Pipeline defect rating method, terminal equipment and storage medium

Abstract

This invention discloses a pipeline defect rating method, terminal device, and storage medium. This method enables quantitative rating of deposits, scum, obstructions, concealed branch pipe connections, foreign object penetration, and residual wall and dam roots. Furthermore, compared to existing rupture rating methods, this method considers the three-dimensional spatial properties of ruptures, significantly improving the accuracy of rupture ratings. This method enables comprehensive rating of pipeline defects, improving their accuracy.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a pipeline defect rating method, terminal equipment, and storage medium. Background Art

[0002] Urban drainage networks are crucial infrastructure and lifelines for urban operations. During operation, pipelines are susceptible to long-term erosion from sewage, rainwater, and external forces, leading to structural and functional damage. Without targeted repairs, these damages can lead to poor urban drainage, urban waterlogging, mixed rainwater and sewage flows, river pollution, and ground collapse. Repairing drainage pipelines requires identifying defects, assessing their grade, and refining materials. Determining the grade of seven defects—deposits, scum, ruptures, obstructions, concealed branch connections, foreign object penetration, and residual wall and dam roots—requires quantitative analysis based on defect morphology, spatial size, and spatial distribution. However, two-dimensional image data lacks three-dimensional spatial information, making it difficult to directly extract and analyze the spatial information of defects.

[0003] Currently, there are two main methods for evaluating the grade of seven types of defects in drainage pipes, including sedimentation, scum, rupture, obstruction, concealed branch connection, foreign body penetration, and residual wall and dam roots:

[0004] 1) Manual Interpretation. Based on extensive experience in video data interpretation, urban drainage pipeline inspection professionals can estimate these seven types of defects based on the image defect's shape, location, and size, thereby deriving a manually assessed defect grade. This method is influenced by factors such as the professional's spatial imagination, experience, and pipeline environment, and is highly subjective and uncertain.

[0005] 2) Deep learning-based defect interpretation technology. Current deep learning-based drainage pipe defect ratings primarily use a Mask r CNN neural network (such as CN113763363A) to identify defects and output a mask. This method then calculates and analyzes the ratio of the mask to the radius, achieving quantitative analysis. This approach is currently only implemented for rupture grade assessment and does not fully consider the spatial properties of the pipe, resulting in certain calculation errors. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a pipeline defect rating method, terminal equipment and storage medium to comprehensively rate pipeline defects and improve the accuracy of pipeline defect rating in response to the shortcomings of the existing technology.

[0007] To solve the above technical problems, the technical solution adopted by the present invention is: a pipeline defect rating method, comprising the following steps:

[0008] The Mask r CNN model is used to identify pipeline images and obtain pipeline defect masks, water surface masks, defect rectangles, and defect categories. The defect categories include sedimentation, scum, rupture, obstruction, branch pipe concealed connection, foreign object penetration, and residual wall and dam root. The defect masks include sedimentation masks, scum masks, rupture masks, obstruction masks, branch pipe concealed connection masks, foreign object penetration masks, and residual wall and dam root masks.

[0009] Obtaining two endpoints A and B of the pipeline defect mask;

[0010] Calculate the center point C of the circle parameter pair of points A and B respectively i Vector AC i , BC i ;

[0011] Calculate the center point C of the circle parameter pair between line AB and points A and B i The ratio of the deposition thickness to the diameter is calculated using the formula (r-d0) / 2r, and the deposition level is determined based on this ratio; where r is the radius of the circle in the circle parameter.

