Method for detecting the internal diameter variation of a cable protection duct

Abstract

The application provides a cable protection pipeline inner diameter change detection method, and relates to the technical field of cable protection pipelines, and comprises the following steps: placing a detection carrier loaded with an annular laser projection module and a coaxial imaging collection module into a cable protection pipeline to be detected, and starting the annular laser projection module to project an annular closed light spot to the inner wall of the pipeline; driving the detection carrier to travel along the axial direction of the pipeline, synchronously collecting a pipeline inner wall image sequence containing the annular light spot and corresponding axial position data of the detection carrier, and binding the image data and the position data one by one according to timestamps. The application has the following advantages: the method does not rely on the passability discrimination of a gauge, can quantitatively describe the slow gradual change process of the inner diameter, and can output the position, length and gradual change rate of an abnormal section and other information, thereby making up for the problem of insufficient hidden gradual change deformation detection capability of the prior art to a certain extent, and helping to improve the fineness of cable protection pipeline state evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable protection pipe, in particular to a cable protection pipe inner diameter change detection method. BACKGROUND

[0002] As an important infrastructure for laying power and communication lines, the smoothness and geometric stability of the internal passage of the cable protection pipe are directly related to the safe operation of the cable and the feasibility of later maintenance and replacement. During long-term service, in addition to the sudden blockage caused by structural damage due to external force or foreign matter intrusion, the pipe may also experience slow compression deformation or thermal expansion and contraction due to external load, uneven foundation settlement, and environmental temperature cycle changes, etc., which may cause the inner diameter of the pipe to gradually decrease or expand along the axial direction. Such inner diameter changes are usually developed in a hidden and slow manner, and are not accompanied by obvious structural fractures or complete blockage, so they are difficult to be effectively identified by conventional detection methods.

[0003] In the prior art, the detection of cable protection pipes mainly relies on pipe endoscopes, sonar detection, mechanical gauges, or electromagnetic / optical defect recognition methods. These technologies have good effects in locating cracks, holes, joint misalignment, and other discrete structural damages, or in determining whether the pipe is completely blocked, but their detection principles are mostly based on threshold discrimination or qualitative observation, and lack the ability to continuously and accurately reconstruct the three-dimensional profile of the pipe inner wall. Especially for cases where the inner diameter changes non-uniformly and gradually along the length direction, existing methods can only give approximate values at local points or overall passability conclusions, and cannot provide complete inner diameter change curves and quantitative parameters, making it difficult for maintenance personnel to accurately assess the remaining available space of the pipe, predict potential risk sections, and provide reliable data support for preventive maintenance or replacement decisions.

[0004] Therefore, there is an urgent need to develop a detection method that can continuously and accurately quantify the gradual changes in the inner diameter of cable protection pipes to address the shortcomings of existing technologies in this regard. SUMMARY

[0005] The purpose of the present application is to provide a cable protection pipe inner diameter change detection method, which solves the problem that the existing cable protection pipe inner diameter detection method in the prior art cannot accurately detect the pipe inner diameter change, and is difficult to locate the deformation position and evaluate the deformation degree.

[0006] To achieve the above-mentioned purpose of the application, the technical solution adopted by the present application is as follows:

[0007] A cable protection pipe inner diameter change detection method, comprising the following steps:

[0008] Put the detection carrier equipped with the annular laser projection module and the coaxial imaging acquisition module into the cable protection pipeline to be detected, and start the annular laser projection module to project an annular closed light spot to the inner wall of the pipeline;

[0009] Drive the detection carrier to travel along the pipeline axis, synchronously acquire the image sequence of the inner wall of the pipeline containing the annular light spot and the corresponding axial position data of the detection carrier, and bind the image data and the position data one by one according to the time stamp;

[0010] Extract the light spot feature from each inner wall image to obtain the profile point set of the cross section of the pipeline at the corresponding position;

[0011] Calculate the equivalent inner diameter value of the corresponding cross section based on the profile point set at each position, combine the matched axial position data, and construct the inner diameter change sequence with the axial distance as the independent variable;

[0012] Perform gradual feature analysis processing on the inner diameter change sequence to generate a continuous change curve of the inner diameter of the pipeline along the axis, and output the inner diameter gradual quantization parameter.

