Drainage pipe desilting and repair assessment method, device and desilting and repair robot

Abstract

The present invention relates to the technical field of dredging assessment, and in particular to a dredging and repair assessment method, device and dredging and repair robot for drainage pipes. The method comprises the following steps: obtaining a multi-directional image of the interior of a drainage pipe after dredging and a pipe ultrasonic detection signal; performing visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional image of the interior of the drainage pipe to construct a three-dimensional pipe panoramic model; performing deep residual sludge recognition on the multi-directional image of the interior of the drainage pipe, and performing sludge spatial distribution analysis to generate residual sludge spatial distribution data; performing reflection deviation compensation optimization on the pipe ultrasonic detection signal, and performing sludge morphology calculation one by one to generate the morphological characteristics of each residual sludge; performing sludge composition characteristic analysis based on the pipe ultrasonic detection signal to obtain the composition characteristics of each residual sludge. The present invention realizes efficient and accurate dredging and repair assessment.

Description

Technical Field

[0001] The present invention relates to the technical field of dredging assessment, and in particular to a dredging and repair assessment method and device for a drainage pipe, and a dredging and repair robot. Background Art

[0002] With the continuous acceleration of urbanization, drainage pipes, as an important part of urban infrastructure, play a vital role in urban rainwater, sewage discharge, flood prevention and disaster prevention. During the long-term use of drainage pipes, due to sediment deposition, garbage blockage and water corrosion, a large amount of silt, sediment and other substances often accumulate in the pipes, resulting in a decrease in the pipe's water flow capacity, and even serious problems such as blockage and leakage. In order to ensure the normal operation of drainage pipes and avoid urban waterlogging and environmental pollution caused by pipe blockage, regular dredging becomes particularly important.

[0003] However, traditional drainage pipe desilting methods rely on manual inspection, manual cleaning or simple mechanical cleaning equipment. These methods have disadvantages such as low efficiency, incomplete cleaning, and easy damage to the pipeline. Especially for some deeply buried pipelines or pipeline parts that cannot be directly contacted, traditional desilting methods often find it difficult to accurately judge the distribution and accumulation of silt, and the cleaning effect is not ideal, and even causes secondary pollution. In addition, manual cleaning is also faced with a large amount of data analysis workload and potential risks in the desilting process, resulting in high operating costs and safety hazards. Therefore, how to obtain the silt accumulation data in the pipeline in real time and accurately, how to quantitatively evaluate the silt distribution and the degree of damage to the pipeline structure, and how to optimize the desilting operation plan based on the evaluation results are still technical problems that need to be solved urgently. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a drainage pipe dredging and repair assessment method, device and dredging and repair robot to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a drainage pipe desilting and repair assessment method, comprising the following steps: Step S1: obtaining a multi-directional image of the interior of the drainage pipe after desilting and a pipe ultrasonic detection signal; performing visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional image of the interior of the drainage pipe to construct a three-dimensional pipe panoramic model; Step S2: performing deep residual sludge identification on the multi-directional image inside the drainage pipe and performing sludge spatial distribution analysis to generate residual sludge spatial distribution data; Step S3: optimizing the reflection deviation compensation of the pipeline ultrasonic detection signal, and calculating the sludge morphology one by one, so as to generate the morphological characteristics of each residual sludge; Step S4: Analyze the sludge composition characteristics according to the pipeline ultrasonic detection signal, so as to obtain the composition characteristics of each residual sludge; Step S5: Rendering the three-dimensional pipeline panoramic model dynamically based on the composition characteristics of each residual sludge location, the morphological characteristics of each residual sludge location, and the residual sludge spatial distribution data, thereby constructing a drainage pipeline residual sludge panoramic rendering model; Step S6: Calculate the residual siltation spatial coverage rate of the panoramic rendering model of the drainage pipe residual siltation, and conduct a comprehensive quantitative analysis of silt removal and restoration to obtain a quantitative assessment report on silt removal and restoration.

[0006] The present invention can fully understand the current situation inside the pipeline by acquiring multi-directional images and ultrasonic signals. The images provide visual information of the pipeline, while the ultrasonic detection signals help to reveal the material status and potential problems of the pipeline. The multi-directional images help to form a geometric structure model of the pipeline, support accurate three-dimensional modeling, and provide a data basis for subsequent analysis. The pipeline ultrasonic detection signal can be used to detect thickness changes, damage, cracks, etc. of the pipeline, provide real data after dredging treatment, and help evaluate the dredging effect. Through image analysis, the residual sludge that still exists after dredging treatment can be accurately identified, providing accurate information for subsequent dredging work, and the distribution of residual sludge can be accurately analyzed, thereby helping to determine which areas need to be cleaned in focus to avoid duplication of labor. According to the depth and distribution of the sludge, the subsequent dredging plan is optimized to improve work efficiency and effect. The reflection deviation compensation optimization can effectively eliminate errors caused by detection equipment or environmental factors, thereby improving the accuracy of ultrasonic signals. By calculating the morphological characteristics of the sludge and understanding the structure of the residual sludge, dredging personnel can be more targeted in treatment and avoid unnecessary intervention, which helps to identify and locate sludge of special forms, especially Especially for the silt blocks that are difficult to remove, provide guidance for repair. The composition analysis of silt can reveal its source and properties, such as organic matter content, inorganic matter content, etc., to provide a basis for subsequent treatment methods. Silt with different components requires different cleaning methods. Composition characteristic analysis helps to formulate more accurate cleaning plans and improve the repair effect. Through composition analysis, the effect after cleaning can be predicted, and whether special treatment or secondary cleaning is necessary can be evaluated. Dynamic rendering can intuitively display the morphological changes of silt inside the pipeline, helping relevant personnel understand the distribution and properties of silt. The rendering model provides managers with intuitive data support to help them make more reasonable decisions and adjust cleaning strategies in time. Through the three-dimensional panoramic model, dredging work becomes more transparent, and relevant personnel have a more comprehensive understanding of the internal situation of the pipeline. By calculating the spatial coverage rate of residual silt, the cleaning effect can be accurately evaluated to help confirm the completion of dredging. Comprehensive quantitative analysis provides a clear evaluation report for dredging and repair work, helps to evaluate the effect of the entire cleaning process, discover potential problems, and provide data support for future maintenance. The quantitative evaluation report can give the team clear goals, avoid repeated cleaning and omissions, and significantly improve work efficiency.

[0007] Preferably, step S1 comprises the following steps: Step S11: obtaining a multi-directional image of the interior of the drainage pipe after desilting and a pipe ultrasonic detection signal; Step S12: visually identifying pipe components on the multi-directional image of the interior of the drainage pipe, and marking a plurality of pipe internal components; Step S13: performing pipeline topological connection analysis according to multiple pipeline internal components, thereby generating a pipeline topological connection relationship; Step S14: performing internal geometric structure arrangement analysis on the multi-directional images of the interior of the drainage pipe to generate internal geometric structure data of the pipe; Step S15: Perform three-dimensional point cloud modeling based on the internal geometric structure data of the pipeline and the pipeline topological connection relationship to construct a three-dimensional pipeline panoramic model.

[0008] The present invention can fully reflect the actual state of the pipeline after desilting, including sludge residue, pipeline damage, corrosion and other problems, by acquiring multi-directional images and pipeline ultrasonic detection signals. Multi-directional images provide visual surface information, and ultrasonic signals can detect pipeline wall thickness, cracks and other potential structural problems. These information are combined to provide comprehensive and accurate data support for subsequent processing. Through image analysis technology, the internal components of the pipeline are visually identified, and different pipeline components (such as pipeline connection ports, valves, turning parts, etc.) are automatically marked to improve recognition efficiency and reduce manual intervention. The position information of each component inside the pipeline is marked to provide accurate positioning data for subsequent topological connection analysis and geometric structure analysis. Automatic marking of pipeline components makes pipeline management more intelligent, avoids omissions, and reduces the probability of human errors. Pipeline topological connection analysis helps to build a global structural model of the pipeline by revealing the connection relationship between the internal components of the pipeline. In this way, the layout, connection method and flow path of the pipeline can be accurately understood. By clarifying the topological connection relationship, the connection points or vulnerable parts with problems can be more accurately identified, which helps to formulate targeted repair plans. During the desilting and repair process, understanding the topological structure of each part of the pipeline helps repair personnel avoid misoperation and unnecessary intervention, and improves the repair effect. Internal geometric structure analysis helps to reveal details such as the spatial distribution inside the pipeline, the shape of the pipeline wall, and the curvature change, providing basic data for evaluating the actual shape of the pipeline. Through geometric structure data, the specific arrangement inside the pipeline can be clearly understood, which helps to evaluate the effect of desilting treatment, such as whether there are structural anomalies that hinder water flow. After analyzing the geometric structure data, it is determined which parts have deep siltation or difficult to clean areas, so as to formulate more efficient and accurate desilting strategies. Through point cloud modeling technology, a highly accurate three-dimensional pipeline panoramic model can be generated to truly reflect the spatial layout and structure of the pipeline. This model intuitively displays the internal geometry of the pipeline and the way the pipeline is connected. The three-dimensional panoramic model provides intuitive and efficient data support for desilting and repair evaluation. Managers and engineers use this model to comprehensively evaluate the desilting effect of the pipeline and find potential repair problems. Through the three-dimensional model, repair personnel can accurately determine the problem area, avoid blind repairs, reduce ineffective work, and improve the efficiency and effectiveness of the entire cleaning process. The 3D panoramic model supports dynamic updates of pipeline status, tracks changes during the desilting process, and helps monitor the progress and effectiveness of the cleaning work.

[0009] Preferably, the three-dimensional point cloud modeling is specifically: Extracting an image in each orientation according to the multi-orientation images inside the drainage pipe; Identify key component points of each orientation image and mark multiple key component points; Perform multi-angle precise registration of images based on multiple key component points to generate multi-angle registration data; Based on the multi-angle registration data, multi-angle stitching processing is performed on the multi-directional images inside the drainage pipe, so as to generate a panoramic image of the inside of the pipe; According to the internal geometric structure data of the pipeline and the pipeline topological connection relationship, the three-dimensional point cloud modeling of the panoramic image inside the pipeline is carried out to construct a three-dimensional pipeline panoramic model.