[0012] Calculate the ratio of the water body surface mask to the water body surface area, and determine the scum grade based on the ratio;

[0013] calculating the width and length of the rupture mask and determining the rupture grade according to the width or length;

[0014] Calculate the distance between the center and the four corner points of any circle parameter pair (a, b, r), and record the two corner points with the smallest distance difference and the minimum difference △d. Obtain the two corner points with the smallest distance difference and the minimum difference for all circle parameter pairs, and obtain a difference set and a set of corner point coordinates corresponding to each difference. Take the circle parameter pair corresponding to the minimum △d in the difference set as the circular cross-section of the pipe where the concealed joint is located. Calculate the ratio of the length between the four corner points of the defective rectangular frame on the circle and the adjacent points not on the circle to the diameter of the circular cross-section of the pipe where the concealed joint is located, and determine the branch pipe concealed joint grade based on this ratio. Where a and b are the coordinates of the circle center.

[0015] The ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is calculated, and the level of the residual wall dam root or obstacle, as well as the level of foreign body penetration, is determined based on the ratio.

[0016] Compared to manual estimation and rating methods, this invention significantly eliminates subjectivity in ratings, improving both credibility and accuracy. Compared to existing artificial intelligence rating algorithms, this invention enables quantitative ratings of deposits, scum, obstructions, concealed branch pipe connections, foreign object penetration, and residual wall and dam roots. Furthermore, compared to existing rupture rating methods, this invention considers the three-dimensional spatial properties of ruptures, significantly improving the accuracy of rupture ratings. This invention enables comprehensive rating of pipeline defects, improving their accuracy.

[0017] Preferably, t A and t B The actual distances represented by the circle parameters corresponding to points A and B to the pixels, (x A ,y A ) and (x B ,y B ) are the actual coordinates of points A and B in the image; t A =D / 2r A ;t B =D / 2r B ; r A 、r B A and B are the radii of the two circles corresponding to the circle parameter pair at points A and B, respectively; D is the pipe diameter. The present invention assigns actual three-dimensional properties to the circular cross section, further improving the rating accuracy.

[0018] In the present invention, the specific implementation process of determining the deposition level includes:

[0019] If the ratio of the deposition thickness to the diameter is in the interval [0.2, 0.3), it is judged as level 1 deposition;

[0020] If the ratio of the deposition thickness to the diameter is in the interval [0.3, 0.4), it is judged as level 2 deposition;

[0021] If the ratio of the deposition thickness to the diameter is in the interval [0.4, 0.5), it is judged as level 3 deposition;

[0022] If the ratio of the deposition thickness to the diameter is greater than 0.5, it is judged as level 4 deposition.

[0023] In the present invention, the specific implementation process of determining the fracture level includes:

[0024] If the ratio of the water surface mask to the water surface area is in the interval [0,0.3), it is judged as level 1 scum;

[0025] If the ratio of the water surface mask to the water surface area is in the interval [0.3, 0.6), it is judged as level 2 scum;

[0026] If the ratio of the water surface mask to the water surface area is greater than 0.6, it is judged as level 3 scum.

[0027] In the present invention, the specific implementation process of determining the fracture level includes:

[0028] If the width is less than 2 mm, it is judged as a grade 1 rupture;

[0029] If the width is greater than 2 mm and less than 20 mm, it is considered a grade 2 rupture;

[0030] If the width is greater than 2 mm and the circumferential difference is less than 2 moments, it is judged as a grade 3 rupture;

[0031] If the circumferential difference is greater than 2 moments, it is a level 4 rupture;

[0032] The circumferential parameter calculation process includes: taking the vector (0, 1) as the direction at time 0, calculating the angles between ACi, BCi and the vector (0, 1), and converting them to a 12-hour clock system to obtain times m and n; taking any point K on the line AB, and determining whether point K is within the defect mask area. If so, the circumferential parameter is calculated as (m, n); otherwise, the circumferential parameter is recorded as (n, m);

[0033] The circumferential difference is the circumferential parameter difference, and the circumferential parameter difference=abs((mn)*30°), where abs is the absolute value.

