Method for quickly judging defect point location and grade of drainage pipeline

By comparing closed-circuit television systems with pre-trained visual model files, the location and severity of defects in drainage pipes can be quickly and accurately determined. This solves the problems of complex methods and poor accuracy in existing technologies, and achieves efficient data support for defect identification and repair.

CN121330320APending Publication Date: 2026-01-13CHINA POWER CONSRTUCTION GRP GUIYANG SURVEY & DESIGN INST CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202511259640.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-04
Publication Date
2026-01-13

AI Technical Summary

Technical Problem

Existing technologies are too complex and inaccurate in determining the location and severity of defects in drainage pipes, making it impossible to quickly and accurately identify defects.

Method used

The system uses a closed-circuit television system to acquire defect locations and visual images in real time, loads visual model files, and compares them with pre-trained models 1-5. The system then uses intelligent interpretation to generate defect record data and produce a report.

Benefits of technology

It enables rapid and accurate identification of defect locations and levels, improving the reliability and efficiency of the assessment and providing precise data for subsequent repairs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121330320A_ABST
    Figure CN121330320A_ABST
Patent Text Reader

Abstract

A method for quickly judging defect point location and grade of a drainage pipeline is realized on the basis of a closed circuit television system, and comprises the following steps: detecting internal defects of the pipeline by adopting a detection lens of the closed circuit television system, and acquiring defect positions and defect visual images in real time; loading a pipeline visual model file in the closed circuit television system, wherein the visual model file is a grading reference model of different defect types; judging the defect type of the defect visual image based on the pipeline visual model file, and selecting a corresponding model in the visual model file for defect grade comparison; and according to a comparison result, judging a defect level and generating a report. A visual model file is trained in advance according to Technical Regulations for Detection and Assessment of Urban Drainage Pipeline, and a closed circuit television system is adopted for detection, so that the judgment method is simple in process, the accuracy and reliability of judgment are improved, a judgment result can be quickly obtained, and the problems of low accuracy and low efficiency of pipeline defect identification in the prior art are solved.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of urban sewage system reconstruction, in particular to a method for quickly judging defect points and grades of a drainage pipeline. BACKGROUND

[0002] With urbanization construction, the coverage rate of drainage pipelines gradually increases. In order to solve problems such as low influent concentration of sewage treatment plants due to defects of the pipe network, and soil and groundwater environment affected by sewage leakage, detecting, judging and repairing the quality of the existing pipe network has become the main direction of future municipal pipe network renovation.

[0003] At present, the main method for defect investigation of the pipe network is to use the detection lens of the closed-circuit television system (CCTV) to take pictures of the internal environment of the pipeline, and then manually interpret the captured videos and pictures. However, due to the difference in perception of different personnel, it is difficult to accurately quantify the size and range of defects, which may lead to misjudgment of defect levels and other phenomena, and ultimately result in improper selection of subsequent treatment measures.

[0004] In order to judge the defect points and grades of the drainage pipeline, the existing technology uses image recognition and grading technology based on deep learning, multi-modal data fusion and intelligent algorithms, intelligent detection equipment and robot technology, and non-contact and stress monitoring technology, each with advantages and disadvantages. In terms of image recognition and grading technology based on deep learning, the existing defect judgment and grading method is too complex, lacks reliable judgment basis, and has a small scope of application.

[0005] For example, the patent document with publication number CN118840587A specifically discloses a drainage pipeline defect classification and grading method and system based on fine-grained image classification. The method includes: constructing and training a fine-grained defect classification and grading model; using the trained model to perform defect classification and grading identification on input drainage pipeline pictures; using an image feature extraction backbone network to extract features from the input drainage pipeline pictures, and inputting the extracted features into a defect classification head and N specific defect classification and grading heads; using the defect classification head to classify the input features, and taking the class corresponding to the maximum value of the N output results as the recognized defect class; using each specific defect classification and grading head to classify the corresponding class of the input features, obtaining the output of the classification head corresponding to the recognized defect class, and calculating the probability that the defect level belongs to the kth level, and taking the level with the maximum probability as the defect level of the drainage pipeline picture. The judgment method is too complex, and the judgment result cannot be obtained quickly, and the accuracy of the judgment is poor.