[0010] By extracting images from each orientation, the present invention can fully understand the various angles and detailed information of the pipeline, providing a basis for subsequent image analysis. The acquisition of multi-directional images can cover different corners inside the pipeline, ensuring that there are no omissions and avoiding blind spots in analysis. Through multi-directional image acquisition, the efficiency of collecting the internal conditions of the pipeline is greatly improved, saving manual inspection time. Through image recognition technology, key components (such as pipeline connections, valves, elbows, etc.) in the pipeline can be automatically marked. This automated recognition greatly reduces the workload and error rate of manual marking. The accurate marking of key component points provides reliable data support for subsequent image registration and geometric analysis. By marking multiple key points, a structured data set is formed, which is convenient for subsequent topological analysis and three-dimensional modeling. Through precise multi-angle registration, images of different perspectives can be aligned, the deviation caused by different shooting angles can be eliminated, and the consistency of image data can be ensured. Multi-angle registration can carefully align various parts inside the pipeline, especially complex structural parts, to ensure seamless connection of images after splicing. Through precise registration, non-overlap or gap phenomena between images are avoided, ensuring the high quality of subsequent splicing data. By stitching based on multi-angle registration data, images from multiple directions can be seamlessly merged into a complete panoramic image, providing clear and coherent visual information for the overall assessment of the pipeline. The panoramic image can show the internal structure of the entire pipeline, including the distribution of different parts of the pipeline and key components, making the pipeline status clear at a glance. The generated panoramic image not only provides visual support for dredging and repair assessment, but also provides reliable data for subsequent geometric structure analysis, repair plan formulation, etc. By combining geometric structure data and topological connection relationships, an accurate three-dimensional point cloud model can be generated to truly reproduce the spatial structure of the pipeline and provide a deeper spatial perspective than a plane image. The three-dimensional point cloud model can accurately display the geometric characteristics, connection relationships and potential structural problems of the pipeline, helping engineers to conduct more accurate dredging and repair assessments. The three-dimensional model provides clearer visual support for pipeline dredging, helps evaluate the location, shape and quantity of residual sludge, guides the implementation of dredging work, and verifies the repair effect. The three-dimensional pipeline panoramic model not only visualizes the spatial layout of the pipeline, but also provides decision makers with intuitive feedback on the repair effect, which is conducive to scientific decision-making and management.

[0011] Preferably, step S2 comprises the following steps: Step S21: performing adaptive brightness enhancement processing on the multi-directional image inside the drainage pipe to obtain an adaptive brightness enhanced image; Step S22: performing super-resolution texture detail optimization on the adaptive brightness enhanced image to construct a detail optimized pipeline image; Step S23: performing deep residual sludge recognition on the detail optimized pipeline image, marking each location of residual sludge in the pipeline; Step S24: performing three-dimensional positioning of each residual sludge in the pipeline space to generate three-dimensional coordinates of each residual sludge; Step S25: Performing sludge spatial distribution analysis on each three-dimensional coordinate of residual sludge, thereby generating residual sludge spatial distribution data.

[0012] The present invention enhances the details of the image by adjusting the brightness and contrast of different areas through adaptive brightness enhancement processing, so that the visualization effect of the image in low brightness or high contrast environment is optimized. It helps to view various details inside the pipeline more clearly, especially dark areas and shadow areas. Brightness enhancement helps to improve the quality of the image under different lighting conditions, thereby ensuring the accuracy of the data in the subsequent processing and analysis process. Adaptive brightness enhancement can be adjusted for different shooting conditions (such as uneven lighting, interference of light sources inside and outside the pipeline, etc.), ensuring that the final image has high clarity under various conditions. Through super-resolution texture detail optimization technology, the details of the image are enhanced, so that the original low-resolution image presents more texture information. In this way, the microstructures inside the pipeline (such as cracks, foreign matter and silt accumulation areas) can be displayed more clearly. Texture detail optimization can significantly improve the clarity of the image, provide more refined image support for subsequent deep residual silt identification, and help discover tiny silt accumulation or pipeline damage. The optimized image has higher detail expression, thereby improving the accuracy and efficiency of subsequent processing steps (such as silt identification and three-dimensional modeling). The deep residual silt identification technology efficiently and accurately identifies the location of silt from the optimized image, especially in images with rich details, which can help quickly locate the accumulation area of ​​silt. The automation technology of this step significantly reduces manual intervention, improves the speed and accuracy of silt identification, and reduces human errors. Each residual silt will be accurately marked, providing detailed geographic information for subsequent spatial positioning and distribution analysis, ensuring the accuracy of silt analysis data. Through three-dimensional positioning technology, the coordinates of each residual silt will be accurately calculated and three-dimensional data will be generated. In this way, not only can the distribution of silt be known, but also its specific location in space can be clearly identified. Compared with traditional planar image analysis, three-dimensional positioning can provide more detailed spatial information to help identify silt accumulation that is difficult to locate in two-dimensional images. Three-dimensional coordinates can help repair personnel accurately determine the severity of silt accumulation and its specific location, avoid unnecessary repair operations, and thus develop a more efficient dredging plan. Through spatial distribution analysis, the spatial layout of residual silt inside the pipeline is comprehensively evaluated, and its depth, distribution breadth and relationship with other parts of the pipeline are analyzed. The analysis results help repair personnel understand the distribution pattern of silt in the pipeline, optimize the dredging plan, and focus on areas with more serious silt accumulation, thereby improving the efficiency of repair. The generated spatial distribution data can quantitatively evaluate the dredging effect of the pipeline, understand the changes in silt distribution before and after cleaning, and provide an objective basis for the subsequent evaluation of the repair effect.

[0013] Preferably, step S3 specifically comprises the following steps: Step S31: performing high-frequency filtering on the pipeline ultrasonic detection signal to obtain a filtered and noise-reduced ultrasonic signal; Step S32: calculating the pipeline reflection deviation of the filtered and denoised ultrasonic signal to generate a pipeline reflection deviation parameter; Step S33: performing reflection deviation compensation optimization on the pipeline reflection deviation parameters to generate deviation compensation ultrasonic parameters; Step S34: Calculate the sludge morphology of each residual sludge location one by one based on the deviation-compensated ultrasonic parameters based on the internal geometric structure data of the pipeline, thereby generating the morphological characteristics of each residual sludge location.

[0014] The present invention can remove high-frequency noise in ultrasonic signals, especially interference signals from the environment or equipment, through high-frequency filtering processing, to ensure the accuracy of subsequent signal analysis. After filtering and noise reduction, the ultrasonic signal is clearer, can better reflect the real situation inside the pipeline, and avoid the influence of noise on sludge identification and other evaluation parameters. By calculating the reflection deviation, the changes in the internal structure of the pipeline, especially the reflection characteristics of damage, cracks or sludge accumulation areas of the pipeline wall, are analyzed to reveal potential problems of the pipeline. The reflection deviation parameter can help accurately identify the morphological changes inside the pipeline and provide more accurate damage information and sludge distribution than conventional detection methods. Reflection deviation compensation optimization eliminates the influence of measurement errors, equipment deviations and other factors, making the processing results of ultrasonic detection signals more accurate and reliable. Compensation optimization can eliminate the influence caused by irregular shape of the pipeline itself, equipment setting errors or external environmental factors, and ensure the accuracy of detection data. Using the compensated ultrasonic signal and the geometric structure data of the pipeline, the morphological characteristics of the residual sludge (such as sludge thickness, shape, accumulation area, etc.) are calculated one by one. This provides specific guidance for sludge repair and cleaning work. Accurate calculation of silt morphology reveals the distribution pattern and accumulation characteristics of silt, helping restoration personnel to develop more scientific dredging plans. Avoid misjudging the scope of silt accumulation, ensure that dredging work is highly targeted, and avoid excessive cleaning or omissions. By accurately calculating the morphological characteristics of each silt, the cleaning path and restoration strategy are optimized, improving restoration efficiency and reducing costs.

[0015] Preferably, the specific steps of step S34 are: Performing acoustic wave reflection waveform analysis on deviation-compensated ultrasonic parameters to generate pipeline acoustic wave reflection waveform curve; Calculate the time difference between adjacent peaks of the pipeline sound wave reflection waveform curve to obtain the propagation time of the reflection signal; According to the pipeline sound wave reflection waveform curve, the reflection signal amplitude change is analyzed to generate the reflection signal amplitude change characteristics; The distance between the sediment and the pipe wall is calculated based on the internal geometric structure data of the pipe and the propagation time of the reflected signal, thereby obtaining the sludge thickness parameter; The size of the silt area is estimated based on the change characteristics of the reflected signal amplitude to generate the size of the silt area; The silt morphology is calculated one by one according to the silt thickness parameters and the silt area size, so as to generate the morphological characteristics of each residual silt.

[0016] The present invention obtains the reflection waveform curve in the pipeline by analyzing the ultrasonic reflection waveform after deviation compensation, reflecting the state of the internal structure of the pipeline, including any cracks, blockages or sludge accumulation areas. The sound wave reflection waveform can reveal problems such as unevenness, damage or accumulation of foreign matter on the pipeline wall, provide clear pipeline condition information, and help evaluate its repair needs. By calculating the time difference between adjacent peaks in the reflection waveform, the propagation time of the reflection signal is determined, which is crucial for estimating the thickness of the sludge. The longer the propagation time, the greater the thickness of the sludge. This step accurately analyzes the propagation characteristics of the signal by quantifying the time difference, helping to determine the sludge accumulation in different areas, thereby obtaining more accurate thickness data. The change in the amplitude of the reflection signal is used to analyze the distribution of sludge in the pipeline. Generally, the accumulation of sludge will cause the reflection amplitude of the sound wave to weaken or change. By analyzing the amplitude change, the distribution area of ​​sludge and its severity can be accurately identified. The amplitude change characteristics can also provide clues about the type of sludge (such as soft sludge, hard sludge, etc.), helping to evaluate the difficulty of cleaning different types of sludge. Combined with the geometric structure data of the pipeline and the propagation time of the reflected signal, the precise distance between the silt and the pipe wall is calculated to obtain the actual thickness of the silt. This method can improve the accuracy of thickness measurement and avoid errors in traditional methods. Through accurate thickness parameters, it can provide a scientific basis for the formulation of dredging plans to avoid excessive cleaning or missing areas. The obtained thickness data is used for visual display to help repair personnel intuitively understand the accumulation of silt and support subsequent dredging and repair operations. The area of ​​silt accumulation can be inferred through the change characteristics of the reflected signal amplitude. Combined with the geometric structure data of the pipeline, it helps to determine the spatial distribution of silt in the pipeline. This step not only provides information on the thickness of the silt, but also enables a comprehensive assessment of the spatial distribution of silt in the pipeline through area estimation. The obtained silt area data provides quantitative information for repair personnel to help evaluate the workload required for cleaning, thereby formulating a reasonable dredging plan and avoiding waste of resources. By combining the thickness and area of ​​the silt, the detailed morphological characteristics of each silt are calculated one by one, including the accumulation shape, distribution range, and concentrated area of ​​the silt. The calculation of morphological characteristics not only helps the current dredging work, but also provides a reference for subsequent pipeline maintenance and inspection, tracking the accumulation of silt through change comparison.