[0013] As an improvement, the specific process of the light spot feature extraction is:

[0014] An adaptive threshold algorithm based on local neighborhood gray entropy is used to segment the inner wall image, and a binary annular light spot region is extracted;

[0015] Perform sub-pixel level edge detection on the annular light spot region to obtain a set of pixel coordinates of the light spot edge;

[0016] Call the pre-calibration parameters of the detection carrier to map and convert the set of edge pixel coordinates in the image coordinate system to the physical coordinate system of the pipeline cross section, to obtain the cross section inner wall profile point set.

[0017] As an improvement, the calculation process of the equivalent inner diameter value is:

[0018] Perform ellipse fitting on the cross section inner wall profile point set using the least squares method to obtain the long semi-axis and short semi-axis parameters of the fitted ellipse;

[0019] Take twice the geometric mean of the long semi-axis and the short semi-axis as the equivalent inner diameter of the current cross section, and calculate the ratio of the difference between the long semi-axis and the short semi-axis to the long semi-axis as an auxiliary quantization parameter of the deformation degree of the pipeline cross section.

[0020] As an improvement, during the acquisition of the inner wall image sequence, a posture correction step is also included:

[0021] Through the inertial measurement unit integrated in the detection carrier, the pitch angle and yaw angle posture data of the detection carrier are acquired in real time;

[0022] Based on the attitude data, a rotation transformation matrix of the detection carrier coordinate system relative to the ideal axis of the pipeline is constructed. An inverse rotation transformation is applied to each coordinate point in the contour point set to correct the measurement deviation introduced by the misalignment between the detection carrier axis and the pipeline axis.

[0023] As an improvement, the axial position data is obtained in any of the following ways:

[0024] A push cable with a length indicator is connected to the tail of the detection carrier. The length of the push cable extended in real time is measured by an encoder as the travel distance.

[0025] A contact-type odometer wheel is installed on the detection carrier, and the distance traveled is calculated by the number of rotations of the odometer wheel and the circumference of the wheel.

[0026] The obtained travel distance data is synchronized with the image acquisition trigger signal to achieve precise matching between each frame of the inner wall image and the corresponding axial position.

[0027] As an improvement, the specific process of the gradient feature parsing is as follows:

[0028] By using a sliding window to traverse the inner diameter change sequence, and performing polynomial fitting on the data points within the window, the inner diameter change trend function of each local segment is obtained.

[0029] The first derivative of the inner diameter variation trend function at each position is obtained to get the inner diameter gradient rate at the corresponding axial position.

[0030] When the absolute value of the gradient rate in a continuous segment exceeds the preset gradient rate threshold, and the length of the segment meets the preset minimum judgment length, the segment is determined to be an abnormal inner diameter gradient segment.

[0031] As an improvement, the axial length of the sliding window is directly proportional to the nominal inner diameter of the pipe, and the value of the proportionality coefficient ranges from 3 to 10.

[0032] The order of the polynomial fitting is adaptively selected based on the number of data points contained in the window. The highest order does not exceed one-fifth of the total number of data points in the window, and the lowest order is not lower than order 2.

[0033] As an improvement, the detection carrier's movement and sampling adopt any of the following working modes:

[0034] Equal-interval stepping mode: The detection carrier moves to the sampling position at a preset step length and then pauses. After completing the acquisition of a single frame of inner wall image, it continues to move.

[0035] Continuous travel mode: The detection carrier maintains a constant axial movement, and the coaxial imaging acquisition module continuously acquires images of the inner wall at a fixed frame rate.

[0036] As an improvement, a risk assessment output step is also included:

[0037] The inner diameter continuous variation curve is compared with the inner diameter parameter of the pipeline design, and the residual available space distribution data of the whole pipeline section is calculated.

[0038] According to the preset multi-level safety threshold, the risk level of each gradual abnormal section is divided, the starting position, section length, maximum gradient rate and risk level information of the abnormal section are marked in the pipeline axial development drawing, and finally a structured detection evaluation report is output.

