Segmentation of Pipe Images for Crack Detection in Buried Sewers

Abstract:   Assessing the condition of underground pipelines such as water lines, sewer pipes, and telecommunication conduits in an automated and reliable manner is vital to the safety and maintenance of buried public infrastructure. To fully automate condition assessment, it is necessary to develop robust data analysis and interpretation systems for defects in buried pipes. This article presents the development of an automated data analysis system for detecting defects in sanitary sewer pipelines. We propose a three-step method to identify and extract cracks from contrast enhanced pipe images. This method is based on mathematical morphology and curvature evaluation that detects crack-like patterns in a noisy pipe camera scanned image. As cracks are the most common defects in pipes and are indicative of the residual structural strength of the pipe, they are the focus of this study. This article discusses its implementation on 225 pipe images taken from different cities in North America and shows that the system performs very well under a variety of pipe conditions.     

Automated defect detection for sewer pipeline inspection and condition assessment

To regularly and proactively assess conditions of sewer infrastructure systems to ensure their structural integrity and continuity of services, it is critical to advance the state of automated pipeline inspection and condition assessment. Currently, a critical issue is to address realistic defect detection that deals with real sewer inspection data. This paper presents the findings of a research project that seeks to enable automated detection of defects in sewer pipelines from inspection videos and images. The need for and the challenges of automated defect detection in sewer infrastructure condition monitoring are presented. Based on a general detection and recognition model established in this paper, a change detection based approach which is tailored to solve the challenges in this sewer pipeline domain is described and illustrated through case study.     

Segmentation of buried concrete pipe images

The enormity of the problem of deteriorating pipeline infrastructure is widely apparent. Since a complete rebuilding of the piping system is not financially realistic, municipal and utility operators require the ability to monitor the condition of buried pipes. Thus, reliable pipeline assessment and management tools are necessary to develop long term cost effective maintenance, repair, and rehabilitation programs. In this paper a simple, robust and efficient image segmentation algorithm for the automated analysis of scanned underground pipe images is presented. The algorithm consists of image pre-processing followed by a sequence of morphological operations to accurately segment pipe cracks, holes, joints, laterals, and collapsed surfaces, a crucial step in the classification of defects in underground pipes. The proposed approach can be completely automated and has been tested on five hundred scanned images of buried concrete sewer pipes from major cities in North America.     

Sewer pipe defects diagnosis assessment using multivariate analysis on CCTV video imagery

Closed circuit television (CCTV) technology has been commonly used to inspect underground pipe defects, and high CCTV image quality is a prerequisite for accurate defect diagnosis. An acceptance criterion for CCTV inspection videos is critical for ensuring accurate diagnosis and preventing disputes between employers and contractors. This paper used multivariate statistical methods to evaluate the overall quality of CCTV images and to define an acceptance criterion for CCTV videos. Numerous CCTV images from a sewer inspection project were assessed and their quality, consisting of similarity in luminance and contrast distortions, was calculated by comparing a set of ideal images. Principal component analysis (PCA) and redundancy analysis (RDA) grouped the CCTV videos into homogeneous segments with similar image quality and provided a visual acceptance criterion for CCTV inspection videos. Furthermore, RDA triplot indicated that the contrast improvement of CCTV images can effectively enhance image quality and increase the diagnosis efficiency.     

Computer Vision Techniques for Automatic Structural Assessment of Underground Pipes

Pipeline surface defects such as cracks cause major problems for asset managers, particularly when the pipe is buried under the ground. The manual inspection of surface defects in the underground pipes has a number of drawbacks, including subjectivity, varying standards, and high costs. An automatic inspection system using image processing and artificial intelligence techniques can overcome many of these disadvantages and offer asset managers an opportunity to significantly improve quality and reduce costs. This article presents a system for the application of computer vision techniques to the automatic assessment of the structural condition of underground pipes. The algorithm consists of image preprocessing, a sequence of morphological operations to accurately extract pipe joints and laterals (where smaller pipe is connected to main bigger pipe), and statistical filters for detection of surface cracks in the pipeline network. The proposed approach can be completely automated and has been tested on over 1,000 scanned images of underground pipes from major cities in North America.     