[0017] Preferably, the specific steps of step S4 are: Step S41: Perform Fourier transform on the deviation-compensated ultrasonic parameters to obtain multiple segments of ultrasonic spectrum data; Step S42: performing spectrum bandwidth analysis on the multiple segments of ultrasonic spectrum data segment by segment to generate spectrum bandwidth characteristics of each segment; Step S43: performing sludge particle estimation based on the spectrum bandwidth characteristics of each segment to obtain particle characteristics in the sludge; Step S44: performing spectrum peak analysis on the multiple segments of ultrasonic spectrum data to obtain the spectrum peak of each segment; Step S45: mining the sludge viscosity based on the spectrum peak of each section to obtain the sludge viscosity characteristics; Step S46: Calculate the resonance frequency of the multiple ultrasonic spectrum data to obtain the degree of sludge consolidation; Step S47: Analyze the sludge composition characteristics according to the sludge consolidation degree, sludge viscosity characteristics and particle characteristics in the sludge, so as to obtain the composition characteristics of each residual sludge.

[0018] The present invention converts the ultrasonic signal in the time domain into a frequency domain signal through Fourier transform, which helps to better understand the information hidden in the signal, especially the properties of the material (such as sludge) in the pipeline. The spectrum data provides the frequency composition of the ultrasonic signal, which can reveal the density, structure and other physical properties of the sludge, which is crucial for analyzing the properties of the sludge. The spectrum bandwidth analysis reveals the size and distribution range of the sludge particles and their influence on the propagation of sound waves. The wider the bandwidth, the larger the sludge particles or the more complex the sludge structure. Different bandwidth characteristics correspond to different sludge types, such as soft mud, hard mud, etc. The change in bandwidth can help classify different types of sludge and provide a basis for remediation strategies. The spectrum bandwidth is closely related to the size of the sludge particles. Through the spectrum bandwidth characteristics, the size of the sludge particles is inferred, and then the physical properties of the sludge such as texture and density are estimated. Understanding the characteristics of sludge particles can help dredging personnel choose appropriate cleaning methods and tools. Silt with larger particles requires more powerful cleaning equipment, while sludge with smaller particles is more suitable for cleaning with high-precision equipment. The particle characteristics also help to classify the sludge (such as fine particles, coarse particles, etc.), so as to adopt different cleaning methods. The spectrum peak reflects the acoustic properties of the material inside the pipeline, including the structural density, rigidity and viscosity of the sludge. Through peak analysis, the characteristics of sludge in different areas can be understood. A high peak means that the sludge is relatively solid or viscous, while a low peak means that the sludge is loose or softened. Peak analysis can effectively evaluate the compaction degree of sludge. The spectrum peak has a certain relationship with the viscosity of the sludge. By analyzing the spectrum peak, the viscosity of the sludge is revealed, and its composition, properties and its impact on the cleaning equipment can be inferred. Silt with different viscosities requires different cleaning methods. Silt with higher viscosity requires special equipment or chemical additives for treatment, while sludge with lower viscosity can be removed by mechanical cleaning. Through the calculation of the resonant frequency, the degree of consolidation and density of the sludge can be understood. A higher resonant frequency usually means that the sludge is relatively solid or has a higher density. The degree of consolidation directly affects the difficulty of cleaning. Silt with a higher degree of consolidation is usually more difficult to clean and requires more professional equipment or methods. Through the calculation of the degree of consolidation, targeted data can be provided to repair personnel to help them adopt appropriate cleaning techniques or methods to reduce ineffective labor. By comprehensively analyzing the degree of consolidation, viscosity, and particle characteristics, we can fully understand the physical and chemical properties of each residual sludge and its impact on pipeline cleaning and repair. This analysis can classify different types of sludge, such as soft mud, hard mud, and mud-sand mixtures, to provide more accurate support for repair. Through accurate sludge component characteristic analysis, the most appropriate cleaning method and repair plan can be selected according to different sludge components, avoiding misjudgment and waste of resources, and ensuring efficient and accurate repair work.

[0019] Preferably, the specific steps of step S5 are: Step S51: performing sludge position matching on the three-dimensional pipeline panoramic model based on the three-dimensional coordinates of each residual sludge location to obtain sludge position matching data; Step S52: Visually marking the three-dimensional pipeline panoramic model with the residual sludge spatial distribution data according to the sludge position matching data, thereby obtaining a residual sludge distribution visualization model; Step S53: Based on the morphological characteristics of each residual sludge, the residual sludge distribution visualization model is dynamically rendered according to the morphological characteristics of each residual sludge, and attribute mapping is performed according to the component characteristics of each residual sludge, so as to construct a panoramic rendering model of residual sludge in the drainage pipeline.

[0020] The present invention can accurately determine the specific position of the silt in the pipeline by matching the three-dimensional coordinates of each residual silt with the three-dimensional pipeline panoramic model, which provides accurate spatial information for subsequent dredging work. Accurate silt position matching can optimize the dredging plan, help repair personnel accurately identify the areas that need to be cleaned, improve work efficiency, and avoid omissions. By mapping the spatial distribution data of the residual silt to the three-dimensional pipeline panoramic model, the silt distribution in the pipeline can be presented in a graphical manner. In this way, the repair personnel can intuitively see which areas are seriously accumulated with silt and which areas are relatively clean. The visualization model can help identify the priority areas in the cleaning work. For example, the more serious silt accumulation areas are highlighted by color, marking, etc., to optimize the dredging process. Dynamic rendering of morphological features can show the morphological changes of silt over time. This dynamic effect can simulate changes in silt deposition and flow, and provide pipeline maintenance personnel with more realistic predictions of pipeline environmental changes. Combined with the composition characteristics of silt, different types of silt (such as soft mud, hard mud, sand, etc.) are accurately marked in the visual model through attribute mapping technology to display the silt composition and characteristics in different areas. This helps repair personnel understand the specific conditions of silt in different areas and choose appropriate cleaning strategies. The combination of dynamic rendering and attribute mapping can provide more information for the repair team, enabling them to formulate more scientific and reasonable dredging plans based on the changing laws and physical properties of silt. For example, consolidated silt and soft silt require different treatment methods, and dynamic rendering makes this difference intuitively reflected.

[0021] Preferably, the specific steps of step S6 are: Step S61: calculating the residual siltation space coverage rate of the drainage pipe residual siltation panoramic rendering model to generate the residual siltation space coverage rate; Step S62: performing a pipeline blockage risk analysis on the drainage pipeline residual siltation panoramic rendering model based on the residual siltation space coverage rate to generate a pipeline residual siltation blockage risk; Step S63: Performing a structural damage risk assessment on the drainage pipe residual siltation panoramic rendering model to obtain pipe structure damage risk data; Step S64: obtaining an original image of the uncleaned drainage pipe; Step S65: Based on the original image of the drainage pipe that has not been cleaned, a comprehensive quantitative analysis of the pipeline structure damage risk data and the pipeline residual silt blockage risk is performed to obtain a quantitative assessment report on the silt removal and repair.

[0022] The present invention can quantitatively evaluate the degree of accumulation of silt in the pipeline by calculating the coverage rate of the residual silt space. The higher the coverage rate, the more serious the accumulation of residual silt in the pipeline, which helps to evaluate the workload that needs to be cleaned. Through this calculation, the uniformity of silt distribution and the area of ​​accumulation in the drainage pipeline can be judged to provide a basis for subsequent dredging work. Areas with high coverage rate need to be cleaned intensively. According to the analysis of the residual silt space coverage rate, it is evaluated which areas have a higher risk of blockage. Excessive accumulation of silt causes pipeline blockage and affects the smooth flow of water. This step helps to identify the risk of blockage early and avoid pipeline function failure by quantitatively analyzing the accumulation of silt. After identifying the high-risk area, repair measures can be taken in advance to reduce the probability of pipeline blockage, which can not only extend the service life of the pipeline, but also reduce the maintenance cost caused by blockage. By analyzing the structural damage problems existing in the residual silt panoramic rendering model, the overall structural health of the pipeline is evaluated, which includes analysis of damage such as cracks, corrosion, and deformation of the pipeline, helping repair personnel identify the vulnerable points of the pipeline. The structural damage risk assessment helps predict problems that may occur in the pipeline in the future, such as pipeline rupture and leakage, thereby providing Take repair and preventive measures before and after the desilting. By evaluating the risk data, the repair team can develop personalized repair plans for specific pipeline damage problems, thereby reducing repair costs and improving repair efficiency. The uncleaned original image is used as the baseline data of the drainage pipeline, which can help compare the changes before and after desilting and evaluate the cleaning effect. The acquisition of the original image provides comparative data for the subsequent quantitative evaluation of repair. By comprehensively analyzing the original image, pipeline structure damage risk data and residual silt blockage risk, a desilting and repair quantitative evaluation report is generated. The report not only provides specific requirements for silt cleaning, but also evaluates the risk points and priorities in the repair process. The quantitative evaluation report provides a clear work guide for the repair team, helping the team understand the overall situation of the cleaning work, the urgency of pipeline repair and the main problems that need to be solved, avoiding blind decision-making. Comprehensive analysis helps determine the priority of repair. For example, areas with high risk of blockage need to be treated first, while areas with serious structural damage need to be repaired or reinforced locally. The quantitative evaluation report provides actual effect feedback for the repair work. The subsequent evaluation of the repair effect is compared with the quantitative report to verify the effectiveness of the repair plan.

[0023] In this specification, a drainage pipe desilting and repair assessment device is provided, which is used to perform the drainage pipe desilting and repair assessment method as described above, comprising: A panoramic model module is used to obtain multi-directional images of the interior of the drainage pipe after desilting and the ultrasonic detection signal of the pipe; perform visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional images of the interior of the drainage pipe to construct a three-dimensional panoramic model of the pipe; A sludge identification module is used to perform deep residual sludge identification on the multi-directional images inside the drainage pipe and perform sludge spatial distribution analysis to generate residual sludge spatial distribution data; A sludge morphology calculation module is used to optimize the reflection deviation compensation of the pipeline ultrasonic detection signal and calculate the sludge morphology one by one, so as to generate the morphological characteristics of each residual sludge; A sludge component characteristic module is used to analyze the sludge component characteristics according to the pipeline ultrasonic detection signal, so as to obtain the component characteristics of each residual sludge; The panoramic rendering module is used to render the dynamic silt morphology change of the three-dimensional pipeline panoramic model according to the composition characteristics of each residual silt, the morphological characteristics of each residual silt and the spatial distribution data of the residual silt, so as to construct a panoramic rendering model of residual silt in the drainage pipeline; The quantitative assessment module is used to calculate the residual siltation spatial coverage rate of the panoramic rendering model of the drainage pipe residual siltation, and conduct a comprehensive quantitative analysis of silt removal and restoration to obtain a quantitative assessment report on silt removal and restoration.

