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.     

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.     

An economic loss model for failure of sewer pipelines

Estimating the costs of failure for sewer pipelines is usually accompanied with uncertainties because of the difficulty in capturing the relationship between the physical and economical characteristics of failed pipelines. To reduce such uncertainties economic loss models are usually used to evaluate the consequences of failure. This paper presents a methodology to estimate economic loss as a result of sewer pipelines’ failure using cost benefit analysis approach. Costs of sewer pipelines’ failure in addition to costs resulting from avoiding such failures are identified and analysed. To validate the proposed methodology, actual costs from a real failure incident were compared with the proposed model outputs. The model could estimate the direct and indirect costs with a deviation ranging between 10–12% and 22–30%, respectively. By implementing the proposed methodology on two case studies, it was found that the indirect costs as a result of sewer pipelines’ failure represent a significant portion ranging between 89 and 94% of the total costs of failure. Also, it was found that costs related to environment, delays to work and traffic disruptions contribute by 12–35% to the indirect costs.     

Multiobjective optimisation algorithm for sewer network rehabilitation

Understanding of deterioration mechanisms in sewers helps asset managers in developing prediction models for estimating whether or not sewer collapse is likely. Effective utilisation of deterioration prediction models along with the development and use of life cycle maintenance cost analysis contribute to reducing operation and maintenance costs in sewer systems. This article presents a model for life-cycle maintenance planning of deteriorating sewer network as a multi-objective optimisation problem that treats the sewer network condition and service life as well as life-cycle maintenance cost (LCMC) as separate objective functions. The developed model utilises Markov chain model for the prediction of the deterioration of the network. A multi-objective genetic algorithm is used to automatically locate an appropriate maintenance scenario that exhibits an optimised tradeoff among conflicting objectives. Monte Carlo simulation is used to account for LCMC uncertainties. The optimisation algorithm provides an improved opportunity for asset managers to actively select near-optimum maintenance scenario that balances life-cycle maintenance cost, condition and service life of deteriorating sewer network. A case study is used to demonstrate the practical features of developed methodology.     

优化神经网络在多传感器火灾探测中的应用

针对BP神经网络在拟合过程中探测精度低、容易陷入局部最优的问题,提出一种基于遗传算法（GA）和模拟退火算法（SA）共同改进的BP神经网络模型,该网络模型可以有效提高火灾识别准确率,同时避免网络过拟合现象,使预测结果跳出局部最优从而达到全局最优。首先,通过GA改进隐藏层结构部分,然后通过SA改进连接权重部分,最后利用优化后的GA-SA-BP模型对火灾实验数据进行信息融合实现火灾探测。实验研究表明,对比单一BP神经网络,经GA和SA改进后的BP神经网络能够有效改善网络拟合能力,并提升火灾探测精度至98.91%。     

热力管道状态检测技术分析

介绍了热力管道的敷设形式,阐述了可通行管道与直埋管道的检测技术,并分析了各种检测方法的优缺点,有利于选择适宜的管道检测技术,提高热力管道的安全系数。     

Automated defect detection in sewer closed circuit television images using histograms of oriented gradients and support vector machine

Condition assessment of municipal sewer pipes using closed circuit television (CCTV) inspections is known to be time consuming, costly, and prone to errors primarily due to operator fatigue or novicity. Automated detection of defects can provide a valuable tool for ensuring the quality, accuracy, and consistency of condition data, while reducing the time and cost of the inspection process. This paper presents an efficient pattern recognition algorithm to support automated detection and classification of pipe defects in images obtained from conventional CCTV inspection videos. The algorithm employs the histograms of oriented gradients (HOG) and support vector machine (SVM) to identify pipe defects. The algorithm involves two main steps: (1) image segmentation to extract suspicious regions of interest (ROI) that represent candidate defect areas; and (2) classification of the ROI using SVM classifier that was trained using sets of HOG features extracted from positive and negative examples of the defect. Proposed algorithm is applied to the problem of detecting tree root intrusions. The performance of linear and radial basis function SVM classifiers evaluated. The algorithm was tested on a set of actual CCTV videos obtained from the cities of Regina and Calgary in Canada. Experimental results demonstrated the viability and robustness of the algorithm.     

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

Semantic 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.     