[0039] A cable protection pipeline inner diameter variation detection system, characterized by comprising:

[0040] The detection carrier is provided with an annular laser projection module, a coaxial imaging acquisition module, an axial positioning module and an attitude measurement module, and is used for stretching into the pipeline to collect inner wall images, axial positions and attitude data;

[0041] The data processing module is used for receiving the image, position and attitude data transmitted by the detection carrier, performing spot feature extraction, equivalent inner diameter calculation, attitude correction and gradual change feature analysis processing, and generating an inner diameter variation curve and gradual change quantization parameters;

[0042] The output display module is used for visualizing the inner diameter variation curve, gradual abnormal section marking and risk assessment results, and outputting a structured detection report.

[0043] The beneficial effects of the present application are: the present method combines annular laser projection and coaxial imaging, can continuously obtain a large amount of cross-sectional profile data along the axial direction of the pipeline, compared with the discrete point sampling method, which is helpful to more comprehensively reflect the gradual change trend of the inner diameter of the pipeline along the length direction.

[0044] The introduction of attitude correction and ellipse fitting processing can weaken the influence of the detection carrier deflection on the measurement results to a certain extent, and can simultaneously output the cross-sectional deformation degree auxiliary parameters, providing multi-dimensional reference information for pipeline structural state evaluation.

[0045] Through sliding window fitting and gradient rate analysis, the slowly developing inner diameter gradual change section can be quantitatively identified and graded, which provides data support for operators to locate potential risk areas and develop maintenance plans. DETAILED DESCRIPTION

[0046] Figure 1 A flowchart of a cable protection pipeline inner diameter variation detection method of the present application.

[0047] Figure 2 An architectural block diagram of a cable protection pipeline inner diameter variation detection system of the present application.

[0048] Figure 3This is a schematic diagram of the light spot feature extraction and inner wall contour reconstruction of the present invention. Detailed Implementation

[0049] To make the content of this invention easier to understand, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Identical components are represented by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, while the terms "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0050] Example 1: Basic Detection Process Example

[0051] like Figure 1 As shown in the figure, this embodiment provides a method for detecting changes in the inner diameter of cable protection pipes. The specific implementation process is as follows:

[0052] The detection carrier, equipped with a ring laser projection module and a coaxial imaging acquisition module, is inserted into the pipe from the inlet end of the pipe to be tested for cable protection. The detection carrier can be a wheeled self-propelled mechanism or a cable traction mechanism, and its outer diameter is smaller than the nominal inner diameter of the pipe. The ring laser projection module is activated to project a closed ring-shaped light spot onto the inner wall of the pipe, forming a complete ring-shaped projection line on the inner wall of the pipe.

[0053] The driving detection carrier moves at a constant speed along the pipeline axis, with the speed controlled between 0.1 m / s and 0.5 m / s. During this movement, the coaxial imaging acquisition module continuously acquires a sequence of images of the pipeline inner wall, including annular light spots, at a fixed frame rate (e.g., 25 frames / second to 60 frames / second). Simultaneously, an axial positioning module, synchronized with the image acquisition trigger signal, obtains the axial position data of the detection carrier corresponding to each frame. After binding the image data and position data one-to-one according to timestamps, the data is transmitted to the data processing module.

[0054] The data processing module sequentially extracts light spot features from each frame of the inner wall image to obtain the inner wall contour point set of the pipe cross-section corresponding to each sampling position. Then, based on the contour point set of each position, it calculates the equivalent inner diameter value of the corresponding cross-section. Combined with the matched axial position data, it constructs an inner diameter change sequence with axial distance as the independent variable. Finally, it performs gradient feature parsing processing on the inner diameter change sequence to generate a continuous change curve of the pipe inner diameter along the axial direction and outputs the inner diameter gradient parameter.

[0055] Example 2: Light Spot Feature Extraction and Contour Reconstruction Example

[0056] This embodiment provides a detailed explanation of the spot feature extraction steps in Embodiment 1.

