Intelligent identification system and method for defects of oil pipeline

Abstract

The present application relates to the technical field of pipeline defect identification, and particularly relates to an intelligent identification system and method for defects of an oil pipeline, the method comprising: acquiring a plurality of images of the oil pipeline to determine feature extraction positions and potential peeling areas; determining whether the potential peeling area is a bulge according to an offset distance; acquiring a front view of the potential peeling area based on the potential peeling area being determined as a bulge to determine a planar range of the bulge, so as to determine a coincidence degree of the planar range and the potential peeling area; determining adjustment of an illumination angle based on the coincidence degree, so as to determine a bulge height, and determining an expected peeling duration according to the bulge height; and the system comprising: a pipeline detection robot and a control center, the pipeline detection robot being used to acquire image information, and the control center being used to analyze the image information acquired by the pipeline robot to determine the bulge and the expected peeling duration corresponding to the bulge. The present application improves the identification accuracy of the bulge and thus improves the detection accuracy of defects.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pipeline defect detection, and in particular to an intelligent identification system and method for defects of an oil pipeline. BACKGROUND

[0002] As the core infrastructure for energy transportation, oil pipelines are prone to defects such as bulges and cracks due to internal and external corrosion, mechanical stress and geological activities. Bulges are difficult to detect because of their diverse shapes and frequent confusion with surface attachments such as rust spots, weld protrusions and anticorrosion patches. Traditional manual inspection relies on experience and is inefficient with a high rate of missed detection of up to 30%. Conventional non-destructive testing techniques such as ultrasonic thickness measurement and eddy current testing can locate abnormalities but are difficult to implement large-scale rapid screening, especially lacking quantitative analysis capabilities for surface topography features.

[0003] In recent years, automated detection technology based on machine vision has developed rapidly, but still faces multiple challenges in practical applications: morphological similarity interference: bulges, weld reinforcement and anticorrosion material accumulation all present local protruding features in two-dimensional images, and traditional edge detection algorithms such as the Canny operator are prone to misjudgment due to texture noise; environmental complexity interference: false images formed by oil stains and rust on the pipeline surface are highly overlapped with real bulges in color and gray scale distribution, and threshold segmentation methods such as the Otsu algorithm are difficult to effectively distinguish; three-dimensional feature loss: monocular cameras cannot capture the three-dimensional height information of bulges, resulting in similar profiles of concave corrosion and convex bulges in two-dimensional projections.

[0004] Chinese Patent Publication No. CN111695482A discloses a pipeline defect identification method, in which a pipeline robot transmits real-time pipeline video, samples the video into key frame images, effectively identifies and captures pipeline defects in the images, and saves the extracted defect images with positioning markers. This invention uses the basic idea of DefectNet to solve the target defect detection as a regression problem. First, a convolutional neural network is used for feature extraction to obtain a feature map of a certain size. Then, multi-scale prediction divides the different scale feature maps into several grids. The center of the defect target falls into which grid, and which grid is responsible for predicting the defect target. Finally, target classification and bounding box regression are performed. Each grid determines the class to which the defect target belongs and adjusts the bounding box position. Compared with existing technologies, this invention improves detection accuracy and efficiency, has high innovation and practical value, and is suitable for promotion.

[0005] Therefore, the pipeline defect identification method has the following problems: This invention saves the extracted defect image location markers to obtain a feature map, and divides the feature map into several grids. The grids are responsible for identifying the defect targets in the area, thereby improving the detection efficiency. However, this invention does not demonstrate how to identify defects, and still does not solve the problem of not being able to accurately identify bulge defects. Summary of the Invention

[0006] To address this issue, the present invention provides an intelligent identification system and method for defects in oil pipelines, thereby overcoming the problem of low defect detection accuracy caused by the inability to accurately identify bulge defects in oil pipelines in the prior art.