[0006] Therefore, in the technology of judging the defect points and grades of the drainage pipeline based on deep learning image recognition and grading, there is an urgent need for a technical solution that is simple in process, high in accuracy, and can quickly judge the defect points and grades of the drainage pipeline. SUMMARY

[0007] To solve the technical problems of the prior art that the process is too complex, the judgment result cannot be obtained quickly, and the accuracy of the judgment is poor, the application provides a method for quickly judging defect points and grades of a drainage pipeline, which is realized based on a closed-circuit television system and includes the following contents: A detection lens of the closed-circuit television system is used to detect defects inside the pipeline, and a defect position and a defect visual image are obtained in real time; A pipeline visual model file is loaded in the closed-circuit television system, the visual model file is a reference model graph for grade division of different defect types, and includes model 1 to model 5; First, the defect shape and characteristics are compared based on the pipeline visual model file and the real-time obtained defect visual image, and the defect type of the defect visual image is judged; Then, the corresponding model is selected in the pipeline visual model file according to the defect type, and the defect grade is compared; Finally, the defect grade is determined according to the comparison result, and a report is generated.

[0008] Further, the generation method of the visual model file is that, based on pipeline key parameters including a pipe diameter, a wall thickness, and a material, and according to the shape and characteristics of functional defects and structural defects and corresponding grade division data in the “Technical Specification for Urban Drainage Pipeline Detection and Evaluation”, the model 1 to the model 5 are respectively established, and then training and learning are performed on site detection to generate the visual model file.

[0009] Further, the establishment of the model 1 to the model 5 includes the following contents: A deformation related to a diameter, a branch pipe dark connection, a deposition defect visual file, and a corresponding grade judgment method are input into the model 1; A tree root, a foreign matter penetration, a scaling, an obstacle, and a residual wall dam root visual file related to a water section loss, and a corresponding grade judgment method are input into the model 2; A misaligned opening, a disconnection defect, a wall thickness, and a fixed length of actual length and width data, and a corresponding grade judgment method are input into the model 3; A defect visual file of a breakage occurring without deformation, and a corresponding grade judgment method are input into the model 4; A defect visual file related to a height of undulation and an elevation angle, and a corresponding grade judgment method are input into the model 5, which is used for undulation detection.

[0010] Further, when the real-time obtained defect visual image has multiple defect types, the visual files of the model 1 to the model 5 are combined with each other for comparison and judgment.

[0011] Further, the grade judgment method of the models 1-5 is: according to the input pipe key parameters, the grade division data in the “Urban Drainage Pipe Detection and Evaluation Technical Regulations” is divided; The model 1 grade division includes deformation 1-BX, 2-BX, 3-BX and 4-BX levels; branch pipe hidden connection 1-AJ, 2-AJ, 3-AJ level; deposition CJ-1, CJ-2, CJ-3, CJ-4 level; The model 2 grade division includes foreign matter penetration 1, 2, 3; tree roots, scaling, obstacles and residual wall dam roots are all 1, 2, 3, 4 levels; The model 3 grade division includes misalignment CK-1, CK-2, CK-3, CK-4 level; dislocation 1, 2, 3, 4 level; The model 4 grade division includes rupture 1-PL, 2-PL, 3-PL, 4-PL level; The model 5 grade division includes fluctuation 1, 2, 3, 4 level.

[0012] Further, when the model 1 is selected for comparison, the following operations should be performed: First, rotate the model 1 until the circular outer ring of the model 1 visual file is basically coincided with most of the pipe inner wall of the defect visual image, and the judgment center of the model 1 visual file is coincided with the center of the defect visual image; Then, compare the height of deformation and sediment, the length of branch pipe hidden connection and the position of each segmentation point of the model 1 visual file through the closed circuit television system.

[0013] Further, when the model 2 is selected for comparison, the following operations should be performed: First, focus the detection lens on the defect position, and draw an arc using the two points of the defect position and the pipe wall and the point farthest from the defect position on the pipe wall; Then, rotate the model 2 until the point farthest from the defect position on the pipe wall is coincided with the farthest point of the drawn arc, and determine the grade by comparing the water cross section.

