Oil pipeline weld defect detection method based on image processing

Abstract

The present invention relates to the field of oil pipeline detection, and more specifically, to an oil pipeline weld defect detection method based on image processing. The method comprises: obtaining a TOFD image of the oil pipeline to be detected, constructing a coordinate system, calculating the probability that each row of pixels belongs to the original signal, eliminating all pixels in the row whose probability is greater than a preset probability threshold to obtain the remaining pixels; calculating the abnormality of the target point, and taking the pixels whose abnormality is greater than the preset abnormality threshold as abnormal points; calculating the same texture probability of any two abnormal points, connecting the abnormal points by using regional growth according to the same texture probability, and obtaining abnormal lines; calculating the total defect degree of the weld, and completing the oil pipeline weld defect detection. Through the technical solution of the present invention, the accuracy of the weld defect detection result can be improved, and the safety of the use of the oil pipeline is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of oil pipeline detection, and more specifically, to an oil pipeline weld defect detection method based on image processing. Background Art

[0002] The inspection of oil pipeline welding quality is crucial to ensure the safety and reliability of the entire energy transportation system. Welding defects such as cracks, pores, and lack of fusion may not only lead to oil leaks, but may even cause serious safety accidents such as explosions. Therefore, strict quality inspection of oil pipeline welds is the key to ensuring the safe operation of pipelines. Welding quality inspection usually includes inspection of the external and internal quality of the weld.

[0003] TOFD (Time-of-Flight Diffraction) is an ultrasonic non-destructive testing technology, which is widely used in weld inspection of oil pipelines. TOFD is based on the diffraction phenomenon produced when ultrasonic waves propagate in the material of the pipeline and encounter defects to detect defects in welds, such as cracks, incomplete penetration, slag inclusions, etc., and uses the diffraction waves at the upper and lower endpoints of the defect to locate the defect.

[0004] The existing Chinese patent application document with publication number CN116263430A discloses a method for determining the degree of harm of pipeline girth weld defects. The method performs radiographic detection and TOFD detection on pipeline girth weld defects respectively; for pipeline girth weld defects detected by radiographic detection and TOFD, pipeline girth weld defects are determined and classified, including surface opening defects and buried defects; surface opening defects are determined as harmful defects; buried defects are determined as acceptable defects and unqualified through TOFD detection and ultrasonic verification of the defective parts exceeding the standard; unqualified defects are divided into point defects and linear defects, and then acceptable defects and harmful defects are determined.

[0005] However, the grayscale value of the defect diffraction signal in TOFD detection is relatively close to the grayscale value of the original signal, which will increase the difficulty of defect identification and lead to inaccurate detection results when detecting pipeline weld defects. Summary of the invention

[0006] In order to solve the problem of inaccurate weld defect detection results, the present invention proposes a petroleum pipeline weld defect detection method based on image processing.

[0007] The invention discloses a method for detecting defects in oil pipeline welds based on image processing, comprising: obtaining a TOFD image of an oil pipeline to be tested, constructing a coordinate system, calculating the probability that each row of pixels belongs to an original signal, the original signal including upper and lower surface echoes, through waves and side wall echoes, eliminating all pixels in the row with a probability greater than a preset probability threshold to obtain remaining pixels, and taking any remaining pixel as a target point; calculating the abnormality degree of the target point, taking the pixel with an abnormality degree greater than a preset abnormality threshold as an abnormal point; calculating the same-texture probability of any two abnormal points, connecting the abnormal points by using regional growth according to the same-texture probability, obtaining an abnormal line, and the same-texture probability satisfies the relationship:

[0008] , Indicates abnormal points and outliers The probability of the same texture is represents the adjustment factor, and Represents abnormal points The degree and point of abnormality The degree of abnormality, and Represents abnormal points The horizontal coordinates and outliers The horizontal axis of Indicates abnormal points and outliers The Euclidean distance of Represents an exponential function; calculates the total defect degree of the weld and completes the oil pipeline weld defect detection.