[0007] To achieve the above objectives, in one aspect, the present invention provides an intelligent identification method for defects in oil pipelines, comprising:

[0008] Acquire several images of the oil pipeline, preprocess the images to determine several feature extraction locations, and determine potential stripping areas based on the feature extraction locations;

[0009] Increase the fluid flow rate in the potential peeling area to determine whether the potential peeling area is a bulge based on the offset distance of the shadow boundary;

[0010] Based on the result that the potential peeling area is determined to be a bulge, a front view of the potential peeling area is obtained. After preprocessing the front view, the planar range of the bulge is determined, and the degree of overlap between the planar range and the potential peeling area is determined.

[0011] The illumination angle is adjusted based on the overlap, the bulge height is determined based on the feature extraction location, and the expected peeling time of the bulge is determined based on the bulge height and the edge gradient of the bulge.

[0012] Furthermore, the process of determining the feature extraction location includes:

[0013] Determine a number of pixels with a gray value less than the first gray value, and connect adjacent positions of each pixel to form a number of pixel regions;

[0014] The pixel region is matched with a set of standard regions to determine each similarity, and the maximum similarity is determined based on each similarity.

[0015] The maximum similarity is compared with the preset similarity.

[0016] The pixel region is determined as the feature extraction location based on the comparison result where the maximum similarity is greater than or equal to the preset similarity.

[0017] Furthermore, the process of determining the potential stripping region includes:

[0018] The line closest to the light source at the feature extraction location is defined as the shadow boundary. The center point of the shadow boundary is determined, and a plane perpendicular to the center line of the oil pipeline is determined based on the center point.

[0019] Using the plane as a reference, the feature extraction locations are mirrored to form a mirror region;

[0020] The region formed by connecting the feature extraction location with the outer edge of the mirror region is the potential stripping region.

[0021] Furthermore, the process of determining the offset distance of the shadow boundary includes:

[0022] After increasing the fluid velocity, the shaded boundary is redefined and recorded as the secondary boundary.

[0023] The distance between the shadow boundary and the secondary boundary is defined as the offset distance.

[0024] Furthermore, the process of determining the properties of the potential stripping region based on the offset distance includes:

[0025] Compare the offset distance with the preset offset distance;

[0026] Based on the comparison result that the offset distance is greater than the preset distance, the potential peeling area is determined to be a bulge.

[0027] Furthermore, the process of determining whether to adjust the lighting angle includes:

[0028] Obtain a front view of the bulge, and determine the planar range of the bulge after preprocessing the front view;

[0029] The area ratio between the plane range and the feature extraction location is determined as the overlap degree;

[0030] The degree of overlap is compared with a first preset degree of overlap and a second preset degree of overlap, respectively;

[0031] Based on the comparison result that the overlap degree is less than the first preset overlap degree, it is determined to reduce the illumination angle;

[0032] Based on the comparison result that the overlap degree is greater than the second preset overlap degree, it is determined to increase the illumination angle.

[0033] Furthermore, the process of reducing the illumination angle includes:

[0034] The first difference value is obtained by subtracting the first preset overlap degree from the overlap degree.

[0035] A first angle adjustment coefficient is determined based on the first difference, and the illumination angle is reduced using the first angle adjustment coefficient.

[0036] The illumination angle is the angle between the light ray and the inner wall of the oil pipeline.

[0037] Furthermore, the process of increasing the adjustment of the illumination angle includes:

[0038] The difference between the degree of overlap and the second preset degree of overlap is used to obtain a second difference value;

[0039] A second angle adjustment coefficient is determined based on the second difference, and the illumination angle is increased by the second angle adjustment coefficient.

[0040] Furthermore, the process of determining the expected peeling time based on the bulge height and edge gradient includes:

[0041] Based on the adjusted illumination angle, the maximum spacing of the feature extraction positions is determined according to the direction parallel to the fluid flow direction. The maximum spacing and the illumination angle are then input into the calculation model to output the bulge height.

[0042] The edge gradient is determined based on the planar range of the bulge;

[0043] The estimated peeling time of the bulge is determined by combining the bulge height and the edge gradient.