[0034] In the present invention, the specific implementation process of determining the concealed connection of the branch pipe includes:

[0035] If the ratio of the length between the four corner points of the defective rectangular frame on the circle and the adjacent points not on the circle to the diameter of the pipe circular section is less than 0.1, it is judged as a level 1 concealed joint;

[0036] If the ratio of the length between the four corner points of the defective rectangle on the circle and the adjacent points not on the circle to the diameter of the pipe circular section is in the interval [0.1, 0.2), it is determined to be a level 2 concealed connection;

[0037] If the ratio of the length between the four corner points of the defective rectangular frame on the circle and the adjacent points not on the circle to the diameter of the pipe circular section is greater than 0.2, it is judged as a level 3 concealed joint.

[0038] In the present invention, the specific implementation process of determining the level of the residual wall dam root and the obstacle includes:

[0039] If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is less than 0.15, it is determined to be a level 1 residual wall or obstacle;

[0040] If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is in the interval [0.15, 0.25), it is determined to be a level 2 residual wall or obstacle;

[0041] If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is in the interval [0.25, 0.5), it is determined to be a level 3 residual wall or dam root or obstacle;

[0042] If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is greater than 0.5, it is judged to be a level 4 residual wall or obstacle.

[0043] In the present invention, the specific implementation process of determining the foreign body penetration level includes:

[0044] If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is less than 0.1, it is judged as a level 1 foreign body penetration;

[0045] If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is in the interval [0.1, 0.3), it is judged as a level 2 foreign body penetration;

[0046] If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is greater than 0.3, it is judged as level 3 foreign body penetration.

[0047] As an inventive concept, the present invention also provides a terminal device, which includes a processor and a memory; the memory stores a computer program / instructions; the processor executes the computer program / instructions stored in the memory; the computer program / instructions are configured to implement the steps of the above-mentioned method of the present invention.

[0048] As an inventive concept, the present invention also provides a computer storage medium having a computer program / instruction stored thereon; the computer program / instruction, when executed by a processor, implements the steps of the above-mentioned method of the present invention.

[0049] As an inventive concept, the present invention also provides a computer program product, including a computer program / instruction; when the computer program / instruction is executed by a processor, the steps of the above method of the present invention are implemented.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1) This method constructs a circular cross-section of the pipeline and combines it with OCR text recognition technology to extract pipeline size parameters (such as the radius of the circle). This gives the circular cross-section actual three-dimensional properties. Based on machine vision technology, this method extracts the pipeline's morphology and completes pixel-level three-dimensional modeling of the pipeline inspection image. This ensures that each pixel value has actual spatial distance properties, thus breaking through the bottleneck of quantitative pipeline analysis based on a single image.

[0052] 2) By combining artificial intelligence and machine vision technology, this approach fully considers the three-dimensional dimensions of pipeline defects and, in line with regulatory requirements, provides a technical means for the quantitative and precise rating of seven defect levels: pipeline deposits, scum, ruptures, obstructions, concealed branch connections, foreign object penetration, and residual wall and dam roots. Compared to manual estimation and rating methods, this method significantly eliminates subjectivity in ratings and improves rating reliability and accuracy. Compared to existing artificial intelligence rating algorithms, this method achieves quantitative ratings for deposits, scum, obstructions, concealed branch connections, foreign object penetration, and residual wall and dam roots. Furthermore, compared to existing rupture rating methods, this method considers the three-dimensional spatial nature of ruptures, significantly improving the accuracy of rupture ratings. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is the process of pipeline cross-section extraction and physical modeling in an embodiment of the present invention;

[0054] Figure 2 a process for rating embodiments of the present invention;

[0055] Figure 3 is the original image of the pipeline;

[0056] Figure 4 This is the pipeline cross-section extraction result of the embodiment of the present invention;

[0057] Figure 5 Schematic diagram of deposition rating according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The specific implementation steps of the embodiment of the present invention include:

[0059] 1. The main steps of defect recognition and contour extraction based on deep learning are as follows:

[0060] 1) Collect images of seven types of defects, including sediment, scum, cracks, obstacles, hidden branch pipe connections, foreign matter penetration, and residual wall and dam roots, and train a Maskr CNN model that can detect sediment, scum, cracks, obstacles, hidden branch pipe connections, foreign matter penetration, residual wall and dam roots, and water surface;

[0061] 2) Based on the trained Mask R-CNN model (HE K, GKIOXARI G, DOLLáRP, et al. Mask R-CNN[C] / / 2017 IEEE International Conference on Computer Vision (ICCV), 2017, 2980-2988.10.1109 / ICCV.2017.322.), pipeline images are identified to obtain masks of the seven defects and the water surface, defect rectangles (bboxes), and defect categories.