[0014] Further, when the model 3 is selected for comparison, the following operations should be performed: A stripe model with black and white separation is generated with 1 / 2 pipe wall thickness as a condition, and more than 2 times is not set with stripes; when the pipe encounters a misalignment defect, directly align the model 3 with the most serious part of the misalignment defect, and then compare.

[0015] Further, when the model 4 is selected for comparison, the following operations should be performed: The position of the rupture arc length of 60 degrees is determined, and the position is used as the division line of the rupture 3-PL level; the rupture has formed an obvious gap, but the shape of the pipeline is not affected and the rupture is not the rupture 2-PL level; the falling off is less than the range of the division line, which is the rupture 3-PL level, and greater than the range of the division line, which is determined as the rupture 4-PL level.

[0016] Further, when the model 5 is selected for comparison, the following operations should be performed: When the upheaval detection is needed, the view angle of the detection lens is adjusted to 0, and then the elevation angle is gradually adjusted so that the outer circle of the mode 5 coincides with most of the area of the inner diameter of the pipeline, and the elevation angle and displacement at this time are recorded; During recording, the pipeline linear model is established through the change of the elevation angle and the displacement, and the whole upheaval state of the pipeline section is generated; then the upheaval level is obtained by comparing the upheaval height with the pipe diameter.

[0017] The method has the beneficial effects that: through the detection lens of the closed-circuit television system (CCTV) in the detection process, the corresponding visual model file pre-trained is selected for comparison with the on-site defects, the defect record data is directly formed by intelligent interpretation of the system, the data is input, and a report is generated; the judgment method process is simple, the judgment result can be quickly obtained, accurate defect judgment data for later repair is provided, the visual model file pre-trained is generated according to the defect type division and grade division of the “Technical Specification for Urban Drainage Pipeline Detection and Evaluation”, and the accuracy and reliability of judgment can be improved, so that the problems of low pipeline defect identification accuracy and low efficiency in the prior art are solved. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 It is a method flow chart for quickly judging the defect points and levels of the drainage pipeline provided by the application; Figure 2 It is a visual effect schematic diagram of the model 1 provided by the application; Figure 3 It is a visual effect schematic diagram of the model 2 provided by the application; Figure 4 It is a visual effect schematic diagram of the model 3 provided by the application; Figure 5 It is a visual effect schematic diagram of the model 4 provided by the application; Figure 6 It is a visual effect schematic diagram of the model 5 provided by the application for determining the upheaval height; Figure 7 It is a rupture level judgment flow chart provided by the application; Figure 8 It is a schematic diagram of the detection route of the upheaval level provided by the application; Figure 9 It is a leakage level judgment flow chart provided by the application; Figure 10 is a pipeline defect judgment logic diagram provided by the present application; Figure 11 is a pipeline interface defect judgment logic diagram provided by the present application. DETAILED DESCRIPTION

[0019] The technical solutions of the present application are further described below, but the scope of protection is not limited to the description.

[0020] The embodiment of the present application provides a method for quickly judging the defect point and grade of a drainage pipeline, which is realized based on a closed-circuit television system, as shown in the figure. Figure 1 The method comprises the following contents: Step S100, detecting the internal defects of the pipeline by using the detection lens of the closed-circuit television system, and acquiring the defect position and defect visual image in real time; Step S200, loading a pipeline visual model file in the closed-circuit television system, wherein the visual model file is a reference model diagram for grade division of different defect types, and comprises model 1-model 5; Step S300, comparing the defect shape and characteristics based on the pipeline visual model file and the real-time acquired defect visual image, and judging the defect type of the defect visual image; The generation method of the visual model file is as follows: based on the key parameters of the pipeline, including the pipe diameter, wall thickness and material, according to the shape and characteristics of the functional defects and structural defects in the “Technical Specification for Urban Drainage Pipeline Detection and Evaluation” and the corresponding grade division data, the model 1-model 5 are respectively established, and then the training and learning of the field detection are performed to generate the visual model file, as shown in the figure. Figures 2-6

[0021] The establishment of the model 1-model 5 comprises the following contents: The deformation related to the diameter, the hidden connection of branch pipes, the deposition defect visual file and the corresponding grade judgment method are input into the model 1; The root, foreign matter penetration, scaling, obstacle and residual wall dam root visual files related to the loss of water section and the corresponding grade judgment method are input into the model 2; The data related to the wall thickness and the actual length and width fixed length in the case of no deformation, and the corresponding grade judgment method are input into the model 3 in the case of existence of the defect of misalignment and dislocation; The defect visual file of the rupture in the case of no deformation and the corresponding grade judgment method are input into the model 4; The defect visual file related to the height of the undulation and the angle of elevation and the corresponding grade judgment method are input into the model 5 for undulation detection.