[0044] On the other hand, the present invention also provides an intelligent identification system for oil pipeline defects using an intelligent identification method for oil pipeline defects, comprising:

[0045] Pipeline inspection robot, used to acquire several images of the inner wall of oil pipelines;

[0046] The control center, which is connected to the pipeline inspection robot, includes,

[0047] A region determination module is used to determine several feature extraction locations based on the image, and to determine potential stripping regions by mirroring the feature extraction locations.

[0048] The defect determination module is used to determine whether the potential peeling area is a bulge based on the offset distance of the shadow boundary, and to determine the planar range of the front view of the bulge to determine the adjustment of the illumination angle based on the degree of overlap.

[0049] The defect development prediction module is used to determine the bulge height based on the adjusted illumination angle to determine the expected peeling time of the bulge.

[0050] Compared with existing technologies, the advantages of this invention are as follows: This invention identifies and determines the feature extraction range by connecting the required pixels. A bulge will cast a shadow on the back side under illumination. Because the direction of illumination in an oil pipeline is fixed, the shadow shape formed by the bulge has a specific shape. The feature extraction position, i.e., the shadow area, can be divided according to the shape characteristics of the pixel area. Based on the determined shadow area, a mirror image is used to determine the potential peeling area, thus initially determining the bottom surface range of the potential peeling area. Further, the fluid flow rate is increased, and the offset distance of the shadow boundary is recorded. If the offset distance is greater than a preset distance, it indicates that the potential peeling area is a bulge. The bulge is caused by corrosion producing gas, which forms a cavity on the metal surface. Under the action of external force, the gas flows to resist external pressure, causing the highest point of the bulge to move. The shadow boundary always passes through the highest point of the bulge. Therefore, by judging the offset of the shadow boundary, it is determined whether the potential peeling area is a bulge. After confirming that it is a bulge, a front view is obtained to determine the degree of overlap between the planar range of the bulge and the range of the potential peeling area. A large potential peeling area indicates that the illumination angle is too small, resulting in an excessively large shadow area. Conversely, a small potential peeling area indicates that the illumination angle is too large, resulting in an excessively small shadow area. By adjusting the illumination angle, the area of ​​the shadow area can be made to be exactly half of the planar range. For a single bulge, under the premise that the area of ​​the shadow area is exactly half of the planar range, there is a one-to-one correspondence between the bulge height and the illumination angle. The bulge height can be determined through the model to solve the problem that a monocular camera cannot capture the three-dimensional height information of the bulge. The higher the bulge height, the shorter the time before the bulge breaks. Therefore, the expected peeling time of the bulge can be predicted based on the bulge height, thereby achieving accurate identification of the bulge and improving the detection accuracy of defects.

[0051] Furthermore, this invention determines several pixels by grayscale values ​​and determines pixel regions based on these pixels. This allows for initial screening of feature extraction locations based on the objective fact that bulges cast shadows under illumination. Within the initial location range, potential stripping regions are further determined. The properties of potential stripping regions are determined based on the objective fact that bulges deform under external forces. Further detection and analysis are performed on the locations identified as bulges, thereby further improving the accuracy of bulge recognition.

[0052] Furthermore, based on the condition that the shadow area can only reach half of the planar range under a specific illumination angle, the present invention inputs the illumination angle and the maximum spacing of the feature extraction position into the model to determine the bulge height. Under the condition of determining the bulge height, the expected peeling time is further determined based on the bulge height, thereby improving the detection accuracy while predicting the development trend of the bulge, and further improving the defect detection accuracy. Attached Figure Description

[0053] Figure 1 This is a flowchart of an intelligent identification method for oil pipeline defects according to an embodiment of the present invention;

[0054] Figure 2 A flowchart for determining the properties of potential stripping regions in an embodiment of the present invention;

[0055] Figure 3 A flowchart for determining whether to adjust the illumination angle in an embodiment of the present invention;

[0056] Figure 4 This is a structural block diagram of an intelligent identification system for oil pipeline defects according to an embodiment of the present invention. Detailed Implementation

[0057] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0058] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0059] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0060] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0061] Please see Figures 1-3 As shown, Figure 1 This is a flowchart of an intelligent identification method for oil pipeline defects according to an embodiment of the present invention; Figure 2 A flowchart for determining the properties of potential stripping regions in an embodiment of the present invention; Figure 3 This is a flowchart illustrating how to determine whether to adjust the illumination angle in an embodiment of the present invention.