[0009] By constructing a coordinate system and calculating the probability that a pixel belongs to the original signal, non-defect related signals are effectively identified and eliminated, thereby improving the pertinence of detection. Potential abnormal points are further screened out by calculating the abnormality of the target point and comparing it with the preset abnormality threshold. These abnormal points are connected using the same texture probability and region growing algorithm to form an abnormal line, which can not only identify a single defect point, but also reveal the continuity and extension trend of the defect.

[0010] Preferably, the calculating the probability that each row of pixels belongs to the original signal includes:

[0011] Calculate the probability of the echo belonging to the upper and lower surfaces and the through wave: , Indicates The probability that the row pixel belongs to the upper and lower surface echoes and the through wave, Indicates The mean gray value of the row pixels, Indicates The entropy value of the row pixel, Represents an exponential function.

[0012] Calculate the probability of belonging to the sidewall echo: , Indicates The probability that a row pixel belongs to the sidewall echo, Indicates The mean gray value of the row pixels, Indicates The entropy value of the row pixel, Represents an exponential function.

[0013] The grayscale mean and entropy of the image are used as features. The grayscale mean reflects the brightness distribution of the image, while the entropy represents the complexity and information content of the image. By combining these features with the exponential function, the possibility of each pixel belonging to a different signal type can be quantified, and the pixels related to the original signal can be identified.

[0014] Preferably, calculating the abnormality degree includes: obtaining the grayscale value mean of the pixel points in the neighborhood of the target point, and normalizing the difference between the grayscale value of the target point and the grayscale value mean as the abnormality degree.

[0015] Preferably, calculating the abnormality degree further includes: obtaining the mean gradient value of the pixel points in the neighborhood of the target point, and normalizing the difference between the gradient value of the target point and the mean gradient value as the abnormality degree.

[0016] Preferably, the total defect degree satisfies the relationship:

[0017] , Indicates the total defect level, Indicates abnormal line The total number of outliers, Represents the total number of remaining pixels. Indicates abnormal line The mean value of the abnormal degree of the abnormal points in Indicates the total number of abnormal lines, Represents the normalization function.

[0018] It is able to comprehensively evaluate the abnormal conditions of welds in oil pipelines, focusing not only on the number of abnormal points but also considering the distribution and severity of these abnormal points.

[0019] Preferably, the total defect degree also includes: obtaining the area between the DC wave area and the echo area, calculating the fluctuation degree according to the signal amplitude in the study area between the DC wave area and the echo area; calculating the quantitative ratio of the total number of abnormal points to the total number of remaining pixel points, adding the fluctuation degree and the quantitative ratio and normalizing the result as the total defect degree.

[0020] Not only the number and distribution of abnormal points are considered, but also the fluctuation degree is introduced to reflect the fluctuation of signal amplitude, so as to capture the abnormal patterns in TOFD images more comprehensively.

[0021] Preferably, the degree of fluctuation includes: calculating the variance of the signal amplitude in the study area; calculating the slope of the line connecting the amplitudes corresponding to two adjacent time points in the study area, and then obtaining the mean slope of the study area; adding the variance and the mean slope and normalizing the result as the degree of fluctuation.

[0022] Preferably, the method of completing the oil pipeline weld defect detection includes: in response to the total defect degree being greater than the maintenance threshold, sending an alarm signal and repairing the oil pipeline; in response to the total defect degree being not greater than the maintenance threshold, calculating the ratio of the number of abnormal lines greater than a preset length to the total number of abnormal lines, when the ratio is greater than one half, repairing or replacing the oil pipeline, and when the ratio is not greater than one half, the oil pipeline continues to be used.

[0023] Through the maintenance threshold, oil pipelines with many abnormal points can be screened out first for maintenance or replacement; however, a low total defect level only means that the number of abnormal points is small, and it cannot guarantee that the oil pipeline can continue to be used. By constraining the generated abnormal lines, the length of the abnormal lines that may affect the oil pipeline can be determined. If the number of abnormal points is small but they can be connected into a long abnormal line, it means that the local abnormality of the oil pipeline is large. When the connected abnormal lines exceed a certain length, it can also reflect that the oil pipeline needs to be repaired or replaced.