Combination of neural network and wavelet transform for islanding detection of distributed generation in a small-scale network

In this paper, an optimum and intelligent method is proposed for islanding detection using wavelet transform. The suggested relay is based on neural network (NN) in which different heuristic algorithms are used for training the NN. In the proposed method, the appropriate signals for detection procedure as well as mother wavelet are selected optimally, based on the mean square error (MSE) concept. Lately, the desired relay is trained by the optimally selected signals using four different algorithms and the optimum condition of the fault detector is identified. Simulation results approved that non detection zone (NDZ) has a significant reduction utilising the proposed intelligent technique. The contributions of the proposed method include presenting an appropriate signal selection method based on MSE, selecting optimum number of relay input signals using the proposed technique, fast training of intelligent relay by using least information, solving threshold selection problem and reduction of NDZ approximately to zero.     

人工神经网络技术在桥梁检测评估中的应用

对现有桥梁的评估方法作了简要介绍,重点讨论了常用的BP人工神经网络模型,并将BP模型应用到桥梁结构检测评估中,指出人工神经网络在桥梁结构的检测评估方面一定有很好的发展前景.     

Pavement defect detection with fully convolutional network and an uncertainty framework

Image segmentation has been implemented for pavement defect detection, from which types, locations, and geometric information can be obtained. In this study, an integration of a fully convolutional network with a Gaussian‐conditional random field (G‐CRF), an uncertainty framework, and probability‐based rejection is proposed for detecting pavement defects. First, a fully convolutional network is designed to generate preliminary segmentation results, and a G‐CRF is used to refine the segmentation. Second, epistemic and aleatory uncertainties in the model and database are considered to overcome the disadvantages of traditional deep‐learning methods. Last, probability‐based rejection is conducted to remove unreasonable segmentations. The proposed method is evaluated on a data set of images that were obtained from 16 highways. The proposed integration segments pavement distresses from digital images with desirable performance. It also provides a satisfactory means to improve the accuracy and generalization performance of pavement defect detection without introducing a delay into the segmentation process.     

Automated detection of surface defects in water and sewer pipes

Automation is gaining momentum in industry, particularly in rehabilitation and inspection works of underground infrastructure facilities. This paper describes a model for automating inspection and identification of surface defects in underground water and sewer pipes. The paper describes the current efforts in identification of surface defects in underground water and sewer mains, and presents an automated system designed to assist infrastructure engineers in diagnosing defects in this class of pipe networks. It describes the general architecture of the system and its basic components, and focuses primarily on four modules designed for automating image acquisition, image processing, features extraction and classification of defects.     

Fine-grained crack segmentation for high-resolution images via a multiscale cascaded network

High-resolution (HR) crack images offer more detailed information for assessing structural conditions compared to low-resolution (LR) images. This wealth of detail proves indispensable in bolstering the safety of unmanned aerial vehicle (UAV)-based inspection procedures and elevating the precision of small crack segmentation. Nonetheless, achieving a balance between segmentation accuracy and GPU memory consumption poses a substantial challenge for deep learning models when processing HR crack images. To overcome this challenge, a novel “HR crack segmentation framework” (HRCSF) is proposed, specifically designed to meticulously segment crack images with resolutions exceeding 4K. First, a multiscale crack feature extraction network (MsCFEN) was proposed with the embedment of the strip pooling operation to enhance the representation of the transverse and longitudinal crack pixels from the complex backgrounds. Subsequently, two cascaded operations were tailored to MsCFEN, enabling a comprehensive refinement process that incorporates both global and local aspects. Furthermore, to fully leverage the potential of each proposed component in the refinement process, the complete architecture was trained using a loss function with embedded boundary optimization. Conclusively, a UAV-based case study was conducted on a real bridge in Changsha, demonstrating HRCSF's practicability in segmenting HR crack images. The implementation of HRCSF allows the UAV to perform crack inspection effectively from a distance of 3 m away from the girder, resulting in a significant 50% reduction in inspection time compared to LR segmentation methods while maintaining high detection accuracy.     

GA优化BP神经网络在多参量火灾探测中的应用

针对BP神经网络的随机权重和阈值稳定性不高的问题,运用遗传算法(GA)对BP神经网络的初始权重和阈值进行优化,提出了一种基于GA优化BP神经网络的多参量数据融合方法以实现火灾探测,提高火灾探测准确率和模型泛化性能,并利用该模型对标准明火和阴燃火中的温度、烟雾浓度和CO浓度进行数据融合实现火灾探测.研究显示,相较单纯BP...     

Estimation of raw sewage discharge resulting from sewer network failures

The raw sewage volume potentially discharged from a sewer network to the environment arising from its component failures is proposed as a measure of overall system reliability. This paper presents a simple method for quickly and properly calculating this volume. The basis for this method is a representation of the sewer network by a combination of Y-like fragments named here as structure-forming elements. Each such element is formally substituted by a fictitious equivalent sewer with a failure rate leading to the same output for the same input. A sequential application of this approach reduces the problem of estimating the discharged sewage volume to elementary subproblems for which the solution is simple. Examples are used to demonstrate the possibilities of this method.     