[0062] 2. Pipeline parameter extraction based on OCR text recognition, the main steps are as follows:

[0063] 1) According to the format of CCTV and QV inspection videos, set the image pixel coordinate parameters of the pipe diameter in the inspection image;

[0064] 2) Based on the coordinate parameters, cut the image and use the tesseract library to identify the pipe diameter D;

[0065] 3. The main steps of pipeline 3D modeling based on machine vision are as follows (flow chart as follows Figure 1 shown):

[0066] 1) Convert the pipeline image into a grayscale image;

[0067] 2) Use a Gaussian smoothing filter kernel of size 3×3 to perform noise reduction on the grayscale image to obtain a smoothed noise-reduced image;

[0068] 3) Use Canny to detect the pipeline smoothed and denoised image to obtain the pipeline edge map;

[0069] 4) Based on the digital image opening operation, the burrs and isolated points in the pipeline edge image are eliminated to obtain the edge binary image.

[0070] 5) Detect circular pipes in the binary edge image after eliminating burrs and isolated points in 4) based on the Hough transform: First, set the threshold conditions for the minimum and maximum number of non-zero feature points that a single circle passes through in the binary edge image. In this embodiment, the number of feature points that a circle passes through is at least 7 and at most 100. Then, detect all circle parameter pairs (a, b, r) that meet the conditions in the binary edge image and form a circle parameter pair set C to detect circular pipes in the binary edge image. Among them, a and b are the coordinates of the center of the circle, and r is the radius of the circle. C is a circle parameter pair set containing multiple circle parameter pairs. For example, r = 400.

[0071] 6) Mean shift clustering: The center coordinates (a j ,b j ) is used as a parameter to draw on the two-dimensional ab plane, and the mean shift algorithm is used to find the center of mass of the area with the most dense center points (a i ,b i ), and set a threshold t1 to eliminate circles in the parameter set whose center-to-center distance exceeds t1. After testing, in this embodiment, t1 is set to 30 pixels to achieve the extraction of the cross section of the drainage pipe. The pixel distance calculation formula is as follows:

[0072]

[0073] 7) Calculate the actual distance represented by one pixel in each cross section and assign it to the circle parameter pair (a, b, r, t), where t is the actual distance represented by one pixel on the circle corresponding to the circle parameter pair (a, b, r) (unit: mm). The calculation formula is as follows: d calculated in 6) is the pipe radius (r in the formula below), and t is the ratio of the actual pipe diameter D identified in 2.2 to 2r:

[0074] t=D / 2r;

[0075] 4. Calculation of pipeline defect rating parameters and defect grade assessment (such as Figure 2 shown):

[0076] 1) Calculate all circles and their corresponding parameter pairs and intersection points where the circle parameter pairs obtained in step 3 intersect with the defect mask obtained by the mask r cnn model in step 1;

[0077] 2) Use the intersection point as a vertex to generate a polygon and calculate the actual length of the polygon edge. The formula is as follows:

[0078]

[0079] Where, (t i ,t j ) is the actual distance corresponding to one pixel distance on the circular cross section of the two vertices i and j, (x i ,y i )、(x j ,y j ) are the coordinates of vertices i and j respectively;

[0080] 3) Calculate the circumferential parameters of the defect in the pipeline based on the polygon: ① Calculate the leftmost point A and the rightmost point B of the defect mask; ② Calculate the center point C of the circle parameter pair of points A and B and the circle where A and B are located i Vector AC i , BC i ; ③ Take the vector (0,1) as the 0 o'clock direction and use the cosine formula to calculate AC i , BC i The angle between the vector (0,1) and the time m and n is obtained by converting it to a 12 o'clock clock format as follows:

