Sewer Defects Research Articles

Automatic sewer pipe defect semantic segmentation based on improved U-Net

At present, technologies based on deep learning methods for automated detection of sewer defects have been developing rapidly. In this study, a novel semantic segmentation network called PipeUNet is proposed for sewer defect segmentation. In order to enhance the feature extraction capability and resolve semantic differences between high level and low level features, a new module named feature reuse and attention mechanism block is added between the original skip connections of U-Net. Focal loss is adopted to solve the class imbalance problem. PipeUNet was trained using the CCTV images with typical defects including crack, infiltration, joint offset and intruding lateral. It was tested by the defects images and normal images to evaluate the network's defect segmentation and detection performance respectively. It achieved the highest Mean Intersection over Union of 76.37% which proved the proposed approach's efficiency. It can process CCTV images at a high speed of 32 images per second.

Deep Learning–Based Automated Detection of Sewer Defects in CCTV Videos

AbstractAutomated interpretation of closed-circuit television (CCTV) inspection videos could improve the speed and consistency of sewer condition assessment. Previous approaches focus on defect cla...

Automatic Detection and Classification of Sewer Defects via Hierarchical Deep Learning

Video and image sources are frequently applied in the area of defect inspection in industrial community. For the recognition and classification of sewer defects, a significant number of videos and images of sewers are collected. These data are then checked by human and some traditional methods to recognize and classify the sewer defects, which is inefficient and error-prone. Previously developed features like SIFT are unable to comprehensively represent such defects. Therefore, feature representation is especially important for defect autoclassification. In this paper, we study the automatic extraction of feature representation for sewer defects via deep learning. Moreover, a complete automatic system for classifying sewer defects is proposed built on a two-level hierarchical deep convolutional neural network, which shows high performance with respect to classification accuracy. The proposed network is trained on a novel data set with over 40 000 sewer images. The system has been successfully applied in the practical production, confirming its robustness and feasibility to real-world applications. The source code and trained model are available at the project website. 1 Note to Practitioners —Automatic defect inspection has become a fundamental research topic in engineering application field. Specifically, sewer defect detection is an important measure for maintenance, renewal, and rehabilitation activities of sewer infrastructure. In the current operation procedure, all the captured videos need to be inspected by experts frame by frame to recognize defects, yielding a significant low inspection rate with a significant amount of time. Previous work has attempted to employ traditional image processing methods for automated sewer defect classification. However, these methods get poor generalization capabilities since they use pre-engineered features. In most cases, sewerage inspection companies have to hire numerous professional inspectors to do this job, thereby consuming a lot of human and material resources. To address this problem, the authors propose an automatic detection and classification method for sewer defects based on hierarchical deep learning. Demonstrated by various experiments, the designed framework achieves a high defect classification accuracy, which can be easily integrated into an automatic sewer defect inspection system. 1 https://github.com/NUAAXQ/SewerDefectDetection

A unified convolutional neural network integrated with conditional random field for pipe defect segmentation

AbstractSemantic segmentation of closed‐circuit television (CCTV) images can facilitate automatic severity assessment of sewer pipe defects by assigning defect labels to each pixel on the image, from which defect types, locations, and geometric information can be obtained. In this study, a unified neural network, namely DilaSeg‐CRF, is proposed by fully integrating a deep convolutional neural network (CNN) with dense conditional random field (CRF) for improving the segmentation accuracy. First, DilaSeg is constructed with dilated convolution and multiscale techniques for producing feature maps with high resolution. The steps of the dense CRF inference algorithm are converted into CNN operations, which are then formulated as recurrent neural network (RNN) layers. The DilaSeg‐CRF is proposed by integrating DilaSeg with the RNN layers. Images containing three common types of sewer defects are collected from CCTV inspection videos and are annotated with ground truth labels, after which the proposed models are trained and evaluated. Experiments demonstrate that the end‐to‐end trainable DilaSeg‐CRF can improve the segmentation significantly, with an increase of 32% and 20% in mean intersection over union (mIoU) values compared with fully convolutional network (FCN‐8s) and DilaSeg, respectively. Our proposed DilaSeg‐CRF also achieves faster inference speed than FCN and eliminates the manual postprocessing for refining the segmentation results.

Designing Smart Sewerbot for the Identification of Sewer Defects and Blockages

Internet of thing (IoT) is a new concept where the term ‘thing’ is associated with the configurable sensors and devices no matter domestic or industrial, whereas bridging up a relationship in between these things and internet protocol is known as Internet of thing. Moreover, the same concept has been introduced in the field of robotics as ‘Internet of Robotic Things (IoRT)’, which is mainly concerned with active sensorization of sensors dully interfaced with any type of robots i.e. autonomous unmanned ground vehicle (UGV). This paper describes the prototyping of an autonomous sewerbot that will not only identify the sewer defects in sewerage pipelines but will also identify the type of blockages using the technique of digital image processing. Furthermore, the deployed configurable sensors will also share the attributes of particular sewerage line on IoT such that temperature, humidity, availability of hazardous gases, exact depth at which it is available and global positioning using GPS module. The paper also provides the brief construction of this mechatronic and amphibian system via which it can extricate the blockages from sewerage lines along with wireless camera surveillance.