[0022] ​Step S400: Then, according to the defect type, select the corresponding model in the pipeline visual model file to compare the defect level. The method for determining the grade of Models 1-5 is as follows: based on the input key parameters of the pipeline, the grade classification is carried out according to the grade classification data in the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines"; The model 1 classification includes deformation levels 1-BX, 2-BX, 3-BX and 4-BX; concealed branch pipe connection levels 1-AJ, 2-AJ and 3-AJ; and deposition levels CJ-1, CJ-2, CJ-3 and CJ-4. Specifically: Deformation level assessment: According to the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines", deformation level 1 is no more than 5% of the pipe diameter; deformation level 2 is 5% to 15% of the pipe diameter; deformation level 3 is 15% to 25% of the pipe diameter; and deformation level 4 is 25% of the pipe diameter. Based on the previously input pipe diameter, Model 1 is divided into 1-BX, 2-BX, 3-BX, and 4-BX. 1-BX represents deformation level 1, which is 5% of the pipe diameter. The determination range of deformation level 1 is between the inner wall of the pipe and the 1-BX line. 2-BX represents deformation level 2, located at the 15% dividing line of the pipe diameter. Deformation level 2 is between the 1-BX and 2-BX lines. 3-BX is divided at 25% of the pipe diameter. Deformation level 3 is between 2-BX and 3-BX. Finally, when the maximum deformation point exceeds the range of 3-BX, it is all deformation level 4, 4-BX.

[0023] Determination of concealed branch pipe connection level: According to the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines", for concealed branch pipe connection level 1 (1-AJ), the length of the branch pipe entering the main pipe is no more than 10% of the main pipe diameter; for concealed branch pipe connection level 2 (2-AJ), the length of the branch pipe entering the main pipe is between 10% and 20% of the main pipe diameter; for concealed branch pipe connection level 3 (3-AJ), the length of the branch pipe entering the main pipe is greater than 20% of the main pipe diameter. Based on the previously input pipe diameter and the above data, visual files are generated using 10% and 20% as dividing lines, respectively.

[0024] Deposition defect level assessment: According to the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines", functional defects are classified as follows: Deposition level 1 (CJ-1): sediment thickness is 20%–30% of the pipe diameter; Deposition level 2 (CJ-2): sediment thickness is 30%–40% of the pipe diameter; Deposition level 3 (CJ-3): sediment thickness is 40%–50% of the pipe diameter; Deposition level 4 (CJ-4): sediment thickness is greater than 50% of the pipe diameter. Based on the previously input pipe diameter and the above data, visual files are generated using 20%, 30%, 40% and the pipe center diameter (50%) as dividing lines.