[0081] angle=0+alpha / 30(alpha>=0);

[0082] angle=6-alpha / 30(alpha<0);

[0083] Where: angle is the moment, the value range is [0,12]; alpha is AC i , BCi The angle with (0,1) ranges from -180° to 180°. ④ Take any point K on line AB and calculate whether point K is within the mask area. If so, the annular parameter is (m,n); otherwise, the annular parameter is (n,m).

[0084] 4) Calculate defect evaluation parameters and implement pipeline rating in accordance with specifications:

[0085] ① For deposition, based on the formula in step 4, step 2), calculate the center point C of the circle parameter pair between line AB and points A and B in step 3). i The actual distance mean d0 (if the AB line is below the center of the circle, d0 is integer, otherwise it is negative), Take the parameter t of the actual distance between the circle parameters corresponding to points A and B and the pixel A and t B 、actual coordinates (x A ,y A ) and (x B ,y B ) are substituted into the formula in step 4, step 2) to calculate d0. The ratio of the deposition thickness to the diameter is calculated using (r-d0) / 2r (this ratio is the rating parameter). If the ratio is in the range [0.2, 0.3), it is classified as a Level 1 deposition; if the ratio is in the range [0.3, 0.4), it is classified as a Level 2 deposition; if the ratio is in the range [0.4, 0.5), it is classified as a Level 3 deposition; if the ratio is greater than 0.5, it is classified as a Level 4 deposition.

[0086] ② For the scum, based on the formula in step 4, step 2), calculate the actual distance between each adjacent inflection point of the polygon corresponding to the scum mask and the water mask (similar to the process of calculating d0 in ①). Combined with the polygon area calculation formula, the actual area S of the scum and water can be obtained. 浮 、S 水 Calculate the ratio of the actual total area of ​​the space corresponding to the scum and the water body (S 浮 / S 水 If the ratio is in the interval [0, 0.3), it is judged as level 1 scum; if the ratio is in the interval [0.3, 0.6), it is judged as level 2 scum; if the ratio is greater than 0.6, it is judged as level 3 scum;

[0087] ③ For the crack, calculate the width and length of the crack mask in 2.1 (in pixels at this time), and calculate the actual width and length (unit: mm) based on the formula in 2). Based on the experience of business personnel, if the width is less than 2mm, it is judged as a Level 1 crack; if the width is greater than 2mm and less than 20mm, it is judged as a Level 2 crack; if the width is greater than 2mm and the circumferential difference calculated in 3) (the circumferential difference is the circumferential parameter difference, converted to an angle system and the absolute value is taken) is less than 2 moments (60°), it is judged as a Level 3 crack; if the circumferential difference calculated in 3) is greater than 2 moments, it is judged as a Level 4 crack;

[0088] ④ For the concealed connection of branch pipes, based on the formula in step 4, step 2), calculate the actual distance between the coordinates of the four corner points and the center of all circle parameter pairs respectively; based on the theorem that the distance between each point on the circle and the center is equal, calculate the difference △d between the two sides with the closest length under the circle parameter pair (a, b, r); take the circle parameter pair corresponding to the smallest △d in the circle parameter pair set as the circular cross section of the pipe where the concealed connection is located (the circle parameter pair represents a circle, and each circle is a pipe cross section, such as Figure 4 Finally, calculate the ratio of the length of the four corner points of the bbox on the circle to the adjacent points not on the circle and the pipe diameter (2r). If the ratio is less than 0.1, it is determined to be a level 1 hidden connection; if the ratio is in the range [0.1, 0.2), it is determined to be a level 2 hidden connection; if the ratio is greater than 0.2, it is determined to be a level 3 hidden connection.