Automated detection of sewer pipe defects in closed-circuit television images using deep learning techniques

Sanitary sewer systems are designed to collect and transport sanitary wastewater and stormwater. Pipe inspection is important in identifying both the type and location of pipe defects to maintain the normal sewer operations. Closed-circuit television (CCTV) has been commonly utilized for sewer pipe inspection. Currently, interpretation of the CCTV images is mostly conducted manually to identify the defect type and location, which is time-consuming, labor-intensive and inaccurate. Conventional computer vision techniques are explored for automated interpretation of CCTV images, but such process requires large amount of image pre-processing and the design of complex feature extractor for certain cases. In this study, an automated approach is developed for detecting sewer pipe defects based on a deep learning technique namely faster region-based convolutional neural network (faster R-CNN). The detection model is trained using 3000 images collected from CCTV inspection videos of sewer pipes. After training, the model is evaluated in terms of detection accuracy and computation cost using mean average precision (mAP), missing rate, detection speed and training time. The proposed approach is demonstrated to be applicable for detecting sewer pipe defects accurately with high accuracy and fast speed. In addition, a new model is constructed and several hyper-parameters are adjusted to study the influential factors of the proposed approach. The experiment results demonstrate that dataset size, initialization network type and training mode, and network hyper-parameters have influence on model performance. Specifically, the increase of dataset size and convolutional layers can improve the model accuracy. The adjustment of hyper-parameters such as filter dimensions or stride values contributes to higher detection accuracy, achieving an mAP of 83%. The study lays the foundation for applying deep learning techniques in sewer pipe defect detection as well as addressing similar issues for construction and facility management.     

Automated classification of civil structure defects based on convolutional neural network

Today, the most commonly used civil infrastructure inspection method is based on a visual assessment conducted by certified inspectors following prescribed protocols. However, the increase in aggressive environmental and load conditions, coupled with the achievement of many structures of the life-cycle end, has highlighted the need to automate damage identification and satisfy the number of structures that need to be inspected. To overcome this challenge, this paper presents a method for automating concrete damage classification using a deep convolutional neural network. The convolutional neural network was designed after an experimental investigation of a wide number of pretrained networks, applying the transfer-learning technique. Training and validation were conducted using a database built with 1352 images balanced between “undamaged”, “cracked”, and “delaminated” concrete surfaces. To increase the network robustness compared to images in real-world situations, different image configurations have been collected from the Internet and on-field bridge inspections. The GoogLeNet model, with the highest validation accuracy of approximately 94%, was selected as the most suitable network for concrete damage classification. The results confirm that the proposed model can correctly classify images from real concrete surfaces of bridges, tunnels, and pavement, resulting in an effective alternative to the current visual inspection techniques.     

A data warehouse-based decision support system for sewer infrastructure management

Since the inception of the Governmental Accounting Standards Board statement-34 (GASB 34) in the United States, local and state governing entities need to inspect sewer systems and collect general information about their properties. Application of the collected information in decision-making processes, however, is often problematic due to the lack of consistency and completeness of infrastructure data. In addition, most techniques involved in decision-making processes are relatively complicated and difficult to implement without a certain level of engineering experience and training. Consequently, the sharing and transferring of pertinent information among stakeholders is not smooth and is frequently limited. This study presents a decision support system (DSS) for the management of sewer infrastructure using data warehousing technology. The proposed decision support system automatically assigns appropriate inspection and renewal methods for each pipeline and estimates associated costs, resulting in effective and practical sewer infrastructure management from various perspectives, with corresponding levels of detail.     

Sewer asset management – state of the art and research needs

ABSTRACT

Sewer asset management gained momentum and importance in recent years due to economic considerations, since infrastructure maintenance and rehabilitation directly represent major investments. Because physical urban water infrastructure has life expectancies of up to 100 years or more, contemporary urban drainage systems are strongly influenced by historical decisions and implementations. The current decisions taken in sewer asset management will, therefore, have a long-lasting impact on the functionality and quality of future services provided by these networks. These decisions can be supported by different approaches ranging from various inspection techniques, deterioration models to assess the probability of failure or the technical service life, to sophisticated decision support systems crossing boundaries to other urban infrastructure. This paper presents the state of the art in sewer asset management in its manifold facets spanning a wide field of research and highlights existing research gaps while giving an outlook on future developments and research areas.     