[0024] The present invention provides a complete picture of the internal structure of the pipeline and the accumulation of silt through the combination of panoramic images and ultrasonic signals, laying the foundation for subsequent analysis and repair work. By analyzing the internal geometric structure of the pipeline, the physical structure of the pipeline can be accurately understood to ensure the accuracy of the model and avoid structural misjudgments during the repair process. By analyzing the spatial distribution of silt, it is possible to assess which areas have more serious silt accumulation, and then conduct targeted cleaning to save cleaning costs. The data provided helps the repair team make more scientific cleaning decisions and predict the accumulation pattern and change trend of silt. The ultrasonic detection signal is used to compensate for the reflection deviation of the silt to ensure accurate silt morphological characteristics, which helps to accurately assess the accumulation degree and morphology of the silt. According to the morphological characteristics of the silt, the repair team selects appropriate tools and methods for cleaning to avoid damage to the pipeline structure. The damage of the structure is analyzed by analyzing the composition characteristics of the silt (such as organic matter, inorganic matter, sediment, etc.), evaluating the physical and chemical properties of the silt, helping to formulate a more targeted cleaning plan. After understanding the composition of the silt, the appropriate cleaning method is selected to reduce pollution and unnecessary losses during the repair process. Through dynamic rendering, the repair team can intuitively see the changes in the morphology of the silt, and understand the dredging progress and silt changes inside the pipeline in real time. The effect of dynamic rendering makes the repair work more visible, helping the repair personnel to understand the repair status of each area and adjust the cleaning strategy in time. The dredging and repair quantitative evaluation report not only provides specific data on silt accumulation, but also evaluates the efficiency and effect of the cleaning process, helping the repair team to adjust the cleaning plan. Through quantitative analysis, the repair team can optimize resource allocation and concentrate on solving areas with serious problems, thereby improving repair efficiency.

[0025] In this specification, a dredging and repair robot is also provided, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the drainage pipe dredging and repair evaluation method described in any one of the above items are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of the steps of a drainage pipe desilting and repair evaluation method according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0027] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0028] The present application example provides a method, device and dredging and repair robot. The execution subjects of the drainage pipe dredging and repair assessment method, device and dredging and repair robot include but are not limited to the following: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0029] See also Figures 1 to 4 The present invention provides a drainage pipe desilting and repair assessment method, the drainage pipe desilting and repair assessment method comprising the following steps: Step S1: obtaining a multi-directional image of the interior of the drainage pipe after desilting and a pipe ultrasonic detection signal; performing visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional image of the interior of the drainage pipe to construct a three-dimensional pipe panoramic model; Step S2: performing deep residual sludge identification on the multi-directional image inside the drainage pipe and performing sludge spatial distribution analysis to generate residual sludge spatial distribution data; Step S3: optimizing the reflection deviation compensation of the pipeline ultrasonic detection signal, and calculating the sludge morphology one by one, so as to generate the morphological characteristics of each residual sludge; Step S4: Analyze the sludge composition characteristics according to the pipeline ultrasonic detection signal, so as to obtain the composition characteristics of each residual sludge; Step S5: Rendering the three-dimensional pipeline panoramic model dynamically based on the composition characteristics of each residual sludge location, the morphological characteristics of each residual sludge location, and the residual sludge spatial distribution data, thereby constructing a drainage pipeline residual sludge panoramic rendering model; Step S6: Calculate the residual siltation spatial coverage rate of the panoramic rendering model of the drainage pipe residual siltation, and conduct a comprehensive quantitative analysis of silt removal and restoration to obtain a quantitative assessment report on silt removal and restoration.

[0030] The present invention can fully understand the current situation inside the pipeline by acquiring multi-directional images and ultrasonic signals. The images provide visual information of the pipeline, while the ultrasonic detection signals help to reveal the material status and potential problems of the pipeline. The multi-directional images help to form a geometric structure model of the pipeline, support accurate three-dimensional modeling, and provide a data basis for subsequent analysis. The pipeline ultrasonic detection signal can be used to detect thickness changes, damage, cracks, etc. of the pipeline, provide real data after dredging treatment, and help evaluate the dredging effect. Through image analysis, the residual sludge that still exists after dredging treatment can be accurately identified, providing accurate information for subsequent dredging work, and the distribution of residual sludge can be accurately analyzed, thereby helping to determine which areas need to be cleaned in focus to avoid duplication of labor. According to the depth and distribution of the sludge, the subsequent dredging plan is optimized to improve work efficiency and effect. The reflection deviation compensation optimization can effectively eliminate errors caused by detection equipment or environmental factors, thereby improving the accuracy of ultrasonic signals. By calculating the morphological characteristics of the sludge and understanding the structure of the residual sludge, dredging personnel can be more targeted in treatment and avoid unnecessary intervention, which helps to identify and locate sludge of special forms, especially Especially for the silt blocks that are difficult to remove, provide guidance for repair. The composition analysis of silt can reveal its source and properties, such as organic matter content, inorganic matter content, etc., to provide a basis for subsequent treatment methods. Silt with different components requires different cleaning methods. Composition characteristic analysis helps to formulate more accurate cleaning plans and improve the repair effect. Through composition analysis, the effect after cleaning can be predicted, and whether special treatment or secondary cleaning is necessary can be evaluated. Dynamic rendering can intuitively display the morphological changes of silt inside the pipeline, helping relevant personnel understand the distribution and properties of silt. The rendering model provides managers with intuitive data support to help them make more reasonable decisions and adjust cleaning strategies in time. Through the three-dimensional panoramic model, dredging work becomes more transparent, and relevant personnel have a more comprehensive understanding of the internal situation of the pipeline. By calculating the spatial coverage rate of residual silt, the cleaning effect can be accurately evaluated to help confirm the completion of dredging. Comprehensive quantitative analysis provides a clear evaluation report for dredging and repair work, helps to evaluate the effect of the entire cleaning process, discover potential problems, and provide data support for future maintenance. The quantitative evaluation report can give the team clear goals, avoid repeated cleaning and omissions, and significantly improve work efficiency.

[0031] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a drainage pipe desilting and repair assessment method of the present invention. In this example, the steps of the drainage pipe desilting and repair assessment method include: Step S1: obtaining a multi-directional image of the interior of the drainage pipe after desilting and a pipe ultrasonic detection signal; performing visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional image of the interior of the drainage pipe to construct a three-dimensional pipe panoramic model; In this embodiment, a high-definition camera or a camera mounted on a drone is used to ensure that it has good image clarity and shooting capabilities in low-light environments. After dredging, the inside of the drainage pipe is photographed from multiple angles and heights to ensure that all pipe areas are covered. A 360-degree camera or multiple fixed cameras can be used for synchronous shooting. Ultrasonic detection equipment is installed inside the pipe to ensure that the equipment can cover the entire length of the pipe and can accurately receive the reflected signal. Start the equipment, obtain ultrasonic detection signal data, and record the time domain and amplitude information of the signal. The collected multi-directional images are denoised, contrast enhanced, and standardized to improve the accuracy of subsequent recognition. Computer vision technology (such as edge detection and contour extraction) is used to extract features inside the pipe, including pipe edges, joints, valves and other components. A deep learning model (such as a convolutional neural network, CNN) is trained to automatically identify pipe components. The model is built and trained using an open source framework (such as TensorFlow or PyTorch). The trained model is used to classify and annotate the pipe components in the image, and the category and position coordinates of the components are output. Based on the identified pipeline components and their positions in the image, 3D reconstruction algorithms (such as structured light, stereo vision, or multi-view geometry) are used to convert 2D image data into 3D point cloud data. The generated point cloud data is processed, including denoising, downsampling, and surface reconstruction, to enhance the accuracy and visualization of the model. The geometric relationships between the components in the 3D model, such as distance, angle, and relative position, are analyzed. This is achieved through computational geometry methods (such as Euclidean distance and angle calculation). The geometric structural integrity of the pipeline interior is evaluated to identify existing defects or anomalies. A complete 3D pipeline panoramic model is generated using the processed 3D point cloud data. Modeling is performed using 3D modeling software (such as Blender, SketchUp, or AutoCAD). The acquired image data is mapped onto the 3D model surface to increase the realism and visualization of the model. The 3D model is enhanced with details to ensure that the shape, size, and position of the pipeline components are consistent with the actual situation. The accuracy and integrity of the model are verified by comparing the original image with the generated 3D model.

[0032] Step S2: performing deep residual sludge identification on the multi-directional image inside the drainage pipe and performing sludge spatial distribution analysis to generate residual sludge spatial distribution data; In this embodiment, the multi-directional image is denoised, and a filter (such as Gaussian filtering or median filtering) is used to reduce the noise in the image to improve the accuracy of sludge identification. Contrast enhancement and histogram equalization techniques are applied to improve the contrast of the image and make the residual sludge more obvious. An image data set containing sludge is collected and annotated to ensure sample diversity and representativeness. The annotation content includes the shape, position and area of ​​the sludge. Select a suitable deep learning model (such as U-Net or MaskR-CNN) for image segmentation and sludge identification. The model is trained using the annotated data set and hyperparameters are optimized to improve the performance of the model. The accuracy of the model is evaluated using a cross-validation method to ensure the generalization ability of the model on unseen data. The trained model is applied to the multi-directional image to automatically segment the sludge. The model will output the category label of each pixel and identify the sludge area. The segmentation results are post-processed to remove small noise areas and smooth the sludge boundaries to improve the accuracy of the recognition results. The boundary and area information of the sludge area is extracted from the segmentation results to generate the contour data of the sludge. The identified sludge area is mapped to the 3D pipeline panoramic model, and the spatial position of the sludge inside the pipeline is determined by combining the previous 3D coordinate data. The spatial distribution of the sludge area is statistically analyzed, including parameters such as the sludge distribution density, volume, thickness and shape. The calculation is performed using geometric analysis methods. A heat map of the sludge spatial distribution is generated to visualize the distribution of sludge inside the pipeline. The heat map is achieved by mapping the sludge distribution data to a color gradient. The generated residual sludge spatial distribution data (including sludge location, area, volume and heat map) is recorded in the database for subsequent analysis and query.

[0033] Step S3: optimizing the reflection deviation compensation of the pipeline ultrasonic detection signal, and calculating the sludge morphology one by one, so as to generate the morphological characteristics of each residual sludge; In this embodiment, the original signal data is obtained from the ultrasonic detection equipment, including the amplitude and time information of the reflected signal. Digital filtering technology (such as a low-pass filter) is applied to remove high-frequency noise in the signal to improve the quality of the signal. The reflection characteristics of the ultrasonic signal are analyzed and a reflection deviation model is established. The source of the deviation is analyzed based on historical data using linear regression or nonlinear fitting methods. The original signal is corrected by the compensation algorithm. Scompensated (t) = Soriginal (t) - D (t), where D (t) is the deviation signal. By comparing the signal waveforms before and after compensation, the compensation effect is evaluated to ensure that the reflection characteristics of the signal are improved. The peak detection algorithm (such as the find_peaks function of SciPy) is used to identify the reflected peaks in the compensated signal and extract the time and amplitude of each peak. The propagation time of the ultrasonic wave from emission to reflection is determined by calculating the time difference between the reflected peaks. The morphological characteristics of the sludge are defined, including parameters such as thickness, volume, density, and shape. Thickness = 1 / 2 × c × t, where c is the propagation speed of the ultrasonic wave in the medium and t is the reflection time. The volume of each sludge is calculated based on the thickness of the sludge and the corresponding cross-sectional area. The geometric formula is used for estimation. The calculated morphological characteristics of each residual silt are recorded in the database, including thickness, volume, density and shape, for subsequent analysis.