[0024] Beneficial effects of the present invention:

[0025] 1. The present invention constructs a coordinate system and calculates the probability that each row of pixels belongs to the original signal, effectively eliminating non-defective signals and retaining pixels that may have defects. It simulates the propagation and reflection process of signals in actual welds, and distinguishes normal signals from abnormal signals through a probability model, laying the foundation for subsequent defect detection.

[0026] 2. The present invention further identifies the real abnormal points by calculating the abnormal degree of the target point and combining the grayscale value and gradient value of the pixel points in the neighborhood. By analyzing the local features of the weld image, the subtle changes that may represent defects are captured, thereby improving the accuracy of defect detection.

[0027] 3. The present invention introduces the concept of isotexture probability, and connects the abnormal points through the method of regional growing to form abnormal lines. Taking into account the spatial distribution and texture characteristics of the abnormal points, the probability model is used to evaluate the correlation between the abnormal points, which helps to identify continuous defective areas and also helps to determine whether local abnormalities can affect the safety of oil pipeline use.

[0028] 4. The present invention calculates the total defect degree and combines it with the maintenance threshold, which helps to take maintenance measures in time and avoid safety accidents caused by defects. It can also improve the automation level of weld defect detection, reduce the subjectivity and labor intensity of manual detection, and help to extend the service life of the pipeline and ensure the safety and efficiency of oil transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] By reading the following detailed description 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 represent the same or corresponding parts, wherein:

[0030] Figure 1 It is a flow chart of a method for detecting oil pipeline weld defects based on image processing according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] 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.

[0032] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0033] The present invention provides a method for detecting oil pipeline weld defects based on image processing. Figure 1 As shown, the oil pipeline weld defect detection method based on image processing includes steps S1 to S4, which are described in detail below.

[0034] S1, obtain the TOFD image of the oil pipeline to be tested, construct a coordinate system, calculate the probability that each row of pixels belongs to the original signal, and eliminate all pixels in the row with a probability greater than a preset probability threshold to obtain the remaining pixels.

[0035] It should be noted that TOFD (Time of Flight Diffraction) is a highly efficient ultrasonic testing technology that detects and evaluates defects by analyzing the propagation time and path differences of ultrasonic waves in materials. In the detection of defects in oil pipeline welds, TOFD technology can provide high-resolution images to help inspectors identify and quantify various defects in welds, such as cracks, pores, and lack of fusion.

[0036] Since the oil pipeline is long and the weld area is large, the ultrasonic TOFD detector needs to work for a long time. The internal electronic components may generate noise due to heat accumulation, which affects the quality of the ultrasonic signal, resulting in a decrease in the quality of the collected ultrasonic image and TOFD image, which has an adverse effect on the subsequent determination of the degree of defects. Therefore, it is necessary to preprocess the ultrasonic image and TOFD image.

[0037] The TOFD detection display includes A-scan signals and TOFD images. The A-scan signals use radio frequency form, while the TOFD images display one-dimensional image lines for each A-scan signal. The position corresponds to the sound path, and the signal amplitude is represented by grayscale. The image lines formed by the continuous A-scan signals collected during the scanning process are spliced ​​into a two-dimensional view along the movement direction of the probe. One axis represents the probe movement distance, and the other axis represents the depth from the scanning surface to the bottom surface, thus forming a TOFD image.

[0038] In one embodiment, ultrasonic images and TOFD images obtained in an ultrasonic TOFD detector are collected. An adaptive filtering algorithm is used to remove noise in ultrasonic images and TOFD images. An automatic time gain compensation (TGC) algorithm is used to compensate for the intensity reduction of ultrasonic signals due to attenuation during propagation, so that structures at different depths in the TOFD image have similar brightness, thereby improving the overall quality of the TOFD image. Among them, the adaptive filtering algorithm and the automatic time gain compensation algorithm are both existing technologies and will not be described in detail here.

[0039] The coordinate system is constructed with the pixel point in the upper left corner of the TOFD image as the coordinate origin, the horizontal right direction as the horizontal axis, and the vertical downward direction as the vertical axis. The row of the coordinate system represents the distance the probe moves along the weld, that is, the scanned position information.