[0062] This invention provides an intelligent method for identifying defects in oil pipelines, comprising:

[0063] Step S1: Acquire several images of the oil pipeline, preprocess the images to determine several feature extraction locations, and determine potential stripping areas based on the feature extraction locations;

[0064] Step S2: Increase the fluid flow rate in the potential peeling area to determine whether the potential peeling area is a bulge based on the offset distance of the shadow boundary.

[0065] Step S3: Based on the result that the potential peeling area is determined to be a bulge, obtain a front view of the potential peeling area, preprocess the front view to determine the planar range of the bulge, and determine the degree of overlap between the planar range and the potential peeling area;

[0066] Step S4: Adjust the illumination angle based on the overlap, determine the bulge height based on the feature extraction location, and determine the expected peeling time of the bulge based on the bulge height and the edge gradient of the bulge.

[0067] Specifically, all images in the embodiments of the present invention need to be preprocessed before analysis. The preprocessing process includes, but is not limited to, grayscale conversion, geometric transformation, denoising, filtering and image enhancement.

[0068] Specifically, the process of determining the feature extraction location includes:

[0069] Determine the first grayscale value of the defect-free location in the oil pipeline;

[0070] Determine a number of pixels with a gray value less than the first gray value, and connect adjacent positions of each pixel to form a number of pixel regions;

[0071] The pixel region is matched with a set of standard regions to determine each similarity, and the maximum similarity is determined based on each similarity.

[0072] The maximum similarity is compared with the preset similarity.

[0073] The pixel region is determined as the feature extraction location based on the comparison result where the maximum similarity is greater than or equal to the preset similarity.

[0074] The first grayscale value is the grayscale value of a simulated oil pipeline at a defect-free location under inspection conditions.

[0075] Specifically, bulges or solder joints have protrusions. When illuminated by light, these protrusions will cast shadows on the side away from the light. The gray values ​​of the shadows are relatively small. Pixels with smaller gray values ​​are connected into a region, and then overlapped according to the shape characteristics of the region to determine the feature extraction location.

[0076] Specifically, the standard region set is a set of pixel regions that simulate the shadows formed on the back side of bulges of different sizes under light illumination, and the direction of light illumination is consistent with the direction of light emitted by the pipeline inspection robot inside the pipeline.

[0077] Specifically, the first gray value is the average gray value of each pixel at the defect-free location, determined by experimental simulation. The preset similarity value range is set to [70%, 95%], and preferably 85% in this embodiment of the invention.

[0078] Understandably, the feature extraction location is the shaded area, representing the location where bulges may exist. During long-term use, the material inside the oil pipeline corrodes under continuous erosion, gradually forming tiny rust spots or patches. As corrosion deepens, the rust spots or patches gradually enlarge and form cavities or bubbles inside the metal. The gases produced by corrosion (such as hydrogen) accumulate in the cavities or bubbles inside the metal, leading to an increase in internal pressure. Under the action of internal pressure, the metal surface begins to bulge, forming bulges. As corrosion continues, the gas inside the bulge continues to accumulate, and the bulge gradually enlarges. The metal layer on the surface of the bulge may become thinner due to corrosion, and even cracks may appear. When the pressure inside the bulge reaches a certain level, the metal surface can no longer withstand this pressure, and the bulge ruptures. After rupture, obvious cracks or holes are formed on the metal surface, exposing the internal corrosion products. The corrosion products are accelerated by the scouring of the fluid and are eventually carried away by the fluid and peeled off from the metal surface.

[0079] Specifically, the process of determining the potential stripping area includes:

[0080] The line closest to the light source at the feature extraction location is defined as the shadow boundary. The center point of the shadow boundary is determined, and a plane perpendicular to the center line of the oil pipeline is determined based on the center point.