[0034] Step S4: Analyze the sludge composition characteristics according to the pipeline ultrasonic detection signal, so as to obtain the composition characteristics of each residual sludge; In this embodiment, the compensated signal data including the amplitude, frequency and time information of the reflected signal are extracted from the ultrasonic detection equipment. The signal is analyzed in the time domain to identify the main features of the signal, such as waveform, peak value and attenuation characteristics. Fast Fourier transform (FFT) is applied to convert the time domain signal into a frequency domain signal to analyze different frequency components. It is implemented using NumPy library or MATLAB. The main peaks in the spectrum are analyzed to extract the amplitude and frequency information of each frequency component. These spectrum characteristics are closely related to the composition characteristics of the sludge. The composition characteristic parameters of the sludge are defined, such as particle size, viscosity, density and organic matter content. These parameters usually affect the propagation characteristics of the ultrasonic wave, which are then reflected in the signal. Linear regression or multiple regression analysis is used to establish a relationship model between the sludge composition characteristics and the ultrasonic signal characteristics. The extracted spectral features are used as independent variables, and the sludge composition characteristics are used as dependent variables for modeling. The model is trained using existing sludge sample data to ensure the accuracy and generalization ability of the model. The cross-validation method is used for evaluation. The newly acquired ultrasonic signal features are input into the trained component feature model to infer the component characteristics. The model outputs the composition characteristics of each residual sludge, including particle size, viscosity, density, etc. The inferred sludge composition characteristics are recorded in the database to ensure data integrity and traceability.

[0035] Step S5: Rendering the three-dimensional pipeline panoramic model dynamically based on the composition characteristics of each residual sludge location, the morphological characteristics of each residual sludge location, and the residual sludge spatial distribution data, thereby constructing a drainage pipeline residual sludge panoramic rendering model; In this embodiment, the component characteristic data of each residual sludge is obtained from the previous steps, including particle size, density, viscosity, etc. The morphological characteristic data of each sludge is obtained, such as thickness, volume and shape. The spatial position and distribution information of the sludge in the three-dimensional pipeline model are collected. All data are integrated into a unified format for subsequent processing. The data is stored in JSON or CSV format to ensure that the information of each sludge is clear. Use three-dimensional modeling software (such as Blender, Maya or Unity) to import the previously constructed three-dimensional pipeline panoramic model. Set the corresponding material and texture according to the component characteristics of each sludge. For example, different colors and transparency are used to represent sludge of different viscosities and particle sizes. Adjust the size and shape of the sludge model according to the morphological characteristics of the sludge (such as thickness and volume) to ensure that the rendered sludge morphology is consistent with the actual situation. Use animation tools to set the dynamic change effect of the sludge. Through keyframe animation, define the change of the shape, volume and position of the sludge over time. Use a physical engine (such as Unity's Rigidbody or Blender's physical simulation) to simulate the flow and deposition process of sludge in the pipeline. Ensure that the dynamic behavior of sludge in the pipeline is more realistic. Dynamically adjust the color, transparency and flow speed of sludge according to its composition and morphological characteristics. For example, sludge with high viscosity flows slowly, while sludge with low viscosity flows faster. Set rendering parameters as needed, including resolution, lighting effects and shadow processing. Ensure that the rendering effect achieves the expected effect. If using a real-time rendering engine (such as Unity), export the dynamic sludge rendering model as an interactive 3D application.

[0036] Step S6: Calculate the residual siltation spatial coverage rate of the panoramic rendering model of the drainage pipe residual siltation, and conduct a comprehensive quantitative analysis of silt removal and restoration to obtain a quantitative assessment report on silt removal and restoration.

[0037] In this embodiment, the panoramic rendering model of the residual siltation in the drainage pipe is imported into a 3D modeling software (such as Blender or MATLAB), and the residual siltation area in the model is identified by using a graphics processing tool or a selection tool in the software. The siltation and clean areas are distinguished by setting a threshold or using a color mapping technique, and the volume of the identified residual siltation area is calculated. The volume of the residual siltation area is calculated using a voxel method (VoxelMethod) or a geometric volume formula. The total available volume of the pipe is also calculated. Pipeline volume = π × r 2×h, where r is the radius of the pipeline, h is the length of the pipeline, and the spatial coverage rate = (residual silt volume / available pipeline volume) × 100%. According to the residual silt spatial coverage rate and silt characteristics, select appropriate dredging methods (such as mechanical dredging, high-pressure water dredging or chemical dredging), evaluate the resources required for dredging, including manpower, equipment and time, make estimates based on the data of historical dredging projects, identify risk factors that affect the dredging process, such as weather, pipeline structure damage, etc., and evaluate the risk level of the dredging project based on the nature and impact of the risk factors. Record basic project information in the report, including project name, location, pipeline specifications, etc., describe in detail the residual silt spatial coverage rate, silt characteristics and the selected dredging method, list the estimated cost and time required for dredging, and provide risk assessment results and recommended countermeasures.

[0038] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: obtaining a multi-directional image of the interior of the drainage pipe after desilting and a pipe ultrasonic detection signal; Step S12: visually identifying pipe components on the multi-directional image of the interior of the drainage pipe, and marking a plurality of pipe internal components; Step S13: performing pipeline topological connection analysis according to multiple pipeline internal components, thereby generating a pipeline topological connection relationship; Step S14: performing internal geometric structure arrangement analysis on the multi-directional images of the interior of the drainage pipe to generate internal geometric structure data of the pipe; Step S15: Perform three-dimensional point cloud modeling based on the internal geometric structure data of the pipeline and the pipeline topological connection relationship to construct a three-dimensional pipeline panoramic model; In this embodiment, a high-definition industrial camera and a 360-degree panoramic camera are used to collect multi-directional images inside the drainage pipe. According to the diameter and length of the pipe, the camera position is reasonably set to ensure that every corner of the pipe is covered. After desilting, an ultrasonic detection instrument is used to detect the inside of the pipe. Ultrasonic sensors should be evenly distributed on the inner wall of the pipe to obtain echo signals at different positions to determine the thickness, cracks and other defects of the pipe. The acquired multi-directional images are stored in a secure network attached storage (NAS), and an efficient image compression format (such as JPEG2000) is used to save storage space. The ultrasonic detection signal is stored in a digital format to ensure the integrity and traceability of the data. The collected multi-directional images are pre-processed, including using a Gaussian filter to remove noise and enhance contrast to improve the accuracy of subsequent recognition. Image segmentation techniques (such as color-based segmentation or edge detection) are used to separate different components inside the pipe (such as valves, joints, and pipe sections) from the background. Machine learning model training: Use a deep learning algorithm (such as a convolutional neural network CNN) for component recognition. First, collect a labeled training data set and train the model to recognize different types of pipeline components. Use the trained model to process the multi-directional images, mark the identified pipeline components, and record their positions and types. Integrate the identified pipeline component data to form a database containing the attributes of each component (such as type, position, and connection relationship). Use graph theory to establish a connection relationship diagram between pipeline components. Each component is regarded as a node in the graph, and the connection is regarded as an edge. Define the connection rules between components (such as the connection between valves and pipes), and determine the connection relationship between components based on the geometric position of the components and the recognition results. Generate a topological connection relationship diagram of the pipeline and output it in a standard format (such as GraphML or GML) for subsequent analysis and visualization. Extract geometric features from the multi-directional images, including the diameter, length, bending radius, etc. of the pipeline. Use computer vision technology (such as Hough transform) to detect the geometric shape of the pipeline. Use the geometric arrangement algorithm to analyze the relative position and arrangement of the internal components of the pipeline and identify the geometric structural features of the pipeline. Save the extracted geometric structural data (such as coordinates, length, angle, etc.) in a standardized format (such as CSV or JSON) for subsequent processing. Generate 3D point cloud data of the pipeline from multi-directional images using laser scanners or 3D reconstruction technology. Obtain high-precision 3D point clouds by matching images from different angles and combining geometric data. Integrate ultrasonic testing data with point cloud data to ensure the accuracy of the model. Apply 3D reconstruction algorithms (such as Poisson Surface Reconstruction) to convert point cloud data into a 3D model and generate a panoramic model of the pipeline. Optimize the generated 3D model to reduce the number of redundant faces and improve rendering efficiency. Use 3D modeling software (such as Blender or MeshLab) for subsequent processing.Import the constructed 3D pipeline panoramic model into the visualization platform, provide interactive viewing functions, and facilitate subsequent analysis and management. Archive the final 3D model and related data for future query and use.

[0039] The three-dimensional point cloud modeling is specifically as follows: Extracting an image in each orientation according to the multi-orientation images inside the drainage pipe; Identify key component points of each orientation image and mark multiple key component points; Perform multi-angle precise registration of images based on multiple key component points to generate multi-angle registration data; Based on the multi-angle registration data, multi-angle stitching processing is performed on the multi-directional images inside the drainage pipe, so as to generate a panoramic image of the inside of the pipe; According to the internal geometric structure data of the pipeline and the pipeline topological connection relationship, the three-dimensional point cloud modeling of the panoramic image inside the pipeline is carried out to construct a three-dimensional pipeline panoramic model.