Construction of underground works in Shanghai, P.R.C.

For nearly thirty years, Shanghai has been constructing underground works designed to improve the city's infrastructure. A number of projects were undertaken using shield tunnelling method in order to determine whether this method could be used successfully in Shanghai's geology, which is characterized by silt and silty or puddly clay. These underground projects have included subway tunnels, sewage and water supply tunnels, road tunnels, and offshore discharge tunnels. This paper describes eight subsurface projects in Shanghai that have used the shield tunnelling method successfully.     

Optical flow techniques for estimation of camera motion parameters in sewer closed circuit television inspection videos

This paper discusses a novel approach for automated analysis and tracking of camera motion in sewer inspection closed circuit television (CCTV) videos. This approach represents an important building block for any system that supports automated analysis and defect detection of CCTV videos. The proposed approach employs optical flow techniques to automatically identify, locate, and extract a limited set of video segments, called regions of interest (ROI), which likely include defects, thus reducing the time and computational requirements needed for video processing. Tracking the camera motion parameters is used to recover the operator actions during the inspection session, which would provide important clues about the location and severity of the ROI. Techniques for estimating the camera travelling distance, position inside the sewer, and direction of motion from optical flow vectors are discussed. The proposed techniques were validated using a representative set of sewer CCTV videos obtained from the cities of Regina and Calgary, Canada.     

Use of compact shield tunneling method in urban underground construction

In Tokyo Metropolis, the rehabilitation and renewal of sewer systems is an urgent issue due to the aged sewers and increases in wastewater and stormwater runoff. In such urban area, shield tunneling confronts various problems, such as high construction costs, adverse effects on living environments, and densely used surface and underground spaces. To solve these problems, the authors developed a new shield tunneling method, which is called the “compact shield” method. This paper describes the concepts used for the construction of shield sewer tunnels without inner linings and an overview of the segments and a shield machine that were newly developed.     

The importance of gully flow modelling to urban flood simulation

ABSTRACT

Urban flood simulation often ignores or simplifies the function of underground gully systems due to data and computational limitations. To discover the influence of gullies on urban flooding, a novel approach is proposed in this study, by fully-coupling a 1D gully flow model (GFM), a 1D sewer flow model (SFM), and a 2D overland flow model (OFM) to simultaneously simulate the flow exchanges between surface, gullies and sewer pipes. This fully-coupled approach is compared with a simplified approach which directly introduces surface water into sewer pipes without being via gullies. The validation results show that the fully-coupled approach considerably reduces the underestimation of flood extent by 27% compared with the simplified approach. Without considering the capacity of lateral tubes between gullies and sewer pipes, the simplified approach over-drains the surface water into sewer pipes. The modelling of gully flow is crucial for correctly evaluating the efficiency of drainage systems.     

Visual sewer inspection: detail of coding system versus data quality?

In order to provide information about the decisions on proactive and reactive maintenance, sewers are visually inspected. Previous research showed that the quality of visual inspection data is questionable. A coding system prescribes which and how defects should be recorded. This article studies the influence of the coding system on quality of inspection data. A database with the examinations of the Dutch sewer inspector course is studied. Through time, 10 photos of the inside of a sewer were evaluated according to two different coding systems: the concise NEN3399:1992 and the more detailed and extensive NEN3399:2004.This article compares both coding systems by evaluating candidate responses to photos showing sewers with clearly visible defects. Results show that added detail in the coding system of 2004 leads to more mistakes. Therefore, it can be concluded that the increase in detail does not lead to more information.     