[0057] likeFigure 3 As shown, for each frame of the pipe inner wall image containing the ring-shaped light spot, an adaptive thresholding algorithm based on local neighborhood gray-level entropy is first used to segment the image. Specifically, a local neighborhood window of size 11×11 pixels is taken as the center of each pixel in the image, and the gray-level entropy value within the window is calculated. Based on the distribution characteristics of the gray-level entropy, the segmentation threshold of each pixel is adaptively determined, and the image is segmented into foreground (ring-shaped light spot region) and background, and the binarized ring-shaped light spot region is extracted.

[0058] For the binarized annular spot region, the Canny edge detection operator combined with a sub-pixel interpolation algorithm is used to perform edge detection, and the pixel coordinate set of the inner and outer edges of the annular spot is obtained, with a positioning accuracy of up to 0.1 pixels.

[0059] The pre-calibration parameters of the detection carrier are called—these calibration parameters include camera intrinsic parameters (focal length, principal point coordinates, distortion coefficients) and the relative pose relationship between the ring laser projection module and the camera—and the set of edge pixel coordinates in the image coordinate system is mapped and transformed to the physical coordinate system of the pipe cross-section through inverse perspective projection transformation, so as to obtain the contour point set of the inner wall of the pipe at the cross-section.

[0060] Example 3: Attitude Correction and Ellipse Fitting Example

[0061] This embodiment adds an attitude correction step based on Embodiment 1 or Embodiment 2, and provides a detailed explanation of the calculation process for the equivalent inner diameter.

[0062] During the acquisition of the inner wall image sequence, the pitch and yaw angle attitude data of the detection vehicle are acquired in real time through the inertial measurement unit (IMU) integrated into the detection vehicle, and the sampling frequency is consistent with the image acquisition frame rate.

[0063] Based on the attitude data, a rotation transformation matrix is ​​constructed for the coordinate system of the detection vehicle relative to the ideal axis of the pipeline. Specifically, the actual attitude of the detection vehicle is decomposed into a pitch rotation transformation about the X-axis and a yaw rotation transformation about the Z-axis, which are combined to obtain a three-dimensional rotation matrix R. Applying an inverse rotation transformation P'=R⁻¹·P to each coordinate point P in the contour point set can correct the measurement deviation introduced by the misalignment between the axis of the detection vehicle and the axis of the pipeline.

[0064] For the set of contour points of the inner wall of the cross section after attitude correction, the least squares method is used to fit an ellipse. Let the general equation of the ellipse be Ax² + Bxy + Cy² + Dx + Ey + F = 0. By minimizing the sum of squared algebraic distances from each contour point to the ellipse, the five parameters of the fitted ellipse are obtained, and then the major semi-axis a and minor semi-axis b of the ellipse are calculated.

[0065] The equivalent inner diameter of the current cross-section, D_eq=2√(a·b), is twice the geometric mean of the major and minor semi-axes (i.e., √(a·b)). Simultaneously, the ratio r=Δ / a of the difference between the major and minor semi-axes of the ellipse, Δ=ab, to the major semi-axe, a, is calculated as an auxiliary quantitative parameter for the degree of pipe cross-section deformation—a larger r value indicates more severe elliptical deformation of the pipe at that cross-section.

[0066] Example 4: Axial Positioning and Working Mode Example

[0067] This embodiment provides a detailed explanation of the method for acquiring the axial position data of the detection carrier and the working mode of the traveling sampling.

[0068] Axial position data can be obtained through either of the following two methods:

[0069] Method 1: A push cable with length markings is connected to the tail of the detection carrier. The cable has evenly spaced graduations or magnetic coded markings along its length. An encoder is installed at the pipe inlet. As the detection carrier moves, the push cable drives the encoder to rotate. The encoder measures the length of the push cable extended in real time, and this length is used as the distance traveled by the detection carrier.

[0070] Method 2: A contact-type odometer wheel is installed on the detection carrier, maintaining elastic contact with the inner wall of the pipe. As the detection carrier moves, the odometer wheel rotates accordingly, and the distance traveled is calculated by multiplying the number of rotations of the odometer wheel by the circumference of the wheel.

[0071] The obtained travel distance data is synchronized with the image acquisition trigger signal to ensure accurate matching between each frame of the inner wall image and the corresponding axial position.