[0040] In this embodiment, the original image data is obtained from the multi-directional image set inside the drainage pipe. Ensure that the images in each direction are effectively captured and the image quality is high enough for subsequent processing. Use image processing tools (such as OpenCV) to segment the original image. By setting a threshold or using an edge detection algorithm (such as the Canny algorithm), the image area of ​​each orientation is identified and the part to be analyzed is extracted. Use a feature extraction algorithm (such as SIFT, SURF or ORB) to detect key points for each orientation image. These algorithms can identify points with unique features in the image, which are suitable for subsequent registration and stitching. Filter out key points with high repeatability and stability to ensure the matching accuracy between different images. Use the RANSAC algorithm to remove outliers and improve the accuracy of key points. The coordinates and related features of the identified key component points are stored in a database for subsequent registration processing. The key points in the images of different orientations are matched by descriptors (such as BRIEF or FREAK) to calculate the relationship between the corresponding points. The relative transformation between images is achieved using matched key point pairs. Use the least squares method or the RANSAC algorithm to estimate the transformation matrix between images. This process will take into account factors such as rotation, translation, and scaling. Based on the registration data, the images in each orientation are transformed. The registration matrix is ​​used to map the images in different orientations to the same coordinate system. The transformed images are merged into a panoramic image using an image stitching algorithm (such as feathering stitching or multiple exposure technology). Ensure that the transition of the stitching area is natural and avoid obvious seams. The stitched panoramic images are post-processed, including denoising, contrast enhancement, and color correction, to improve image quality. The generated panoramic image of the pipeline interior is saved in a high-resolution standard format for subsequent use and display. Combine the previously acquired internal geometric structure data of the pipeline (such as diameter, length, bending radius, etc.) with the panoramic image to construct a 3D spatial model. Use laser scanning or optical 3D reconstruction technology to extract point cloud data from the panoramic image. Each pixel corresponds to a point in 3D space, and a dense point cloud is generated through depth information. Use 3D reconstruction algorithms (such as PoissonSurfaceReconstruction or MarchingCubes) to convert the point cloud data into a mesh model to construct a 3D panoramic model of the pipeline. According to the pipeline topology connection relationship, each component is reasonably connected to form a complete pipeline model, ensuring that its geometric and topological characteristics are consistent. Use 3D visualization tools (such as Blender or Unity) to display the constructed 3D pipeline panoramic model and provide interactive viewing functions. Archive the 3D model and related data for future query and analysis.

[0041] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing adaptive brightness enhancement processing on the multi-directional image inside the drainage pipe to obtain an adaptive brightness enhanced image; Step S22: performing super-resolution texture detail optimization on the adaptive brightness enhanced image to construct a detail optimized pipeline image; Step S23: performing deep residual sludge recognition on the detail optimized pipeline image, marking each location of residual sludge in the pipeline; Step S24: performing three-dimensional positioning of each residual sludge in the pipeline space to generate three-dimensional coordinates of each residual sludge; Step S25: Performing sludge spatial distribution analysis on each three-dimensional coordinate of residual sludge, thereby generating residual sludge spatial distribution data.

[0042] In this embodiment, an image that needs to be enhanced in brightness is selected from the multi-directional images of the drainage pipe obtained in the early stage. The brightness distribution of the image is analyzed by using an image processing tool (such as OpenCV), the brightness of the current image is evaluated, and the area with insufficient brightness is identified. The adaptive histogram equalization (AHE) or CLAHE (Contrast Limited Adaptive Histogram Equalization) algorithm is used to enhance the local contrast of the image. These methods enhance the details of the local area by equalizing the image blocks, while preventing the noise amplification caused by excessive enhancement. After the processing is completed, the adaptive brightness enhanced image is saved for subsequent detail optimization and analysis. The adaptive brightness enhanced image is processed using super-resolution reconstruction technology (such as SRCNN or GAN-based method). These methods generate clearer and more detailed images by learning the mapping relationship between low-resolution images and high-resolution images. During the super-resolution processing, special attention is paid to the texture details and features inside the pipeline to ensure that the generated image has higher visual clarity and details. The pipeline image with optimized details is processed using image segmentation technology (such as threshold-based segmentation, region growing or image segmentation network). By analyzing the texture features of the image, the residual sludge area is identified. If there is enough annotated data, the convolutional neural network (CNN) is trained for semantic segmentation to automatically identify and mark the sludge area in the pipeline image. The two-dimensional coordinates of the identified pipeline sludge area are converted into three-dimensional coordinates using the internal geometric parameters of the pipeline and the known internal and external parameters of the camera. Use triangulation or structured light method for spatial positioning. If there is a depth camera, use the depth image to directly obtain the three-dimensional coordinates of each sludge point. Store the three-dimensional coordinates (X, Y, Z) of each sludge point in the database to ensure the accuracy and traceability of the data. Perform statistical analysis on the recorded three-dimensional coordinates of residual sludge to calculate the sludge density, volume and distribution characteristics of each area. Use spatial analysis tools (such as GIS software) for visualization and analysis. Use machine learning methods (such as clustering algorithms) to model the spatial distribution of sludge and identify the laws and trends of sludge distribution. Summarize the analysis results and generate a sludge spatial distribution data report, including statistical charts and visualization results, to facilitate subsequent decision-making and management. Archive the generated spatial distribution data for subsequent use and query.

[0043] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: performing high-frequency filtering on the pipeline ultrasonic detection signal to obtain a filtered and noise-reduced ultrasonic signal; Step S32: calculating the pipeline reflection deviation of the filtered and denoised ultrasonic signal to generate a pipeline reflection deviation parameter; Step S33: performing reflection deviation compensation optimization on the pipeline reflection deviation parameters to generate deviation compensation ultrasonic parameters; Step S34: Calculate the sludge morphology of each residual sludge location one by one based on the deviation-compensated ultrasonic parameters based on the internal geometric structure data of the pipeline, thereby generating the morphological characteristics of each residual sludge location.

[0044] In this embodiment, raw ultrasonic signals are obtained from an ultrasonic detector. These signals contain noise and interference, which affect the accuracy of subsequent analysis. First, the signal spectrum is analyzed to identify the frequency range of high-frequency noise, which usually includes electromagnetic interference and noise generated by the device itself. A suitable high-pass filter (such as a Butterworth or Chebyshev filter) is designed to filter out low-frequency noise and retain only useful high-frequency signals. The designed filter is applied to the ultrasonic signal, and the signal is filtered to remove high-frequency noise to obtain a filtered and denoised ultrasonic signal. The filtering effect is checked by comparing the signal waveforms before and after filtering to ensure signal clarity and effectiveness. The reflected wave features, including reflection time, amplitude, and frequency, are extracted from the filtered and denoised ultrasonic signal. A signal processing algorithm (such as peak detection) is used to identify the start and end positions of the reflected wave. The time delay of the signal reflection is calculated based on the speed of ultrasonic wave propagation in the pipeline, thereby deducing the actual position of the inner wall of the pipeline. The actually measured reflection time is compared with the theoretical value (based on the numerical value of the geometric structure of the pipeline). The reflection deviation is calculated by the formula: reflection deviation = actual reflection time - theoretical reflection time. A parameter list containing the deviation value of each reflection point is generated to facilitate subsequent compensation optimization processing. According to the reflection deviation parameters, a compensation model is designed. The relationship between the deviation and the actual pipeline state is established by using linear regression or polynomial fitting method. The compensation algorithm is applied to correct the original ultrasonic signal to generate compensated ultrasonic parameters. This process usually includes adding and subtracting the deviation of each reflection point to adjust the signal. A mathematical model between the position and morphology of the sludge is established by combining the geometric structure data of the pipeline (such as diameter, length, curvature, etc.) with the deviation-compensated ultrasonic parameters. A computational geometry algorithm is used to calculate the morphological characteristics of the sludge one by one based on the compensated ultrasonic data, including thickness, width, volume, etc. Voxel processing technology is used to simulate the three-dimensional morphology of the sludge in the pipeline. The morphological characteristic data of each residual sludge, including various parameters (such as shape, volume, distribution), are generated and stored in the database for subsequent analysis and management.

[0045] In this embodiment, the specific steps of step S34 are: Performing acoustic wave reflection waveform analysis on deviation-compensated ultrasonic parameters to generate pipeline acoustic wave reflection waveform curve; Calculate the time difference between adjacent peaks of the pipeline sound wave reflection waveform curve to obtain the propagation time of the reflection signal; According to the pipeline sound wave reflection waveform curve, the reflection signal amplitude change is analyzed to generate the reflection signal amplitude change characteristics; The distance between the sediment and the pipe wall is calculated based on the internal geometric structure data of the pipe and the propagation time of the reflected signal, thereby obtaining the sludge thickness parameter; The size of the silt area is estimated based on the change characteristics of the reflected signal amplitude to generate the size of the silt area; The silt morphology is calculated one by one according to the silt thickness parameters and the silt area size, so as to generate the morphological characteristics of each residual silt.

[0046] In this embodiment, the reflected signal data is extracted from the compensated ultrasonic parameters to ensure the integrity and accuracy of the data. The extracted reflected signal data is plotted as a waveform curve using signal processing software (such as Matplotlib in MATLAB or Python). By setting appropriate time axis and amplitude axis, the changes in the ultrasonic reflection waveform are clearly displayed. The main reflected wave peaks are marked in the waveform graph for subsequent analysis. By analyzing the waveform graph, the main characteristics and change trends of the reflected signal are identified. The peak detection algorithm (such as the find_peaks function in SciPy) is used to identify adjacent peaks in the waveform curve, and the time and amplitude of each peak are extracted. The time difference between adjacent peaks is calculated to obtain the propagation time of the reflected signal. This process is implemented by simple subtraction, and the calculated reflected signal propagation time is recorded in the database for subsequent analysis and use. The amplitude of each peak is extracted from the sound wave reflection waveform curve to form an amplitude data sequence. The change trend of the amplitude data is analyzed, including statistical features such as the maximum, minimum, average and standard deviation of the amplitude. These features are calculated using data analysis tools (such as Pandas). According to the propagation speed of ultrasound in the pipeline (usually about 1480m / s in water) and the propagation time, the distance between the silt and the pipe wall is calculated. Silt thickness = (propagation speed × propagation time) / 2. The relationship model between the amplitude of the reflected signal and the silt area is established based on experimental data or historical data. Generally, the larger the amplitude, the thicker or larger the silt layer. The area of ​​each silt is inferred using the amplitude change characteristics. This is achieved through a linear regression model or a nonlinear fitting method. A three-dimensional morphological model of the silt is established based on the silt thickness and area data. This is achieved through a geometric model construction method, where the silt is usually regarded as a uniform layer and its shape is derived using thickness and area. The morphological characteristics of the silt, such as volume, surface area, shape factor, etc., are extracted from the model to form complete silt morphological characteristic data. The morphological characteristic data of each residual silt are sorted and stored for subsequent analysis, visualization, and decision support.

[0047] In this embodiment, step S4 includes the following steps: Step S41: Performing Fourier transform on the deviation-compensated ultrasonic parameters to obtain multiple segments of ultrasonic spectrum data; Step S42: performing spectrum bandwidth analysis on the multiple segments of ultrasonic spectrum data segment by segment to generate spectrum bandwidth characteristics of each segment; Step S43: sludge particle estimation is performed based on the spectrum bandwidth characteristics of each segment to obtain particle characteristics in the sludge; Step S44: performing spectrum peak analysis on the multiple segments of ultrasonic spectrum data to obtain the spectrum peak of each segment; Step S45: mining the sludge viscosity based on the spectrum peak of each section to obtain the sludge viscosity characteristics; Step S46: Calculate the resonance frequency of the multiple ultrasonic spectrum data to obtain the degree of sludge consolidation; Step S46: Analyze the sludge composition characteristics according to the sludge consolidation degree, sludge viscosity characteristics and particle characteristics in the sludge, so as to obtain the composition characteristics of each residual sludge.