[0040] Therefore, the pixels are divided into rows and the probability of each row of pixels belonging to the original signal is calculated. The original signal includes upper and lower surface echoes, direct waves and side wall echoes.

[0041] It should be noted that the upper and lower surface echoes represent the waves reflected by the ultrasonic wave from the upper and lower surfaces of the material, respectively. In the TOFD image, these echoes usually appear as two bright lines, corresponding to the upper and lower surfaces of the material, respectively; the through wave is the signal that the ultrasonic wave generated by the transmitting probe directly passes through the inspected workpiece without any reflection or diffraction and is directly captured by the receiving probe. The existence of the through wave indicates that the ultrasonic wave can penetrate the material and is an important reference for evaluating the continuity of the material; the side wall echo may appear as a different signal from the through wave and the surface echo in the TOFD image, which can be used to identify possible defects or irregularities on the side wall.

[0042] Calculate the probability of the echo belonging to the upper and lower surfaces and the through wave:

[0043] , Indicates The probability that the row pixel belongs to the upper and lower surface echoes and the through wave, Indicates The mean gray value of the row pixels, Indicates The entropy value of the row pixel, Represents an exponential function.

[0044] Calculate the probability of belonging to the sidewall echo:

[0045] , Indicates The probability that a row pixel belongs to the sidewall echo, Indicates The mean gray value of the row pixels, Indicates The entropy value of the row pixel, Represents an exponential function.

[0046] It should be noted that in TOFD images, grayscale values ​​are usually used to represent the amplitude of the signal. Direct waves, upper surface waves, and lower surface waves usually appear as very bright areas in the image because these signals are relatively strong. Direct waves are ultrasonic waves that propagate along the surface of the workpiece and arrive at the receiving end first, while upper surface waves and lower surface waves are signals reflected from the upper and lower surfaces of the workpiece. Side wall echoes usually appear as darker areas in the image because these signals are relatively weak and may appear on both sides of the image. The above types of waves are relatively fixed in position in the TOFD image.

[0047] All pixels in the row whose probability is greater than a preset probability threshold are eliminated to obtain the remaining pixels. Exemplarily, the preset probability threshold is set to 0.5, and the specific value can be set by those skilled in the art.

[0048] When calculating the probability of belonging to the original signal for any row of pixels, four situations may occur, namely: the first is that the probability of belonging to the upper and lower surface echoes and the through wave is greater than the preset probability threshold, and the probability of belonging to the side wall echo is greater than the preset probability threshold. At this time, the maximum value of the two probabilities is used as the signal to which the pixels in this row belong, and all the pixels in this row need to be eliminated; the second is that the probability of belonging to the upper and lower surface echoes and the through wave is not greater than the preset probability threshold, and the probability of belonging to the side wall echo is not greater than the preset probability threshold. At this time, all the pixels in this row are retained as the remaining pixels; the third is that the probability of belonging to the upper and lower surface echoes and the through wave is not greater than the preset probability threshold, and the probability of belonging to the side wall echo is greater than the preset probability threshold. At this time, the pixels in this row belong to the side wall echo, and all the pixels in this row need to be eliminated; the fourth is that the probability of belonging to the upper and lower surface echoes and the through wave is greater than the preset probability threshold, and the probability of belonging to the side wall echo is not greater than the preset probability threshold. At this time, the pixels in this row belong to the upper and lower surface echoes and the through wave, and all the pixels in this row need to be eliminated.

[0049] At this point, the remaining pixel points can be obtained, and these remaining pixel points are defect diffraction signals, oil pipeline feature points or noise points.

[0050] Each defect texture generated by the defect diffraction signal in the TOFD image is straight, that is, if the pixels are on the same defect texture, they will be in a similar horizontal position and the degree of abnormality will be similar.

[0051] S2, calculate the abnormality degree of the target point, and take the pixel points whose abnormality degree is greater than the preset abnormality threshold as abnormal points.