[0072] The movement and sampling of the detection carrier can be carried out using any of the following working modes:

[0073] Equal-interval stepping mode: The detection carrier moves to the sampling position in a preset step size (such as 50mm or 100mm) and pauses. After the coaxial imaging acquisition module completes the acquisition of a single frame of the inner wall image, it continues to move to the next sampling position. This mode is suitable for high-precision sampling of key sections.

[0074] Continuous travel mode: The detection carrier maintains a constant axial speed (speed deviation controlled within ±5%), and the coaxial imaging acquisition module continuously acquires images of the inner wall at a fixed frame rate. This mode is suitable for rapid general inspection of long-distance pipelines.

[0075] Example 5: Gradient Feature Analysis and Risk Assessment Output Example

[0076] This embodiment provides a detailed explanation of the steps for parsing and processing gradient features and outputting risk assessments.

[0077] A sliding window is used to traverse the sequence of inner diameter changes. The axial length L_w of the sliding window is directly proportional to the nominal inner diameter D of the pipe, L_w = k·D, where the proportionality coefficient k ranges from 3 to 10. For example, for a pipe with a nominal inner diameter of 200mm, the axial length of the window can be selected between 600mm and 2000mm.

[0078] A polynomial fitting is performed on the data points within the window. The fitting order n is adaptively selected based on the number of data points m within the window: n = min(floor(m / 5), 4), and n ≥ 2, meaning the highest order does not exceed one-fifth of the total number of data points within the window, and the lowest order is not lower than order 2. The fitting yields the trend function f(x) of the inner diameter change in each local segment.

[0079] The first derivative f'(x) of the inner diameter change trend function f(x) at each location is obtained to get the inner diameter gradient rate at the corresponding axial position, in mm / m. When the absolute value of the gradient rate in a continuous section exceeds the preset gradient rate threshold (e.g., 5 mm / m), and the length of the section meets the preset minimum judgment length (e.g., not less than 5 times the nominal inner diameter of the pipe), the section is judged as an abnormal inner diameter gradient section.

[0080] By comparing the continuous inner diameter variation curve with the pipeline design inner diameter parameters, the remaining usable space distribution data for the entire pipeline section is calculated. Based on preset multi-level safety thresholds (e.g., Level 1 threshold: gradient rate > 5 mm / m; Level 2 threshold: gradient rate > 10 mm / m; Level 3 threshold: gradient rate > 20 mm / m), the risk level of each gradient anomaly section is classified. The starting position, section length, maximum gradient rate, and risk level information of the anomaly section are marked on the pipeline axial development diagram. Finally, a structured inspection and evaluation report is output, including a pipeline inner diameter variation curve for the entire section, a summary table of anomaly sections, and maintenance recommendations.

[0081] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting changes in the inner diameter of a cable protection conduit, characterized in that, Includes the following steps: The detection carrier equipped with a ring laser projection module and a coaxial imaging acquisition module is placed inside the protective pipe of the cable under test, and the ring laser projection module is activated to project a ring-shaped closed light spot onto the inner wall of the pipe. The driving detection carrier moves along the pipeline axis, simultaneously acquiring a sequence of images of the pipeline inner wall containing annular light spots and the corresponding axial position data of the detection carrier, and binding the image data and position data one-to-one according to timestamps; Light spot features are extracted from each frame of the inner wall image to obtain the set of contour points of the inner wall of the pipe cross-section at the corresponding location; The equivalent inner diameter of the corresponding cross section is calculated based on the contour point set at each location. Combined with the matched axial position data, an inner diameter variation sequence with axial distance as the independent variable is constructed. The inner diameter variation sequence is subjected to gradual feature analysis processing to generate a continuous variation curve of the pipe inner diameter along the axial direction, and the inner diameter gradual variation parameter is output.

2. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 1, characterized in that, The specific process for extracting the light spot features is as follows: An adaptive thresholding algorithm based on local neighborhood gray-level entropy is used to segment the inner wall image and extract the binarized annular light spot region. Subpixel-level edge detection is performed on the annular light spot region to obtain the set of pixel coordinates of the light spot edge; By calling the pre-calibration parameters of the detection carrier, the set of edge pixel coordinates in the image coordinate system is mapped and transformed to the physical coordinate system of the pipe cross-section, thus obtaining the set of contour points of the inner wall of the cross-section.

3. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 1, characterized in that, The calculation process for the equivalent inner diameter value is as follows: The least squares method is used to fit an ellipse to the set of points on the inner wall of the cross section, and the parameters of the major and minor axes of the fitted ellipse are obtained. The equivalent inner diameter of the current cross section is taken as twice the geometric mean of the major and minor semi-axes. At the same time, the ratio of the difference between the major and minor semi-axes of the ellipse to the major semi-axe is calculated as an auxiliary quantitative parameter for the degree of deformation of the pipe cross section.

4. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 3, characterized in that, The process of acquiring the inner wall image sequence also includes an attitude correction step: The pitch and yaw attitude data of the test vehicle are collected in real time by an inertial measurement unit integrated into the test vehicle. Based on the attitude data, a rotation transformation matrix of the detection carrier coordinate system relative to the ideal axis of the pipeline is constructed. An inverse rotation transformation is applied to each coordinate point in the contour point set to correct the measurement deviation introduced by the misalignment between the detection carrier axis and the pipeline axis.

5. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 1, characterized in that, The axial position data is obtained through any of the following methods: A push cable with a length indicator is connected to the tail of the detection carrier. The length of the push cable extended in real time is measured by an encoder as the travel distance. A contact-type odometer wheel is installed on the detection carrier, and the distance traveled is calculated by the number of rotations of the odometer wheel and the circumference of the wheel. The obtained travel distance data is synchronized with the image acquisition trigger signal to achieve precise matching between each frame of the inner wall image and the corresponding axial position.

6. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 1, characterized in that, The specific process of the gradient feature parsing is as follows: By using a sliding window to traverse the inner diameter change sequence, and performing polynomial fitting on the data points within the window, the inner diameter change trend function of each local segment is obtained. The first derivative of the inner diameter variation trend function at each position is obtained to get the inner diameter gradient rate at the corresponding axial position. When the absolute value of the gradient rate in a continuous segment exceeds the preset gradient rate threshold, and the length of the segment meets the preset minimum judgment length, the segment is determined to be an abnormal inner diameter gradient segment.

7. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 6, characterized in that, The axial length of the sliding window is directly proportional to the nominal inner diameter of the pipe, and the proportionality coefficient ranges from 3 to 10. The order of the polynomial fitting is adaptively selected based on the number of data points contained in the window. The highest order does not exceed one-fifth of the total number of data points in the window, and the lowest order is not lower than order 2.

8. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 7, characterized in that, The detection carrier travels and samples are collected using any of the following operating modes: Equal-interval stepping mode: The detection carrier moves to the sampling position at a preset step size and then pauses. After completing the acquisition of a single frame of inner wall image, it continues to move. Continuous movement mode: The detection carrier moves axially at a constant speed, and the coaxial imaging acquisition module continuously acquires images of the inner wall at a fixed frame rate.

9. The method for detecting changes in the inner diameter of a cable protection conduit according to claim 1, characterized in that, It also includes the risk assessment output step: By comparing the continuous inner diameter variation curve with the pipe design inner diameter parameters, the remaining available space distribution data for the entire pipe section is calculated. Based on preset multi-level safety thresholds, the risk levels of each gradual abnormal section are classified, and the starting position, section length, maximum gradual change rate and risk level information of the abnormal section are marked on the pipeline axial development diagram, and finally a structured inspection and evaluation report is output.

10. The cable protection duct inner diameter change detection system according to claim 1, characterized in that, include: The detection carrier is equipped with a ring laser projection module, a coaxial imaging acquisition module, an axial positioning module and an attitude measurement module, which are used to extend into the inside of the pipe to collect images of the inner wall, axial position and attitude data; The data processing module is used to receive image, position and attitude data transmitted by the detection carrier, perform spot feature extraction, equivalent inner diameter calculation, attitude correction and gradual feature analysis, and generate inner diameter change curve and gradual variable parameters. The output display module is used to visualize the inner diameter change curve, the annotation of gradual abnormal sections and the risk assessment results, and output a structured inspection report.

Images