[0048] In this embodiment, the time domain signal is extracted from the deviation compensation ultrasonic parameter to ensure that the signal is continuous and has no obvious missing. The time domain signal is converted into a frequency domain signal using a fast Fourier transform (FFT) algorithm. The calculation is performed using the NumPy library of Python or the FFT function in MATLAB. The obtained spectrum data includes frequency components and their corresponding amplitude information to form multiple ultrasonic spectrums. For each segment of spectrum data, its frequency range is determined, including the lowest frequency and the highest frequency. The bandwidth of each segment is calculated, bandwidth = f high - f low, where f high and f low are the highest and lowest frequencies of the spectrum, respectively. According to existing research or experimental data, a relationship model between spectrum bandwidth and sludge particle size is established. Generally, the larger the bandwidth, the wider the particle size distribution. The particle characteristics in the sludge are inferred using the spectrum bandwidth characteristics, such as the average diameter and distribution range of the particles, and the particle characteristics are obtained using statistical analysis methods (such as mean and standard deviation calculations). Each segment of spectrum data is analyzed using a peak detection algorithm (such as the find_peaks function of SciPy) to identify the main peaks in the spectrum. The frequency and amplitude of each peak are extracted, and the main features of each spectrum are recorded. According to existing literature or experimental data, a relationship model between the spectrum peak and the viscosity of the silt is established. Generally, the change in the spectrum peak is positively correlated with the viscosity of the silt. Use the spectrum peak data to infer the viscosity characteristics of the silt. This is achieved through linear regression or nonlinear fitting methods. The resonant frequency is determined as the frequency point with the largest amplitude in the spectrum. Generally, there is a certain correlation between the degree of consolidation of the silt and the resonant frequency. The resonant frequency is extracted from each section of the spectrum data, and the frequency value with the highest amplitude is usually selected as the resonant frequency. The degree of consolidation of the silt, the viscosity characteristics and the particle characteristics are integrated to form a comprehensive feature data set. Use multiple regression analysis or machine learning algorithms (such as random forests or support vector machines) to establish a silt component feature analysis model to infer the component characteristics of each residual silt.

[0049] In this embodiment, step S5 includes the following steps: Step S51: performing sludge position matching on the three-dimensional pipeline panoramic model based on the three-dimensional coordinates of each residual sludge location to obtain sludge position matching data; Step S52: Visually marking the three-dimensional pipeline panoramic model with the residual sludge spatial distribution data according to the sludge position matching data, thereby obtaining a residual sludge distribution visualization model; Step S53: Based on the morphological characteristics of each residual sludge, the residual sludge distribution visualization model is dynamically rendered according to the morphological characteristics of each residual sludge, and attribute mapping is performed according to the component characteristics of each residual sludge, so as to construct a panoramic rendering model of residual sludge in the drainage pipeline.

[0050] In this embodiment, the three-dimensional coordinate data of each residual sludge is extracted to ensure that the data format is consistent, usually in XYZ coordinate format, and the coordinate system of the pipeline model and the sludge coordinates (such as the right-hand coordinate system and the left-hand coordinate system) is checked. If they are inconsistent, coordinate conversion is required. The sludge coordinates are converted to the coordinate system of the pipeline model using linear transformation (such as translation and rotation). According to the needs, the following method is used: by calculating the distance between each sludge point and the surface point of the pipeline model, the nearest matching point is found. If the sludge data and the pipeline model are both point clouds, the ICP (Iterative Closest Point) algorithm is used for alignment, and each sludge coordinate is traversed to calculate its distance to the surface point of the pipeline model. The distance between each point on the surface of the pipeline model, record the position of the matching point corresponding to each silt coordinate on the pipeline model, select suitable visualization software, such as ParaView, Unity, or Blender, which supports the display and interaction of 3D models, import the 3D pipeline panoramic model into the visualization tool, and ensure that the scale and unit of the model are correct, load the silt position matching data, and ensure that the data format is compatible with the visualization tool, use different colors, shapes or sizes to mark the location of the silt, for example: use red to represent high-concentration silt and green to represent low-concentration silt, adjust the size of the mark according to the size of the silt volume, add marks to the pipeline model, and use the API of the visualization tool Or use scripting functions (such as Python scripts) for automated processing, perform visual checks to ensure that the markings are accurate, the sludge position is consistent with the geometric shape of the pipeline model, and compare with the actual data to confirm the authenticity of the visualization results. Select visualization tools or engines that support dynamic rendering, such as UnrealEngine, Three.js, or Unity. These tools support real-time rendering and dynamic effects. Extract the morphological characteristics (such as volume, shape) and composition characteristics (such as viscosity, particle composition) of each residual sludge from the database to ensure the format and integrity of the data. According to the morphological characteristics of the sludge, set the dynamic simulation effect, for example: according to the state change of the sludge ( The system can dynamically change the shape of the silt, use keyframe animation or physical simulation technology to show the movement and change of the silt in the pipe, write scripts in the selected rendering engine, control the realization of dynamic effects, and perform color and transparency mapping according to the composition characteristics of the silt. For example, different colors are used to represent different component ratios, and the transparency of the silt is adjusted according to the viscosity characteristics. Silt with high viscosity is displayed as opaque, while silt with low viscosity is relatively transparent. The results of dynamic rendering and attribute mapping are integrated into a complete panoramic rendering model of residual silt in the drainage pipe, and the final model is saved in a format suitable for display and analysis (such as interactive web pages, virtual reality formats, etc.) to ensure that it can be used and displayed later.

[0051] In this embodiment, step S6 includes the following steps: Step S61: calculating the residual siltation space coverage rate of the drainage pipe residual siltation panoramic rendering model to generate the residual siltation space coverage rate; Step S62: performing a pipeline blockage risk analysis on the drainage pipeline residual siltation panoramic rendering model based on the residual siltation space coverage rate to generate a pipeline residual siltation blockage risk; Step S63: Performing a structural damage risk assessment on the drainage pipe residual siltation panoramic rendering model to obtain pipe structure damage risk data; Step S64: obtaining an original image of the uncleaned drainage pipe; Step S65: Based on the original image of the drainage pipe that has not been cleaned, a comprehensive quantitative analysis of the pipeline structure damage risk data and the pipeline residual silt blockage risk is performed to obtain a quantitative assessment report on the silt removal and repair.

[0052] In this embodiment, a panoramic rendering model of residual silt in the drainage pipe is loaded from the database to ensure the accuracy and integrity of the model. A three-dimensional calculation method (such as volume integration or bounding box method) is used to calculate the volume of all residual silt in the pipe. A computer graphics library (such as Open3D or PCL) is used to extract the three-dimensional point cloud data of the silt and perform volume calculation. The total volume inside the pipe is calculated, which is usually obtained by a simple geometric formula (such as the volume formula of a cylinder). The coverage rate = residual silt volume / pipeline volume. The calculated residual silt space coverage rate is stored in the database to form the basic data for subsequent analysis and to select the appropriate risk Evaluation models, such as logistic regression models or decision tree models, are analyzed based on coverage and other relevant factors (such as pipe diameter, flow, etc.). The residual siltation space coverage, pipe geometric characteristics (such as diameter, length) and flow data are obtained from the database. The residual siltation space coverage and pipe characteristic data are input into the model to obtain the blockage risk score or probability of each pipe. The Python Scikit-learn library is used for modeling and analysis. The analysis results (pipeline blockage risk) are recorded in the database for subsequent use. The structural characteristic data of the drainage pipes, including pipe material, age, history, etc., are loaded from the database. damage records, etc., select a suitable structural damage risk assessment model, such as a method based on fuzzy logic or probability model, input the pipeline feature data into the assessment model, calculate the structural damage risk of each pipeline, and the assessment model takes into account factors such as the weight of silt and the carrying capacity of the pipeline. Obtain the original image of the drainage pipeline from the monitoring camera or drone to ensure that the image quality is good and covers all areas in the pipeline. Store the acquired image data in a database or file system to ensure that it is easily accessed and processed. Preprocess the original image, including denoising, contrast enhancement, etc., to improve the effect of subsequent analysis. Integrate the uncleaned original image, pipeline structure damage risk data and pipeline blockage risk data into an analysis framework, select a suitable comprehensive analysis model, such as multivariate regression analysis or weighted scoring model, and conduct an assessment in combination with various risk factors. Based on the integrated data, apply the selected model to conduct a comprehensive quantitative analysis of dredging and repair, evaluate the necessity and priority of cleaning and repair, calculate the required repair cost, time and potential risks, so as to provide decision support, and organize the analysis results into a quantitative assessment report on dredging and repair, including summary, analysis methods, results and recommendations, etc. Format and export the report using a report generation tool (such as LaTeX or Word).

[0053] In this embodiment, a drainage pipe desilting and repair assessment device is provided, which is used to perform the drainage pipe desilting and repair assessment method as described above, including: A panoramic model module is used to obtain multi-directional images of the interior of the drainage pipe after desilting and the ultrasonic detection signal of the pipe; perform visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional images of the interior of the drainage pipe to construct a three-dimensional panoramic model of the pipe; A sludge identification module is used to perform deep residual sludge identification on the multi-directional images inside the drainage pipe and perform sludge spatial distribution analysis to generate residual sludge spatial distribution data; A sludge morphology calculation module is used to optimize the reflection deviation compensation of the pipeline ultrasonic detection signal and calculate the sludge morphology one by one, so as to generate the morphological characteristics of each residual sludge; A sludge component characteristic module is used to analyze the sludge component characteristics according to the pipeline ultrasonic detection signal, so as to obtain the component characteristics of each residual sludge; The panoramic rendering module is used to render the dynamic silt morphology change of the three-dimensional pipeline panoramic model according to the composition characteristics of each residual silt, the morphological characteristics of each residual silt and the spatial distribution data of the residual silt, so as to construct a panoramic rendering model of residual silt in the drainage pipeline; The quantitative assessment module is used to calculate the residual siltation spatial coverage rate of the panoramic rendering model of the drainage pipe residual siltation, and conduct a comprehensive quantitative analysis of silt removal and restoration to obtain a quantitative assessment report on silt removal and restoration.