[0089] ⑤ For penetrating foreign objects, residual wall roots, and obstacles: First, calculate the circle parameter pair containing points A and B in 3) based on the circle parameter pair set. Then, calculate the ratio of the defect mask area in 2.1 to the circle pixel area corresponding to the circle parameter pair (a, b, r, t). For residual wall roots or obstacles, if the ratio is less than 0.15, it is classified as a Class 1 residual wall root or obstacle; if the ratio is in the range [0.15, 0.25), it is classified as a Class 2 residual wall root or obstacle; if the ratio is in the range [0.25, 0.5), it is classified as a Class 2 residual wall root or obstacle; if the ratio is greater than 0.5, it is classified as a Class 4 residual wall root or obstacle. For foreign body penetration, if the ratio is less than 0.1, it is judged as level 1 foreign body penetration; if the ratio is in the interval [0.1, 0.3), it is judged as level 2 foreign body penetration; if the ratio is greater than 0.3, it is judged as level 3 foreign body penetration.

[0090] In the embodiment of the present invention, the original pipeline image (such as Figure 3 As shown) using the method of the embodiment of the present invention, according to the farthest distance between the polygon and any of the intersecting circles, the maximum distance is 139 mm, which is 34.86% of the diameter (as shown Figure 5As shown in the figure, the thickness of the deposit is between 30% and 40% according to the Technical Specification for Inspection and Assessment of Urban Drainage Pipelines (CJJ 181-2012). Therefore, the deposit is classified as Level 2. The positions of the leftmost and rightmost points of the deposit on the circle are calculated and converted to a 12-hour format, resulting in a circumferential parameter of 0408 for the defect.

Claims

1. A pipeline defect rating method, characterized in that: The following steps are involved: The Mask r CNN model is used to identify pipeline images and obtain pipeline defect masks, water surface masks, defect rectangles, and defect categories. The defect categories include sedimentation, scum, rupture, obstruction, branch pipe concealed connection, foreign object penetration, and residual wall and dam root. The defect masks include sedimentation masks, scum masks, rupture masks, obstruction masks, branch pipe concealed connection masks, foreign object penetration masks, and residual wall and dam root masks. Obtaining two endpoints A and B of the pipeline defect mask; Calculate the center point C of the circle parameter pair of points A and B respectively i Vector AC i , BC i ; Calculate the center point C of the circle parameter pair between line AB and points A and B i The ratio of the deposition thickness to the diameter is calculated using the formula (r-d0) / 2r, and the deposition level is determined based on this ratio; where r is the radius of the circle in the circle parameter. Calculate the ratio of the water body surface mask to the water body surface area, and determine the scum grade based on the ratio; calculating the width and length of the rupture mask and determining the rupture grade according to the width or length; Calculate the distance between the center and the four corner points of any circle parameter pair (a, b, r), and record the two corner points with the smallest distance difference and the minimum difference △d. Obtain the two corner points with the smallest distance difference and the minimum difference for all circle parameter pairs, and obtain a difference set and a set of corner point coordinates corresponding to each difference. Take the circle parameter pair with the smallest △d in the difference set as the circular cross-section of the pipe where the concealed joint is located. Calculate the ratio of the length between the four corner points of the defective rectangular frame on the circle and the adjacent points not on the circle to the diameter of the circular cross-section of the pipe where the concealed joint is located, and determine the branch pipe concealed joint grade based on this ratio. Where a and b are the coordinates of the circle center. Calculate the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r). Based on this ratio, determine the level of the remaining wall or dam root or obstacle, as well as the level of foreign body penetration. ;t A and t B The actual distances represented by the circle parameters corresponding to points A and B to the pixels, (x A , y A ) and (x B , y B ) are the actual coordinates of points A and B respectively; t A =D / 2r A ;t B =D / 2r B ; r A 、r B are the radii of the two circles corresponding to the circle parameters of points A and B respectively; D is the pipe diameter.

