormality, internal damage identification of the conveyor belt can be achieved based on X-ray detection.

[0061] Specifically, the method for realizing the internal damage identification of the conveyor belt based on X-ray detection includes: In response to the abnormality level being not less than the warning threshold, the conveyor belt area corresponding to the current X-ray image is determined to be abnormal, reflecting that the internal damage of the conveyor belt detected at the current moment is relatively serious, and the conveyor belt needs to be stopped. An early warning is issued to remind maintenance personnel to repair the corresponding conveyor belt, thereby completing the identification of internal damage to the conveyor belt based on X-ray detection.

[0062] Implementers can set the warning threshold according to specific implementation circumstances, for example, 0.7.

[0063] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A conveyor belt internal damage identification method based on X-ray detection, characterized in that: include: Real-time acquisition of pre-processed images of X-ray images of the interior of the conveyor belt; Construct a grayscale histogram of the preprocessed image, and remove background pixels to obtain a target image containing only wire rope pixels, and record the wire rope pixels as main pixels; construct a rectangular coordinate system of the target image to obtain the angles between the main pixels; calculate the abnormal probability of the main pixel based on the running direction angle of the conveyor belt in the target image and the sum of the angles between any main pixel and an adjacent pixel pair, where the adjacent pixel pair is two pixels that do not belong to the same wire rope as the main pixel and are adjacent to the main pixel; record the main pixel whose abnormal probability is greater than the abnormal threshold as a curved pixel; The curved pixels are clustered according to their gradient values ​​and abnormal probabilities to obtain several curved regions; the degree of curvature abnormality of the curved region is calculated based on the proportion of curved pixels in any curved region to all main pixels and the mean value of the abnormal probability of curved pixels; the main pixels whose grayscale values ​​in the grayscale histogram are within the fracture range are recorded as fracture pixels, and the ratio of the number of fracture pixels to the number of background pixels in the preprocessed image is taken as the degree of fracture abnormality; the normalized result of the sum of the curvature abnormality degree and the fracture abnormality degree is taken as the degree of abnormality of the preprocessed image; According to the degree of abnormality, internal damage identification of the conveyor belt can be realized based on X-ray detection.

2. The method for identifying internal damage of a conveyor belt based on X-ray detection according to claim 1, characterized in that: The pre-processed image is obtained by filtering and denoising the X-ray image using a mean filtering algorithm, and then equalizing the denoised X-ray image using a histogram equalization algorithm to obtain the pre-processed image.

3. The method for identifying internal damage of a conveyor belt based on X-ray detection according to claim 1, characterized in that: The rectangular coordinate system of the target image is constructed, comprising: Any of the four corners of the target image is used as the coordinate origin of the rectangular coordinate system, and is extended along the boundary of the target image as the horizontal and vertical coordinates of the rectangular coordinate system, and the horizontal and vertical coordinates increase in sequence.

4. The method for identifying internal damage of a conveyor belt based on X-ray detection according to claim 1, characterized in that: The abnormal probability satisfies the following relationship: ； In the formula, The target image The abnormal probability of the main pixel point is is the running direction angle of the conveyor belt in the target image, For the The sum of the angles between the main pixel and the adjacent pixel pairs, For the The main pixel and the adjacent pixel pair The angle between the main pixels, For the The main pixel and the adjacent pixel pair The angle between the main pixels, is the normalization function, is the absolute value symbol.

5. The method for identifying internal damage of a conveyor belt based on X-ray detection according to claim 1, characterized in that: The curved pixels are clustered according to the gradient values ​​and abnormal probabilities of the curved pixels to obtain several curved areas, including: Calculate the probability that any two curved pixels belong to the same curved area , and Respectively and The gradient value of the curved pixel, and Respectively and The abnormal probability of curved pixels is is the natural exponential function, is the absolute value symbol; In response to is not less than the bending threshold, and the two bending pixels are considered to belong to the same bending area; The probability that all curved pixels belong to the same curved region is obtained, and they are connected by region growing to obtain several curved regions.

6. The method for identifying internal damage of a conveyor belt based on X-ray detection according to claim 1, characterized in that: The degree of abnormal bending in the bending area satisfies the following relationship: ； In the formula, The target image The degree of curvature anomaly in the curvature region is the number of bending regions, The target image The number of curved pixels in the curved area, is the number of all primary pixels in the target image, The target image The mean abnormal probability of curved pixels in the curved area, is the normalization function.

7. The method for identifying internal damage of a conveyor belt based on X-ray detection according to claim 1, characterized in that: The method for obtaining the fracture range is: The gray value at the peak farthest from the coordinate origin in the gray histogram is recorded as the fracture gray value, and the preset value range centered on the fracture gray value is the fracture range.

8. The method for identifying internal damage of a conveyor belt based on X-ray detection according to claim 1, characterized in that: The method for realizing internal damage identification of a conveyor belt based on X-ray detection includes: In response to the abnormality level being not less than the warning threshold, the conveyor belt area corresponding to the current X-ray image is determined to be abnormal, and a warning prompt is issued to complete the identification of internal damage to the conveyor belt based on X-ray detection.

Images