[0081] Using the plane as a reference, the feature extraction locations are mirrored to form a mirror region;

[0082] The region formed by connecting the feature extraction location with the outer edge of the mirror region is the potential stripping region.

[0083] Specifically, the potential stripping area is a preliminary area obtained by mirroring the shaded part. The mirrored area and the original potential stripping area often cannot form a closed shape. Therefore, it is necessary to connect the outer edges of the two parts after mirroring to form a closed shape.

[0084] Specifically, the process of determining the offset distance of the shadow boundary includes:

[0085] After increasing the fluid velocity, the shaded boundary is redefined and recorded as the secondary boundary.

[0086] The distance between the shadow boundary and the secondary boundary is defined as the offset distance.

[0087] It is understandable that the bulge contains gas, which will shift under different flow rates. The highest point of the bulge will shift away from the light source, thus causing the shadow boundary to also shift away from the light source.

[0088] Specifically, the process of determining the properties of the potential stripping region based on the offset distance includes:

[0089] Compare the offset distance with the preset offset distance;

[0090] Based on the comparison result that the offset distance is greater than the preset distance, the potential peeling area is determined to be a bulge;

[0091] Based on the comparison result that the offset distance is equal to the preset distance, the potential peeling area is determined to be a non-bulge area.

[0092] Specifically, the preset distance is the offset distance of light propagating in the same fluid at different flow rates. It can be understood that as the flow rate of the fluid increases, the propagation of light under the same conditions will be offset. The specific offset distance is determined by experiments.

[0093] Specifically, the process of determining whether to adjust the angle of illumination includes:

[0094] Obtain a front view of the bulge, and determine the planar range of the bulge after preprocessing the front view;

[0095] The area ratio between the plane range and the feature extraction location is determined as the overlap degree;

[0096] The degree of overlap is compared with a first preset degree of overlap and a second preset degree of overlap, respectively;

[0097] The illumination angle is reduced based on the comparison result that the overlap degree is less than the first preset overlap degree;

[0098] The illumination angle is increased based on the comparison result that the overlap degree is greater than the second preset overlap degree;

[0099] Based on the comparison result that the overlap degree is greater than or equal to the first preset overlap degree and less than or equal to the second preset overlap degree, it is determined that the illumination angle will not be adjusted.

[0100] Specifically, the first preset overlap degree is set to a value range of [95%, 100%], and 98% is preferred in this embodiment of the invention; the second preset overlap degree is set to a value range of [100.1%, 103%], and 102% is preferred in this embodiment of the invention.

[0101] It is understandable that an overlap of less than the first preset overlap indicates that the potential peeling area is smaller than the planar area of ​​the bulge, which means that the illumination angle is too large and should be appropriately reduced to increase the area of ​​the potential peeling area. An overlap of greater than the second preset overlap indicates that the potential peeling area is larger than the planar area of ​​the bulge, which means that the illumination angle is too small and should be appropriately increased to reduce the area of ​​the potential peeling area.

[0102] Specifically, the process of reducing the illumination angle includes:

[0103] The first difference value is obtained by subtracting the first preset overlap degree from the overlap degree.

[0104] A first angle adjustment coefficient is determined based on the first difference, and the illumination angle is reduced using the first angle adjustment coefficient.

[0105] The illumination angle is the angle between the light ray and the inner wall of the oil pipeline.

[0106] Specifically, the formula for calculating the first angle adjustment coefficient is as follows:

[0107]

[0108] Where X1 represents the first angle adjustment coefficient and ΔC1 represents the first difference.

[0109] Specifically, the process of increasing the adjustment of the illumination angle includes:

[0110] The difference between the degree of overlap and the second preset degree of overlap is used to obtain a second difference value;

[0111] A second angle adjustment coefficient is determined based on the second difference, and the illumination angle is increased by the second angle adjustment coefficient.