2. The pipeline defect rating method according to claim 1, characterized in that: The specific implementation process for determining the sedimentation grade includes: If the ratio of the deposition thickness to the diameter is in the interval [0.2, 0.3), it is judged as level 1 deposition; If the ratio of the deposition thickness to the diameter is in the interval [0.3, 0.4), it is judged as a level 2 deposition; If the ratio of the deposition thickness to the diameter is in the interval [0.4, 0.5), it is judged as level 3 deposition; If the ratio of the deposition thickness to the diameter is greater than 0.5, it is judged as level 4 deposition.

3. The pipeline defect rating method according to claim 1, characterized in that: The specific implementation process of determining the rupture grade includes: If the ratio of the water surface mask to the water surface area is in the interval [0, 0.3), it is judged as level 1 scum; If the ratio of the water surface mask to the water surface area is in the interval [0.3, 0.6), it is judged as level 2 scum; If the ratio of the water surface mask to the water surface area is greater than 0.6, it is judged as level 3 scum.

4. The pipeline defect rating method according to claim 1, characterized in that: The specific implementation process of determining the rupture grade includes: If the width is less than 2 mm, it is judged as a grade 1 rupture; If the width is greater than 2 mm and less than 20 mm, it is considered a grade 2 rupture; If the width is greater than 2 mm and the circumferential difference is less than 2 moments, it is judged as a grade 3 rupture; If the circumferential difference is greater than 2 moments, it is a level 4 rupture; The circumferential parameter calculation process includes: taking the vector (0, 1) as the direction at time 0, calculating the angles between ACi, BCi and the vector (0, 1), and converting them to a 12-hour clock system to obtain times m and n; taking any point K on the line AB, and determining whether point K is within the defect mask area. If so, the circumferential parameter is calculated as (m, n); otherwise, the circumferential parameter is recorded as (n, m); The circumferential difference is the circumferential parameter difference, and the circumferential parameter difference=abs((mn)*30°), wherein abs is the absolute value.

5. The pipeline defect rating method according to claim 1, characterized in that: The specific implementation process of determining the concealed connection of branch pipes includes: If the ratio of the length between the four corner points of the defective rectangular frame on the circle and the adjacent points not on the circle to the diameter of the pipe circular section is less than 0.1, it is judged as a level 1 concealed joint; If the ratio of the length between the four corner points of the defective rectangle on the circle and the adjacent points not on the circle to the diameter of the pipe circular section is in the interval [0.1, 0.2), it is determined to be a level 2 concealed connection; If the ratio of the length between the four corner points of the defective rectangular frame on the circle and the adjacent points not on the circle to the diameter of the pipe circular section is greater than 0.2, it is judged as a level 3 concealed joint.

6. The pipeline defect rating method according to claim 1, characterized in that: The specific implementation process of determining the level of residual wall dam roots and obstacles includes: If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is less than 0.15, it is determined to be a level 1 residual wall or dam root or obstacle; If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is in the interval [0.15, 0.25), it is determined to be a level 2 residual wall or dam root or obstacle; If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is in the interval [0.25, 0.5), it is determined to be a level 3 residual wall or dam root or obstacle; If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is greater than 0.5, it is judged to be a level 4 residual wall or obstacle.

7. The pipeline defect rating method according to claim 1, characterized in that: The specific implementation process of determining the foreign body penetration level includes: If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is less than 0.1, it is judged as a level 1 foreign body penetration; If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is in the interval [0.1, 0.3), it is judged as a level 2 foreign body penetration; If the ratio of the defect mask area to the circle pixel area corresponding to the circle parameter pair (a, b, r) is greater than 0.3, it is judged as level 3 foreign body penetration.

8. A terminal device, characterized in that: The invention comprises a processor and a memory; the memory stores a computer program / instruction; the processor executes the computer program / instruction stored in the memory; the computer program / instruction is configured to implement the steps of the method according to any one of claims 1 to 7.

9. A computer storage medium having a computer program / instruction stored thereon; characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program / instructions; characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images