[0054] The present invention provides a complete picture of the internal structure of the pipeline and the accumulation of silt through the combination of panoramic images and ultrasonic signals, laying the foundation for subsequent analysis and repair work. By analyzing the internal geometric structure of the pipeline, the physical structure of the pipeline can be accurately understood to ensure the accuracy of the model and avoid structural misjudgments during the repair process. By analyzing the spatial distribution of silt, it is possible to assess which areas have more serious silt accumulation, and then conduct targeted cleaning to save cleaning costs. The data provided helps the repair team make more scientific cleaning decisions and predict the accumulation pattern and change trend of silt. The ultrasonic detection signal is used to compensate for the reflection deviation of the silt to ensure accurate silt morphological characteristics, which helps to accurately assess the accumulation degree and morphology of the silt. According to the morphological characteristics of the silt, the repair team selects appropriate tools and methods for cleaning to avoid damage to the pipeline structure. The damage of the structure is analyzed by analyzing the composition characteristics of the silt (such as organic matter, inorganic matter, sediment, etc.), evaluating the physical and chemical properties of the silt, helping to formulate a more targeted cleaning plan. After understanding the composition of the silt, the appropriate cleaning method is selected to reduce pollution and unnecessary losses during the repair process. Through dynamic rendering, the repair team can intuitively see the changes in the morphology of the silt, and understand the dredging progress and silt changes inside the pipeline in real time. The effect of dynamic rendering makes the repair work more visible, helping the repair personnel to understand the repair status of each area and adjust the cleaning strategy in time. The dredging and repair quantitative evaluation report not only provides specific data on silt accumulation, but also evaluates the efficiency and effect of the cleaning process, helping the repair team to adjust the cleaning plan. Through quantitative analysis, the repair team can optimize resource allocation and concentrate on solving areas with serious problems, thereby improving repair efficiency.

[0055] The present invention also provides a dredging and repair robot, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the drainage pipe dredging and repair evaluation method as described in any one of the above items are implemented.

[0056] Those skilled in the art clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0057] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media for storing program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks or optical disks.

[0058] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, the scope of the present invention is limited by the attached method rather than the above description, and it is intended that all changes falling within the meaning and scope of equivalent elements of the application documents are included in the present invention.

[0059] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A drainage pipe desilting and repair assessment method, characterized in that: The following steps are involved: Step S1: obtaining a multi-directional image of the interior of the drainage pipe after desilting and a pipe ultrasonic detection signal; performing visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional image of the interior of the drainage pipe to construct a three-dimensional pipe panoramic model; Step S2: performing deep residual sludge identification on the multi-directional image inside the drainage pipe and performing sludge spatial distribution analysis to generate residual sludge spatial distribution data; Step S3: optimizing the reflection deviation compensation of the pipeline ultrasonic detection signal, and calculating the sludge morphology one by one, so as to generate the morphological characteristics of each residual sludge; Step S4: Analyze the sludge composition characteristics according to the pipeline ultrasonic detection signal, so as to obtain the composition characteristics of each residual sludge; Step S5: Rendering the three-dimensional pipeline panoramic model dynamically based on the composition characteristics of each residual sludge location, the morphological characteristics of each residual sludge location, and the residual sludge spatial distribution data, thereby constructing a drainage pipeline residual sludge panoramic rendering model; Step S6: Calculate the residual siltation spatial coverage rate of the panoramic rendering model of the drainage pipe residual siltation, and conduct a comprehensive quantitative analysis of silt removal and restoration to obtain a quantitative assessment report on silt removal and restoration.

2. The drainage pipe desilting and repair assessment method according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: obtaining a multi-directional image of the interior of the drainage pipe after desilting and a pipe ultrasonic detection signal; Step S12: visually identifying pipe components on the multi-directional image of the interior of the drainage pipe, and marking a plurality of pipe internal components; Step S13: performing pipeline topological connection analysis according to multiple pipeline internal components, thereby generating a pipeline topological connection relationship; Step S14: performing internal geometric structure arrangement analysis on the multi-directional images of the interior of the drainage pipe to generate internal geometric structure data of the pipe; Step S15: Perform three-dimensional point cloud modeling based on the internal geometric structure data of the pipeline and the pipeline topological connection relationship to construct a three-dimensional pipeline panoramic model; The three-dimensional point cloud modeling is specifically as follows: Extracting an image in each orientation according to the multi-orientation images inside the drainage pipe; Identify key component points of each orientation image and mark multiple key component points; Perform multi-angle precise registration of images based on multiple key component points to generate multi-angle registration data; Based on the multi-angle registration data, multi-angle stitching processing is performed on the multi-directional images inside the drainage pipe, so as to generate a panoramic image of the inside of the pipe; According to the internal geometric structure data of the pipeline and the pipeline topological connection relationship, the three-dimensional point cloud modeling of the panoramic image inside the pipeline is carried out to construct a three-dimensional pipeline panoramic model.

3. The drainage pipe desilting and repair evaluation method according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: performing adaptive brightness enhancement processing on the multi-directional image inside the drainage pipe to obtain an adaptive brightness enhanced image; Step S22: performing super-resolution texture detail optimization on the adaptive brightness enhanced image to construct a detail optimized pipeline image; Step S23: performing deep residual sludge recognition on the detail optimized pipeline image, marking each location of residual sludge in the pipeline; Step S24: performing three-dimensional positioning of each residual sludge in the pipeline space to generate three-dimensional coordinates of each residual sludge; Step S25: Performing sludge spatial distribution analysis on each three-dimensional coordinate of residual sludge, thereby generating residual sludge spatial distribution data.

4. The drainage pipe desilting and repair assessment method according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: performing high-frequency filtering on the pipeline ultrasonic detection signal to obtain a filtered and noise-reduced ultrasonic signal; Step S32: calculating the pipeline reflection deviation of the filtered and denoised ultrasonic signal to generate a pipeline reflection deviation parameter; Step S33: performing reflection deviation compensation optimization on the pipeline reflection deviation parameters to generate deviation compensation ultrasonic parameters; Step S34: Calculate the sludge morphology of each residual sludge location one by one based on the deviation-compensated ultrasonic parameters based on the internal geometric structure data of the pipeline, thereby generating the morphological characteristics of each residual sludge location.

5. The drainage pipe desilting and repair assessment method according to claim 4, characterized in that: The specific steps of step S34 are: Performing acoustic wave reflection waveform analysis on deviation-compensated ultrasonic parameters to generate pipeline acoustic wave reflection waveform curve; Calculate the time difference between adjacent peaks of the pipeline sound wave reflection waveform curve to obtain the propagation time of the reflection signal; According to the pipeline sound wave reflection waveform curve, the reflection signal amplitude change is analyzed to generate the reflection signal amplitude change characteristics; The distance between the sediment and the pipe wall is calculated based on the internal geometric structure data of the pipe and the propagation time of the reflected signal, thereby obtaining the sludge thickness parameter; The size of the silt area is estimated based on the change characteristics of the reflected signal amplitude to generate the size of the silt area; The silt morphology is calculated one by one according to the silt thickness parameters and the silt area size, so as to generate the morphological characteristics of each residual silt.

6. The drainage pipe desilting and repair assessment method according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: Perform Fourier transform on the deviation-compensated ultrasonic parameters to obtain multiple segments of ultrasonic spectrum data; Step S42: performing spectrum bandwidth analysis on the multiple segments of ultrasonic spectrum data segment by segment to generate spectrum bandwidth characteristics of each segment; Step S43: performing sludge particle estimation based on the spectrum bandwidth characteristics of each segment to obtain particle characteristics in the sludge; Step S44: performing spectrum peak analysis on the multiple segments of ultrasonic spectrum data to obtain the spectrum peak of each segment; Step S45: mining the sludge viscosity based on the spectrum peak of each section to obtain the sludge viscosity characteristics; Step S46: Calculate the resonance frequency of the multiple ultrasonic spectrum data to obtain the degree of sludge consolidation; Step S47: Analyze the sludge composition characteristics according to the sludge consolidation degree, sludge viscosity characteristics and particle characteristics in the sludge, so as to obtain the composition characteristics of each residual sludge.

7. The drainage pipe desilting and repair assessment method according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: performing sludge position matching on the three-dimensional pipeline panoramic model based on the three-dimensional coordinates of each residual sludge location to obtain sludge position matching data; Step S52: Visually marking the three-dimensional pipeline panoramic model with the residual sludge spatial distribution data according to the sludge position matching data, thereby obtaining a residual sludge distribution visualization model; Step S53: Based on the morphological characteristics of each residual sludge, the residual sludge distribution visualization model is dynamically rendered according to the morphological characteristics of each residual sludge, and attribute mapping is performed according to the component characteristics of each residual sludge, so as to construct a panoramic rendering model of residual sludge in the drainage pipeline.

8. The drainage pipe desilting and repair assessment method according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: calculating the residual siltation space coverage rate of the drainage pipe residual siltation panoramic rendering model to generate the residual siltation space coverage rate; Step S62: performing a pipeline blockage risk analysis on the drainage pipeline residual siltation panoramic rendering model based on the residual siltation space coverage rate to generate a pipeline residual siltation blockage risk; Step S63: Performing a structural damage risk assessment on the drainage pipe residual siltation panoramic rendering model to obtain pipe structure damage risk data; Step S64: obtaining an original image of the uncleaned drainage pipe; Step S65: Based on the original image of the drainage pipe that has not been cleaned, a comprehensive quantitative analysis of the pipeline structure damage risk data and the pipeline residual silt blockage risk is performed to obtain a silt removal and repair quantitative assessment report.

9. A drainage pipe desilting and repair assessment device, characterized in that: The method for evaluating the desilting and repairing of a drainage pipe according to claim 1 comprises: A panoramic model module is used to obtain multi-directional images of the interior of the drainage pipe after desilting and the ultrasonic detection signal of the pipe; perform visual recognition of pipe components and internal geometric structure arrangement analysis on the multi-directional images of the interior of the drainage pipe to construct a three-dimensional panoramic model of the pipe; A sludge identification module is used to perform deep residual sludge identification on the multi-directional images inside the drainage pipe and perform sludge spatial distribution analysis to generate residual sludge spatial distribution data; A sludge morphology calculation module is used to optimize the reflection deviation compensation of the pipeline ultrasonic detection signal and calculate the sludge morphology one by one, so as to generate the morphological characteristics of each residual sludge; A sludge component characteristic module is used to analyze the sludge component characteristics according to the pipeline ultrasonic detection signal, so as to obtain the component characteristics of each residual sludge; The panoramic rendering module is used to render the dynamic silt morphology change of the three-dimensional pipeline panoramic model according to the composition characteristics of each residual silt, the morphological characteristics of each residual silt and the spatial distribution data of the residual silt, so as to construct a panoramic rendering model of residual silt in the drainage pipeline; The quantitative assessment module is used to calculate the residual siltation spatial coverage rate of the panoramic rendering model of the drainage pipe residual siltation, and conduct a comprehensive quantitative analysis of silt removal and restoration to obtain a quantitative assessment report on silt removal and restoration.

10. A dredging and repairing robot, comprising a memory and a processor, wherein a computer program is stored in the memory, characterized in that: When the processor executes the computer program, the steps of the drainage pipe desilting and repair evaluation method according to any one of claims 1 to 8 are implemented.

Images