Pixel-level thin crack detection on road surface using convolutional neural network for severely imbalanced data

Because roads are the major backbone of the transportation network, research about crack detection on road surfaces has been popular in computer and infrastructure engineering. When training a convolutional neural network (CNN) for pixel-level road crack detection, three common challenges include (1) the data are severely imbalanced, (2) crack pixels can be easily confused with normal road texture and other visual noises, and (3) there are many unexplainable characteristics regarding the CNN itself. When it comes to very fine and thin cracks, these challenges are exaggerated and a new challenge is introduced, as there can be a discrepancy between the actual width and the annotated width of a crack. To tackle all these challenges of thin crack detection, this paper proposes a new variant of CNN named ThinCrack U-Net, designed to provide thin results upon pixel-level crack detection on road surfaces. The main contribution is to demystify how pixel-level thin crack detection results are affected by different loss functions as well as various combinations of the U-Net components. The experimental results show that ThinCrack U-Net yields a significant performance boost in CrackTree260, from 65.71% to 94.48% F-measure, compared to the existing variant of U-Net previously proposed in the context of pixel-level thin crack detection. Finally, this paper locates the source of undesirable result thickness and solves it with the balanced usage of downsampling/upsampling layers and atrous convolution. Unlike suggested by previous works, different loss functions show no significant impact on ThinCrack U-Net, whereas normalization layers are proved crucial in pixel-level thin crack detection.     

Experimental evaluation of a vibration-based leak detection technique for water pipelines

Conventional water pipeline leak detection surveys employ labour-intensive acoustic techniques, which are usually expensive and not amenable for continuous monitoring of distribution systems. Many previous studies attempted to address these limitations by proposing and evaluating a myriad of continuous, long-term monitoring techniques. However, these techniques have difficulty to identify leaks in the presence of pipeline system complexities (e.g. T-joints), offered limited compatibility with popular pipe materials (e.g. PVC), and were in some cases intrusive in nature. Recently, a non-intrusive pipeline surface vibration-based leak detection technique has been proposed to address some of the limitations of the previous studies. This new technique involves continuous monitoring of the change in the cross-spectral density of surface vibration measured at discrete locations along the pipeline. Previously, the capabilities of this technique have been demonstrated through an experimental campaign carried out on a simple pipeline set-up. This paper presents a follow-up evaluation of the new technique in a real-size experimental looped pipeline system located in a laboratory with complexities, such as junctions, bends and varying pipeline sizes. The results revealed the potential feasibility of the proposed technique to detect and assess the onset of single or multiple leaks in a complex system.     

基于神经网络算法的多参数火灾探测系统设计

为进一步拓宽神经网络算法在火灾探测领域的应用,利用以阴燃火、明火和无火发生概率作为输出结果的国家标准火实验数据,在MATLAB平台上训练了一个多层BP神经网络,此网络对于测试集样本拟合的决定性系数均达到了0.95以上.以AT89C52为主控芯片,在Proteus中设计出了一种能满足火灾探测目的的仿真电路图.通过将训练好...     

基于微粒群算法和模糊神经网络的火灾探测研究

针对火灾探测的特点,将模糊系统和神经网络有机结合,实现模糊系统设计参数的自动调整。采用符合国家标准明火、阴燃火以及厨房环境下的干扰火等作为模糊神经网络的训练样本和测试样本,依据模糊神经网络算法要求,完成了网络结构的设计,并给出相应的计算模型,利用微粒群算法对网络的权值进行学习与训练。结果表明,该算法在探测国家标准火的火灾状态方面具有有效性和可行性。     

基于卷积神经网络的视频烟雾探测追踪技术研究

火灾探测预警技术是有效降低火灾损失、辅助扑救火灾,保护人民生命财产安全的重要技术保障,是对烟雾进行探测较为有效的手段之一。目前,大部分烟雾探测报警装置主要设置于室内空间场所,仅具备探测和报警功能,同时误报率相对较高,也无法同步传递实时视频画面信息,对室外空间区域也无法进行探测。针对上述情况,基于视频监控系统对烟雾进行实时探测研究。通过对CNN架构进行改进,在EfficientNet中加入残差模块Res-EfficientNet,更精准的探测和识别烟雾。通过STRCF实现对烟雾的精度定位。为提高探测准确率,还考虑了烟雾偏振传输特性,如烟雾的扩散和半透明状态。为了能够更好地探测视频中的烟雾,将空间频率的能量作为滤波器的一维约束项,在基准数据集上进行了试验,试验结果表明,准确率提高了3%。