[0052] In one embodiment, calculating the degree of abnormality includes: obtaining the mean grayscale value of pixels in the neighborhood of the target point, and normalizing the difference between the grayscale value of the target point and the mean grayscale value as the degree of abnormality.

[0053] Exemplarily, the preset abnormal threshold is set to 0.6, and the specific value can be set by those skilled in the art.

[0054] At this point, the abnormal points can be screened out from the remaining pixels.

[0055] In one embodiment, calculating the degree of abnormality further includes: obtaining the mean gradient value of the pixels in the neighborhood of the target point, and normalizing the difference between the gradient value of the target point and the mean gradient value as the degree of abnormality.

[0056] S3, calculate the same-texture probability of any two abnormal points, connect the abnormal points by using region growing according to the same-texture probability, and obtain the abnormal line.

[0057] In one embodiment, the same texture probability satisfies the relationship:

[0058] , Indicates abnormal points and outliers The probability of the same texture is represents the adjustment factor, and Represents abnormal points The degree and point of abnormality The degree of abnormality, and Represents abnormal points The horizontal coordinates and outliers The horizontal axis of Indicates abnormal points and outliers The Euclidean distance of Represents an exponential function.

[0059] The adjustment factor is used for normalization. For example, and The value range is , The value range is In this case, the adjustment factor is 2.5, which can normalize the absolute value of the difference to the range of 0 to 1.

[0060] It represents the cosine value of the angle between the line connecting the two outliers and the horizontal axis. The angle should be less than or equal to 90 degrees. The closer the cosine value is to 1, the closer the horizontal lines of the two outliers are. When the cosine value is close to 1, the smaller the difference in the degree of anomaly is, the greater the probability of the two outliers being of the same texture.

[0061] According to the above method, the isotexture probability between any two abnormal points is obtained, and the abnormal points are connected by region growing to obtain at least one abnormal line.

[0062] S4, calculate the total defect degree of the weld and complete the oil pipeline weld defect detection.

[0063] In one embodiment, the total defect level satisfies the relationship:

[0064] , Indicates the total defect level, Indicates abnormal line The total number of outliers, Represents the total number of remaining pixels. Indicates abnormal line The mean value of the abnormal degree of the abnormal points in Indicates the total number of abnormal lines, Represents the normalization function.

[0065] In response to the total defect degree being greater than a maintenance threshold, an alarm signal is sent and the oil pipeline is repaired.

[0066] In response to the total defect degree being no greater than the maintenance threshold, the ratio of the number of abnormal lines greater than a preset length to the total number of abnormal lines is calculated. When the ratio is greater than one-half, the oil pipeline is repaired or replaced. When the ratio is not greater than one-half, the oil pipeline continues to be used.

[0067] One-half is an exemplary value, and a specific value may be given by a person skilled in the art according to the length of the oil pipeline and other properties.

[0068] Preferably, the inspection threshold is set to 0.89.

[0069] In one embodiment, the total defect degree also includes: obtaining the area between the DC wave area and the echo area, calculating the fluctuation degree according to the signal amplitude in the study area between the DC wave area and the echo area; calculating the ratio of the total number of abnormal points to the total number of remaining pixel points, adding the fluctuation degree and the ratio and normalizing the result as the total defect degree.

[0070] The sampling time of the ultrasonic image is from 0 to From the first second, calculate the difference between the amplitude per second and the initial amplitude. When Seconds is the starting time of the DC wave, and it traverses from left to right. When it satisfies and When Seconds is the termination moment of the DC wave. Indicates Amplitude per second, Indicates Amplitude per second, represents the initial amplitude, Indicates Amplitude per second, Indicates Amplitude per second, Represents the amplitude difference threshold, which is preferably set to 0.6.

[0071] From Seconds start, when satisfied When seconds as the echo start time, traverse from right to left, when it meets and When Seconds is taken as the echo termination time. Indicates Amplitude per second, Indicates Amplitude per second, represents the initial amplitude, Indicates Amplitude per second, Indicates Amplitude per second, Indicates the amplitude difference threshold.

[0072] The positions of the DC wave and the echo are relatively fixed.