[0112] Specifically, the formula for calculating the second angle adjustment coefficient is as follows:

[0113]

[0114] Where X2 represents the second angle adjustment coefficient, and ΔC2 represents the second difference.

[0115] Specifically, the process of determining the expected peeling time based on the bulge height and edge gradient includes:

[0116] Based on the adjusted illumination angle, the maximum spacing of the feature extraction positions is determined according to the direction parallel to the fluid flow direction. The maximum spacing and the illumination angle are then input into the calculation model to output the bulge height.

[0117] The edge gradient is determined based on the planar range of the bulge;

[0118] The estimated peeling time of the bulge is determined by combining the bulge height and the edge gradient.

[0119] Specifically, the maximum spacing is actually half the distance of the bulge plane range along a direction parallel to the fluid flow direction.

[0120] Specifically, the computational model is trained based on a large amount of accurate data from the bulge recognition process. The computational model may be a neural network model or a support vector machine, and there is no specific limitation.

[0121] It is understandable that the edge gradient represents the clarity of the edge, which is determined based on the gray-scale difference of the bump edge. The higher the height of the bump, the greater the edge clarity of the bump, that is, the greater the edge gradient, and the shorter the time before the rupture and peeling.

[0122] Specifically, the formula for determining the expected peeling time of the bulge based on the bulge height and the edge gradient is as follows:

[0123]

[0124] Where S represents the expected peeling time, K represents the material constant, E represents the edge gradient, H represents the bulge height, ΔC represents the difference in thermal expansion coefficients between the inner surface of the oil pipeline and the pipeline substrate, T represents the fluid temperature in the oil pipeline, and ε represents the critical strain at the interface between the inner surface of the oil pipeline and the pipeline substrate.

[0125] It is understandable that the inner surface of oil pipelines often has a coating, and the difference in expansion coefficients refers to the difference in expansion coefficients between the coating and the substrate.

[0126] Please see Figure 4 The diagram shown is a structural block diagram of an intelligent identification system for oil pipeline defects according to an embodiment of the present invention.

[0127] Specifically, embodiments of the present invention also provide an intelligent identification system for oil pipeline defects using an intelligent identification method for oil pipeline defects, comprising:

[0128] Pipeline inspection robot, used to acquire several images of the inner wall of oil pipelines;

[0129] The control center, which is connected to the pipeline inspection robot, includes,

[0130] A region determination module is used to determine several feature extraction locations based on the image, and to determine potential stripping regions by mirroring the feature extraction locations.

[0131] The defect determination module is used to determine whether the potential peeling area is a bulge based on the offset distance of the shadow boundary, and to determine the planar range of the front view of the bulge to determine the adjustment of the illumination angle based on the degree of overlap.

[0132] The defect development prediction module is used to determine the bulge height based on the adjusted illumination angle to determine the expected peeling time of the bulge.

[0133] Specifically, the pipeline inspection robot moves passively under normal inspection conditions based on the pressure difference of the fluid. After determining that the fluid flow rate should be increased, its own power system is activated to increase the speed of the pipeline inspection robot to increase the fluid flow rate and increase the pressure of the fluid on the bulge, causing the bulge to change.

[0134] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for intelligent identification of defects in oil pipelines, characterized in that, include: Acquire several images of the oil pipeline, preprocess the images to determine several feature extraction locations, and determine potential stripping areas based on the feature extraction locations; Increase the fluid flow rate in the potential peeling area to determine whether the potential peeling area is a bulge based on the offset distance of the shadow boundary; Based on the result that the potential peeling area is determined to be a bulge, a front view of the potential peeling area is obtained. After preprocessing the front view, the planar range of the bulge is determined, and the degree of overlap between the planar range and the potential peeling area is determined. The illumination angle is adjusted based on the overlap, the bulge height is determined based on the feature extraction location, and the expected peeling time of the bulge is determined based on the bulge height and the edge gradient of the bulge.