Simulation modeling techniques for underground infrastructure construction processes

Simulation is an efficient and cost-effective tool for decision-making and analyzing real-world systems and repetitive construction processes. Tunneling and trenchless construction processes are excellent candidates for the utilization of computer simulation due to their repetitive nature. This paper presents six simulation tools that have been developed over the last five years and implemented to plan and manage a range of several applications in underground infrastructure construction. The purpose of the tools, modeling framework, modeling logic, inputs, and outputs for tunneling, soil type prediction, sewer condition forecasting, pipeline routing, horizontal directional drilling, and trenchless pipe replacement are presented. The successful development and implementation of the tools presented in this paper further illustrate the usefulness of employing simulation for pre-planning and decision-making to reduce uncertainty inherent in construction projects involving underground infrastructure systems.     

Automated recognition of surface defects using digital color image processing

One of the computerized technologies for advanced infrastructure inspection methods is the application of digital image processing. Digital image processing methods have been developed for steel bridge coating inspections for the past few years. The rust percentages on steel bridge coating surfaces can be reliably computed through the use of digital image processing methods. However, previous researchers solely focused on the determination of the degree of rust defects on the steel surfaces in percentage. Therefore, an automated processor that can recognize the existence of bridge coating rust defects needs to be developed. This paper presents the development of a rust defect recognition method to determine whether rust defects exist in a given digital image by processing digital color information. For the development of the image processor, color image processing is employed, instead of grayscale image processing commonly used in previous researches, since rust defects are distinctive in color against background.     

Attention-guided multiscale neural network for defect detection in sewer pipelines

Sanitary sewer systems are major infrastructures in every modern city, which are essential in protecting water pollution and preventing urban waterlogging. Since the conditions of sewer systems continuously deteriorate over time due to various defects and extrinsic factors, early intervention in the defects is necessary to prolong the service life of the pipelines. However, prior works for defect inspection are limited by accuracy, efficiency, and economic cost. In addition, the current loss functions in object detection approaches are unable to handle the imbalanced data well. To address the above drawbacks, this paper proposes an automatic defect detection framework that accurately identifies and localizes eight types of defects in closed-circuit television videos based on a deep neural network. First, an effective attention module is introduced and used in the backbone of the detector for better feature extraction. Then, a novel feature fusion mechanism is presented in the neck to alleviate the problem of feature dilution. After that, an efficient loss function that can reasonably adjust the weight of training samples is proposed to tackle the imbalanced data problem. Also, a publicly available dataset is provided for defect detection tasks. The proposed detection framework is robust against the imbalanced data and achieves a state-of-the-art mean average precision of 73.4%, which is potentially applied in realistic sewer defect inspections.     

Automated defect detection tool for closed circuit television (cctv) inspected sewer pipelines

In sewer networks, the economic effects and costs that result from a pipeline failure are rising sharply. As a result, there is huge demand for inspection and rehabilitation of sewer pipelines. In addition to being inaccurate, current practices of sewer pipelines inspection are time consuming and may not keep up with the deterioration rates of the pipelines. This papers presents the development of an automated tool to detect some defects such as: cracks, deformation, settled deposits and joint displacement in sewer pipelines. The automated approach is dependent upon using image-processing techniques and several mathematical formulas to analyze output data from Closed Circuit Television (CCTV) camera images. The automated tool was able to detect cracks, displaced joints, ovality and settled deposits in pipelines using CCTV camera inspection output footage using two different datasets. To examine the performance of the proposed detection methodology, confusion matrices were constructed, in which true positives for crack, settled deposits and displaced joints were 74%, 53% and 65%. As for the ovality, all defects in the images were detected successfully. Although these values could indicate low performance, however the proposed methodology could be improved if additional images were used. Given that one inspection session can result in hundreds of CCTV camera footage, introducing an automated tool would help yield faster results. Additionally, given the subjective nature of evaluating the severity of defects, it would result in more systematic outputs since the current method rely heavily on the operator's experience.     

淮南市中央公园下穿隧道抗浮地下水位分析

根据淮南市山南新区南纬六路下穿隧道抗浮地下水位工程的实际情况,从场区地形地貌、地表水系规划、水文地质条件、地下水位控制要素等方面进行了分析,合理确定设计抗浮地下水位,以期从最大程度上满足工程建设的安全运营。