[0025] The model 2 classification includes foreign object penetration at levels 1, 2, and 3; tree roots, scale, obstacles, and the roots of dilapidated walls and dams are classified as levels 1, 2, 3, and 4. Specifically: Foreign objects penetrating the pipe for a length less than or equal to 10% of the pipe diameter are classified as Class 1; foreign objects penetrating the pipe for a length between 10% and 30% of the pipe diameter are classified as Class 2; foreign objects penetrating the pipe for a length greater than 30% of the pipe diameter are classified as Class 3. The dividing line is at 10% and 30% of the pipe diameter. Roots, scale, obstructions, and remnants of walls and dams are categorized as follows: Level 1: 15% or less into the pipe diameter; Level 2: 15% to 25% into the pipe diameter; Level 3: 25% to 50% into the pipe diameter; Level 4: greater than 50% into the pipe diameter. The dividing lines are 15%, 25%, and 50%, respectively. The model 3 level classification includes misalignment levels CK-1, CK-2, CK-3, and CK-4; and disengagement levels 1, 2, 3, and 4. Specifically: Misalignment grade determination: According to the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines", for misalignment, the wall thickness determined in the early stage is used directly to generate the grade. Misalignment not greater than 1 / 2 of the wall thickness is grade 1 (CK-1); misalignment between 1 / 2 and 1 of the wall thickness is grade 2 (CK-2); misalignment between 1 and 2 times the wall thickness is grade 3 (CK-3); misalignment more than 2 times the wall thickness is grade 4 (CK-4). Disconnection level judgment: Disconnection with a small amount of soil squeezed in is level 1; directly generated disconnection no more than 20mm is level 2; disconnection of 20-50mm is level 3; disconnection greater than 50mm is level 4; considering that disconnection level 1 and 2 defects generally do not need to be treated, this time we directly use 20 and 50mm as the dividing line; The Model 4 classification includes fracture levels 1-PL, 2-PL, 3-PL, and 4-PL. Specifically: According to the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines", a Class 1 rupture (1-PL) is defined as follows: fine cracks are visible on the pipe wall, or a small amount of deposits emerge from the fine cracks, or there is slight peeling, or multiple of the above three conditions exist; a Class 2 rupture (2-PL) has formed a clear gap, but the pipe shape is not affected and there is no detachment at the rupture site; a Class 3 rupture (3-PL) is defined as follows: the circumferential coverage of the remaining fragments at the rupture or detachment of the pipe wall is no greater than 60° of the arc length; a Class 4 rupture (4-PL) is defined as follows: the circumferential coverage of the cracks, fissures, or broken edges of the pipe material is greater than 60° of the arc length; or the circumferential coverage of the detached pipe wall material is greater than 60° of the arc length, or any one or more of the above conditions exist. The process for determining the fracture level is as follows: Figure 7As shown, first determine if the pipe is deformed. If it is deformed, compare it with the deformation and rupture models. If there is no deformation, determine if there is a significant gap in the pipe. If there is no gap, it is classified as a rupture level 1 (1-PL). If there is a gap, determine if the pipe has detached. If there is no gap, it is classified as a rupture level 2 (2-PL). If there is a gap, determine if the detachment of the pipe is greater than 60° of the arc length. If not, it is classified as a rupture level 3 (3-PL). If so, it is classified as a rupture level 4 (4-PL).

[0026] The model 5 level division includes fluctuation levels 1, 2, 3, and 4.

[0027] Specifically: Level 1 is defined as follows: Level 1 is defined as the height of the fluctuation being less than or equal to 20% of the pipe diameter; Level 2 is defined as the height of the fluctuation being between 20% and 35% of the pipe diameter; Level 3 is defined as the height of the fluctuation being between 35% and 50% of the pipe diameter; and Level 4 is defined as the height of the fluctuation being greater than 50% of the pipe diameter.

[0028] Any leakage in sewage pipes will affect the concentration of sewage over time. Therefore, it is necessary to conduct supplementary assessments and repairs in the early stages of potential leakage defects such as foreign objects penetrating or tree roots entering the pipes. Leakage is classified as follows: Level 1 is when water flows out from the defect point and drips continuously from the defect point, flowing along the pipe wall; Level 2 is when water flows away from the pipe wall; Level 3 is when the water surface area of ​​the leak is no more than 1 / 3 of the pipe cross-section, and Level 4 is when it is more than 1 / 3 of the pipe cross-section. The method for determining the leakage level is entered into Model 3. The specific method for determining the leakage level is as follows: Figure 9 As shown: Determine whether the water droplets flow along the pipe wall. If yes, it is classified as Level 1; if no, determine whether a water-passing cross-section is formed. If no, it is classified as Level 2; if yes, determine whether the water-passing cross-section is greater than 1 / 3 of the pipe cross-section. If no, it is classified as Level 3; if yes, it is classified as Level 4.

[0029] When selecting Model 1 for comparison, the following operations should be performed: First, rotate Model 1 until the outer circle of the visual file of Model 1 basically coincides with most of the inner wall of the pipe in the defect visual image, and the judgment center of the visual file of Model 1 coincides with the center of the defect visual image. Then, the height of deformation and sediment, the length of the hidden connection of the branch pipe, and the position of each segment point in the visual file of Model 1 were compared using a closed-circuit television system.