2. The intelligent identification method for oil pipeline defects according to claim 1, characterized in that, The process of determining the feature extraction location includes: Determine a number of pixels with a gray value less than the first gray value, and connect adjacent positions of each pixel to form a number of pixel regions; The pixel region is matched with a set of standard regions to determine each similarity, and the maximum similarity is determined based on each similarity. The maximum similarity is compared with the preset similarity. The pixel region is determined as the feature extraction location based on the comparison result where the maximum similarity is greater than or equal to the preset similarity.

3. The intelligent identification method for oil pipeline defects according to claim 2, characterized in that, The process for determining the potential stripping region includes: The line closest to the light source at the feature extraction location is defined as the shadow boundary. The center point of the shadow boundary is determined, and a plane perpendicular to the center line of the oil pipeline is determined based on the center point. Using the plane as a reference, the feature extraction locations are mirrored to form a mirror region; The region formed by connecting the feature extraction location with the outer edge of the mirror region is the potential stripping region.

4. The intelligent identification method for oil pipeline defects according to claim 3, characterized in that, The process of determining the offset distance of the shadow boundary includes: After increasing the fluid velocity, the shaded boundary is redefined and recorded as the secondary boundary. The distance between the shadow boundary and the secondary boundary is defined as the offset distance.

5. The intelligent identification method for oil pipeline defects according to claim 4, characterized in that, The process of determining the properties of the potential stripping region based on the offset distance includes: Compare the offset distance with the preset offset distance; Based on the comparison result that the offset distance is greater than the preset distance, the potential peeling area is determined to be a bulge.

6. The intelligent identification method for oil pipeline defects according to claim 5, characterized in that, The process of determining whether to adjust the angle of the light includes: Obtain a front view of the bulge, and determine the planar range of the bulge after preprocessing the front view; The area ratio between the plane range and the feature extraction location is determined as the overlap degree; The degree of overlap is compared with a first preset degree of overlap and a second preset degree of overlap, respectively; Based on the comparison result that the overlap degree is less than the first preset overlap degree, it is determined to reduce the illumination angle; Based on the comparison result that the overlap degree is greater than the second preset overlap degree, it is determined to increase the illumination angle.

7. The intelligent identification method for oil pipeline defects according to claim 6, characterized in that, The process of reducing the illumination angle includes: The first difference value is obtained by subtracting the first preset overlap degree from the overlap degree. A first angle adjustment coefficient is determined based on the first difference, and the illumination angle is reduced using the first angle adjustment coefficient. The illumination angle is the angle between the light ray and the inner wall of the oil pipeline.

8. The intelligent identification method for oil pipeline defects according to claim 6, characterized in that, The process of increasing the illumination angle includes: The difference between the degree of overlap and the second preset degree of overlap is used to obtain a second difference value; A second angle adjustment coefficient is determined based on the second difference, and the illumination angle is increased by the second angle adjustment coefficient.

9. The intelligent identification method for oil pipeline defects according to claim 8, characterized in that, The process of determining the expected peeling time based on the bulge height and edge gradient includes: Based on the adjusted illumination angle, the maximum spacing of the feature extraction positions is determined according to the direction parallel to the fluid flow direction. The maximum spacing and the illumination angle are then input into the calculation model to output the bulge height. The edge gradient is determined based on the planar range of the bulge; The estimated peeling time of the bulge is determined by combining the bulge height and the edge gradient.

10. An intelligent identification system for oil pipeline defects using the intelligent identification method for oil pipeline defects according to any one of claims 1-9, characterized in that, include: Pipeline inspection robot, used to acquire several images of the inner wall of oil pipelines; The control center, which is connected to the pipeline inspection robot, includes, A region determination module is used to determine several feature extraction locations based on the image, and to determine potential stripping regions by mirroring the feature extraction locations. The defect determination module is used to determine whether the potential peeling area is a bulge based on the offset distance of the shadow boundary, and to determine the planar range of the front view of the bulge to determine the adjustment of the illumination angle based on the degree of overlap. The defect development prediction module is used to determine the bulge height based on the adjusted illumination angle to determine the expected peeling time of the bulge.

Images