[0073] The degree of fluctuation includes: calculating the variance of the signal amplitude in the study area; calculating the slope of the line connecting the amplitudes corresponding to two adjacent time points in the study area, and then obtaining the mean slope of the study area; adding the variance and the mean slope and normalizing the result as the degree of fluctuation.

[0074] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0075] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting defects in oil pipeline welds based on image processing, characterized in that: include: Get the TOFD image of the oil pipeline to be tested, build a coordinate system, and calculate the probability that each row of pixels belongs to the original signal. The original signal includes upper and lower surface echoes, direct waves, and side wall echoes. Calculate the probability of belonging to upper and lower surface echoes and direct waves: , Indicates The probability that the row pixel belongs to the upper and lower surface echoes and the through wave, Indicates The mean gray value of the row pixels, Indicates The entropy value of the row pixel, represents the exponential function; Calculate the probability of belonging to the sidewall echo: , Indicates The probability that a row pixel belongs to the sidewall echo; Eliminate all pixels in the row whose probability is greater than a preset probability threshold to obtain the remaining pixels, and use any remaining pixel as the target point; Calculate the abnormality degree of the target point, and take the pixel points whose abnormality degree is greater than the preset abnormality threshold as abnormal points; Calculate the probability of the same texture of any two outliers, connect the outliers using region growing based on the probability of the same texture, and obtain the outlier line. The probability of the same texture satisfies the relationship: , Indicates abnormal points and outliers The probability of the same texture is represents the adjustment factor, and Represents abnormal points The degree and point of abnormality The degree of abnormality, and Represents abnormal points The horizontal coordinates and outliers The horizontal axis of Indicates abnormal points and outliers The Euclidean distance of Calculate the total defect degree of the weld to complete the oil pipeline weld defect detection.

2. The oil pipeline weld defect detection method based on image processing according to claim 1 is characterized in that: Calculating the abnormality degree includes: The mean grayscale value of the pixels in the neighborhood of the target point is obtained, and the difference between the grayscale value of the target point and the mean grayscale value is normalized as the degree of abnormality.

3. The oil pipeline weld defect detection method based on image processing according to claim 1 is characterized in that: Calculating the abnormality level further includes: Get the mean gradient value of the pixels in the neighborhood of the target point, and normalize the difference between the gradient value of the target point and the mean gradient value as the degree of abnormality.

4. The oil pipeline weld defect detection method based on image processing according to claim 1 is characterized in that: The total defect degree satisfies the relationship: , Indicates the total defect level, Indicates abnormal line The total number of outliers, Represents the total number of remaining pixels. Indicates abnormal line The mean value of the abnormal degree of the abnormal points in Indicates the total number of abnormal lines, Represents the normalization function.

5. The oil pipeline weld defect detection method based on image processing according to claim 1 is characterized in that: The total defect level also includes: The area between the DC wave area and the echo area is obtained, and the fluctuation degree is calculated according to the signal amplitude in the research area between the DC wave area and the echo area; The ratio of the total number of abnormal points to the total number of remaining pixels is calculated, and the fluctuation degree is added to the ratio and the result is normalized as the total defect degree.

6. The oil pipeline weld defect detection method based on image processing according to claim 5 is characterized in that: The degree of fluctuation includes: Calculate the variance of the signal amplitude in the study area; Calculate the slope of the line connecting the amplitudes of two adjacent time points in the study area, and then obtain the mean slope of the study area; The variance and the slope mean are added and normalized to obtain the volatility.

7. The method for detecting oil pipeline weld defects based on image processing according to claim 1 is characterized in that: The above-mentioned method of completing the oil pipeline weld defect detection includes: In response to the total defect degree being greater than a maintenance threshold, an alarm signal is sent to perform maintenance on the oil pipeline; In response to the total defect degree being no greater than the maintenance threshold, the ratio of the number of abnormal lines greater than a preset length to the total number of abnormal lines is calculated. When the ratio is greater than one-half, the oil pipeline is repaired or replaced. When the ratio is not greater than one-half, the oil pipeline continues to be used.

Images