[0030] When selecting Model 2 for comparison, the following operations should be performed: First, focus the detection lens on the defect location, and draw an arc using the two points where the defect location meets the pipe wall and the point on the pipe wall furthest from the defect location. Then, rotate model 2 until the point on the pipe wall furthest from the defect location coincides with the furthest point of the drawn arc. By comparing the water flow cross section, the level is determined.

[0031] When selecting Model 3 for comparison, the following operations should be performed: Using half the pipe wall thickness as a condition, a striped model with transparency and alternating black and white stripes is generated; stripes are not set if the thickness exceeds twice the pipe wall thickness. When the pipe encounters a misalignment defect, model 3 is directly aligned with the most severe part of the misalignment defect before comparison.

[0032] When selecting Model 4 for comparison, the following operations should be performed: The location of the rupture arc length 60° is determined and used as the dividing line for rupture level 3-PL; if a clear gap has been formed at the rupture point, but the shape of the pipe is not affected and there is no detachment, it is classified as rupture level 2-PL; if the detachment is less than the range of the dividing line, it is classified as rupture level 3-PL, and if it is greater than the range of the dividing line, it is classified as rupture level 4-PL.

[0033] When selecting Model 5 for comparison, the following operations should be performed: When fluctuation detection is required, adjust the viewing angle of the detection lens to 0, and then gradually adjust the elevation angle so that the outer ring of mode 5 coincides with most of the inner diameter of the pipe. Record the elevation angle and displacement at this time. During recording, a linear model of the pipeline is established by measuring changes in elevation angle and displacement to generate the entire undulation state of the pipe segment; then, the undulation level is obtained by comparing the undulation height with the pipe diameter. The travel path for undulation level detection is as follows: Figure 8 As shown.

[0034] By adjusting the slope of the entire pipeline section, the overall operation of the pipeline network can be made smoother through optimized design, restoring its designed flow capacity.

[0035] In this embodiment, when the real-time acquired defect visual image contains multiple defect types, the visual file composed of the combination of Model 1 to Model 5 is compared and judged.

[0036] Step S500: Determine the defect level based on the defect level comparison results and generate a report.

[0037] In this embodiment, defect determination is divided into pipe and pipe joint. The determination logic is detailed below. Figure 10 and Figure 11 First, the system learns the basic shapes of defects to enable basic defect type identification, and then fully automates defect level assessment. This logic considers that CCTV inspections require pipe sealing, and in cases of no water or water level less than 20%, scum is not assessed. Furthermore, considering that scale, obstacles, and residual wall / dam roots have consistent classification standards and treatment methods, these defects are not further differentiated.

[0038] The method for rapidly determining the location and level of defects in drainage pipelines provided in this invention uses the detection lens of a closed-circuit television system to select appropriate visual files and compare them with on-site defects during the detection process. The system directly uses intelligent interpretation to form defect record data, inputs the data, and generates a report. This provides accurate defect judgment data for subsequent repairs, thus solving the problems of low accuracy and low efficiency in pipeline defect identification in the prior art.

[0039] The above-disclosed embodiments are merely specific examples of the present invention. However, the present invention is not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for rapidly determining the location and severity of defects in drainage pipes, characterized in that, Based on a closed-circuit television system, including the following: The inspection lens of the closed-circuit television system is used to detect defects inside the pipeline and to acquire the location and visual image of the defects in real time. Load the pipeline visual model file into the closed-circuit television system. The visual model file is a reference model for classifying different defect types, including Model 1 to Model 5. First, the defect shape and features are compared with the real-time acquired defect visual image based on the pipeline visual model file to determine the defect type of the defect visual image; Then, based on the defect type, the corresponding model is selected from the pipeline visual model file for defect level comparison; Finally, based on the defect level comparison results, the defect level is determined and a report is generated.

2. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 1, characterized in that, The method for generating the visual model file is as follows: based on the key parameters of the pipeline, including pipe diameter, wall thickness, and material, and according to the shape and characteristics of functional and structural defects in the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines" and the corresponding grade classification data, the models 1-5 are established respectively, and then the visual model file is generated through on-site inspection training.

3. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 2, characterized in that, The establishment of Models 1-5 includes the following: Visual files of diameter-related deformations, concealed connections of branch pipes, and deposition defects, along with corresponding grade judgment methods, are entered into Model 1. Visual files related to tree roots, foreign object penetration, scaling, obstacles, and residual dam roots, along with corresponding level judgment methods, related to water cross-sectional loss, are entered into Model 2. After defects such as misalignment or disconnection are found, the data of wall thickness and actual length and width on site, as well as the corresponding grade judgment method, are entered into the model 3. Visual files showing cracks without deformation, along with the corresponding level judgment method, are entered into Model 4. Visual files of defects related to undulation height and elevation angle, along with corresponding level judgment methods, are entered into Model 5 for undulation detection.

4. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 3, characterized in that, When the real-time acquired defect visual image contains multiple defect types, the visual file is compared and judged by combining the visual files of Model 1-Model 5.

5. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 4, characterized in that, The method for determining the grade of Models 1-5 is as follows: based on the input key parameters of the pipeline, the grade classification is carried out according to the grade classification data in the "Technical Specification for Inspection and Evaluation of Urban Drainage Pipelines"; The model 1 classification includes deformation levels 1-BX, 2-BX, 3-BX and 4-BX; concealed branch pipe connection levels 1-AJ, 2-AJ and 3-AJ; and deposition levels CJ-1, CJ-2, CJ-3 and CJ-4. The model 2 classification includes foreign object penetration at levels 1, 2, and 3; tree roots, scale, obstacles, and the roots of dilapidated walls and dams are classified as levels 1, 2, 3, and 4. The model 3 level classification includes misalignment levels CK-1, CK-2, CK-3, and CK-4; and disengagement levels 1, 2, 3, and 4. The Model 4 classification includes fracture levels 1-PL, 2-PL, 3-PL, and 4-PL. The model 5 level division includes fluctuation levels 1, 2, 3, and 4.

6. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 5, characterized in that, When selecting Model 1 for comparison, the following operations should be performed: First, rotate Model 1 until the outer circle of the visual file of Model 1 basically coincides with most of the inner wall of the pipe in the defect visual image, and the judgment center of the visual file of Model 1 coincides with the center of the defect visual image. Then, the height of deformation and sediment, the length of the hidden connection of the branch pipe, and the position of each segment point in the visual file of Model 1 were compared using a closed-circuit television system.

7. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 5, characterized in that, When selecting Model 2 for comparison, the following operations should be performed: First, focus the detection lens on the defect location, and draw an arc using the two points where the defect location meets the pipe wall and the point on the pipe wall furthest from the defect location. Then, rotate model 2 until the point on the pipe wall furthest from the defect location coincides with the furthest point of the drawn arc. By comparing the water flow cross section, the level is determined.

8. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 5, characterized in that, When selecting Model 3 for comparison, the following operations should be performed: Using half the pipe wall thickness as a condition, a striped model with transparency and alternating black and white stripes is generated; if the thickness exceeds twice the pipe wall thickness, no stripes are set. When the pipe encounters a misalignment defect, model 3 is directly aligned with the most severe part of the misalignment defect before comparison.

9. The method for rapidly determining the location and severity of defects in drainage pipes as described in claim 5, characterized in that, When selecting Model 4 for comparison, the following operations should be performed: The location of the rupture arc length 60° is determined and used as the dividing line for rupture level 3-PL; if a clear gap has been formed at the rupture point, but the shape of the pipe is not affected and there is no detachment, it is classified as rupture level 2-PL; if the detachment is less than the range of the dividing line, it is classified as rupture level 3-PL, and if it is greater than the range of the dividing line, it is classified as rupture level 4-PL.

10. The method for rapidly determining the location and level of defects in drainage pipes as described in claim 5, characterized in that, When selecting Model 5 for comparison, the following operations should be performed: When fluctuation detection is required, adjust the viewing angle of the detection lens to 0, and then gradually adjust the elevation angle so that the outer ring of mode 5 coincides with most of the inner diameter of the pipe. Record the elevation angle and displacement at this time. During recording, a linear model of the pipeline is established by measuring changes in elevation angle and displacement to generate the entire undulation state of the pipe segment; then, the undulation level is obtained by comparing the undulation height with the pipe diameter.

Citation Information

Patent Citations

  • Fine-grained image classification-based drainage pipeline defect classification and grading method and system

    CN118840587A

  • Pipeline defect quantitative evaluation method

    CN118608518A