An integrated assessment approach to prevent risk of sewer exfiltration

Regular inspection, assessment, and rehabilitation are necessary to maintain the performance levels of sewer pipelines. The primary objective of this research is to develop a causality-based model by incorporating twenty-two sewer defects, including erosion voids. The Decision Making Trial Evaluation Laboratory (DEMATEL) was used to study the causality relationship between the defects, based on a designed questionnaire, to conclude the influencing and influenced distress. The results suggested that 59% of the sewer structural and operational defects were influencing ones. Besides, the existing state of the sewer pipelines was evaluated on a scale from 1 (Excellent) to 5 (Critical) by integrating the DEMATEL technique with the Quality Function Deployment (QFD) method. The QFD-DEMATEL model was implemented on a case study to compare the model’s classifications of the pipelines’ states with the actual ones. Based on the comparison, the average accuracy and true negative rate (TNR) of the classifications were 74% and 79% respectively. Also, the average invalidity percentage (AIP) was 41.32%, causing the validity percentage to exceed 50%. The integrated approach keeps decision-makers informed about their sewer pipelines to prioritize pipelines rehabilitation. Thus, defective sewers are reduced and so are the exfiltration cases.

Multiattribute Utility Theory Deployment in Sewer Defects Assessment

AbstractAssessing the condition of sewer pipelines is a backbone process to plan for rehabilitation and maintenance work. The closed circuit-television (CCTV) method is the widely adopted method to...

Management of risk of tree and shrub root intrusion into sewers

The major objective of this study was to determine the frequency of root intrusion into sewers and the sizes of the roots reported. The data were used to develop two methods: one for determining risks related to root intrusions into sewers and sewage flow blockage and the other for establishing the category of probability of root intrusion into sewers with structural defects.The data on root intrusions into sewers and sewer defects allowing such intrusions were collected through CCTV surveys conducted by the Kielce University of Technology for more than 29km of concrete and vitrified clay sewers. The frequency of root intrusions into sewers and the sizes of the roots were determined for sewers in Polish cities. The root sizes were classified as one of the five categories of probability of sewage flow blockage, proposed on the basis of the ratio of the cross-sectional area of the roots intruding into a sewer to the total cross-sectional area of the sewer. The study involved proposing a method for determining the category of root intrusion consequences for sewers with roots already growing inside and a method for determining the sewage flow blockage risks related to root intrusion. The latter method can be used to establish the category of root intrusion probability on the basis of structural sewer defects allowing root intrusion as well as factors not related to the sewer, e.g. tree species or the distance of the tree from the sewer.From the investigations it is evident that root intrusions into sewers are serious problems in Poland and in other countries. The methods proposed in this paper are important tools to be used for proper management of sewer systems.

원심력철근콘크리트관의 결함에 따른 심각도 평가 -이음부 사례를 중심으로-

Sewer joint-related defect is one of the most common domestic sewer defects along with the lateral pipe problem. However, there are currently no criteria that precisely assess the joint-related sewer defects. Therefore, this study examined the joint-related sewer defects found in domestic circumstances, classified them according to the suggested defect code, and presented the examples of defect pictures. Each defect code was organized as the process of out of pipeline alignment (OPA) which shows the progress in deterioration. Each defect was classified into 5 grades depending on appropriate repair and rehabilitation method. The result of this study is expected to be useful for domestic CCTV inspectors to assess the sewer condition and helpful for managers to make a decision of repair and rehabilitation.

Rapid detection of sewer defects and blockages using acoustic-based instrumentation

Sewer flooding incidents in the UK are being increasingly associated with the presence of blockages. Blockages are difficult to deal with as although there are locations where they are more likely to occur, they do occur intermittently. In order to manage sewer blockage pro-actively sewer managers need to be able to identify the location of blockages promptly. Traditional closed-circuit television (CCTV) inspection technologies are slow and relatively expensive so are not well suited to the rapid inspection of a network. This is needed if managers are to be able to address sewer blockages proactively. This paper reports on the development of an acoustic-based sensor. The sensor was tested in a full scale sewer pipe in the laboratory and it was shown that it is able to find blockages and identify structural aspects of a sewer pipe such as a manhole and lateral connection. Analysis of the received signal will locate a blockage and also provide information on its character. The measurement is very rapid and objective and so inspections can be carried out at much faster rates than using existing CCTV technologies.

Assessing sewer–groundwater interaction at the city scale based on individual sewer defects and marker species distributions

By combining a sewer defect database and hydrogeological information, it has been attempted to assess sewer–groundwater interaction at the scale of the city of Rastatt (SW-Germany). Comprehensive hydrochemical samplings, including a series of new marker species, have been conducted in the urban aquifer and used for validation. The iodated X-ray contrast media, amidotrizoic acid and iothalamic acid, can be considered as highly specific for wastewater influence and have been found in several groundwater observation wells.

Quantifying Sub-Surface Discharges from Individual Sewer Defects

Quantifying Sub-Surface Discharges from Individual Sewer Defects