Now showing 1 - 10 of 20
  • Publication
    Texture Analysis Based Damage Detection of Ageing Infrastructural Elements
    To make visual data a part of quantitative assessment for infrastructure maintenance management, it is important to develop computer-aided methods that demonstrate efficient performance in the presence of variability in damage forms, lighting conditions, viewing angles, and image resolutions taking into account the luminous and chromatic complexities of visual data. This article presents a semi-automatic, enhanced texture segmentation approach to detect and classify surface damage on infrastructure elements and successfully applies them to a range of images of surface damage. The approach involves statistical analysis of spatially neighboring pixels in various color spaces by defining a feature vector that includes measures related to pixel intensity values over a specified color range and statistics derived from the Grey Level Co-occurrence Matrix calculated on a quantized grey-level scale. Parameter optimized non-linear Support Vector Machines are used to classify the feature vector. A Custom-Weighted Iterative model and a 4-Dimensional Input Space model are introduced. Receiver Operating Characteristics are employed to assess and enhance the detection efficiency under various damage conditions.
      551Scopus© Citations 96
  • Publication
    Semantic Segmentation of Underwater Imagery Using Deep Networks Trained on Synthetic Imagery
    Recent breakthroughs in the computer vision community have led to the emergence of efficient deep learning techniques for end-to-end segmentation of natural scenes. Underwater imaging stands to gain from these advances, however, deep learning methods require large annotated datasets for model training and these are typically unavailable for underwater imaging applications. This paper proposes the use of photorealistic synthetic imagery for training deep models that can be applied to interpret real-world underwater imagery. To demonstrate this concept, we look at the specific problem of biofouling detection on marine structures. A contemporary deep encoder–decoder network, termed SegNet, is trained using 2500 annotated synthetic images of size 960 × 540 pixels. The images were rendered in a virtual underwater environment under a wide variety of conditions and feature biofouling of various size, shape, and colour. Each rendered image has a corresponding ground truth per-pixel label map. Once trained on the synthetic imagery, SegNet is applied to segment new real-world images. The initial segmentation is refined using an iterative support vector machine (SVM) based post-processing algorithm. The proposed approach achieves a mean Intersection over Union (IoU) of 87% and a mean accuracy of 94% when tested on 32 frames extracted from two distinct real-world subsea inspection videos. Inference takes several seconds for a typical image.
      353Scopus© Citations 41
  • Publication
    Quantifying the Health Impacts of Active Travel: Assessment of Methodologies
    (Taylor & Francis, 2015-05-12) ; ;
    In the past several years, active travel (walking and cycling) has increasingly been recognized as an effective means of improving public health by increasing physical activity and by avoiding the negative externalities of motorized transport. The impacts of increased active travel on mortality and morbidity rates have been quantified through a range of methodologies. In this study, the existing publications in this field of research have been reviewed to compare and contrast the methodologies adapted and to identify the key considerations and the best practices. The publications were classified in terms of the health summary outcomes and exposure variables considered, the model structures used in the studies and the impact of these choices on the results. Increased physical activity was identified as the most important determinant of the health impacts of active travel but different ways of quantifying these health impacts can lead to substantial differences in the scale of the impact. Further research is required into the relationship between increased physical activity and health effects in order to reach consensus on the most reliable modelling approach for this important determinant of benefits. Critical discussions on other exposure variables have also been provided to ascertain best practices. Additionally, a logical flow of the modelling processes (and their variations) has also been illustrated which can be followed for developing future studies into the health impacts of active travel.
      581Scopus© Citations 46
  • Publication
    High dynamic range image processing for non-destructive-testing
    (Taylor & Francis, 2011-10-17) ; ;
    This paper proposes High Dynamic Range (HDR) imaging as a protocol for nondestructive-testing for the first time. HDR imaging protocol has been experimentally validated on images of pitting corrosion in this paper and has been further applied to isolate a range of image backgrounds that arise out of a number of environmental conditions. The superiority of HDR imaging over a standard imaging process has been demonstrated. The results indicate that the proposed protocol is not limited by the examples considered in this paper and can readily be applied to a number of infrastructure maintenance management related applications. The protocol complements and improves standard visual inspection techniques and expert opinions. This approach immediately lends itself to further mathematical and statistical analyses, both qualitative and quantitative.
      399Scopus© Citations 5
  • Publication
    Perception of safety of cyclists in Dublin City
    In recent years, cycling has been recognized and is being promoted as a sustainable mode of travel. The perception of cycling as an unsafe mode of travel is a significant obstacle in increasing the mode share of bicycles in a city. Hence, it is important to identify and analyze the factors which influence the safety experiences of the cyclists in an urban signalized multi-modal transportation network. Previous researches in the area of perceived safety of cyclists primarily considered the influence of network infrastructure and operation specific variables and are often limited to specific locations within the network. This study explores the factors that are expected to be important in influencing the perception of safety among cyclists but were never studied in the past. These factors include the safety behavior of existing cyclists, the users of other travel modes and their attitude toward cyclists, facilities and network infrastructures applicable to cycling as well as to other modes in all parts of an urban transportation network. A survey of existing cyclists in Dublin City was conducted to gain an insight into the different aspects related to the safety experience of cyclists. Ordered Logistic Regression (OLR) and Principal Component Analysis (PCA) were used in the analysis of survey responses. This study has revealed that respondents perceive cycling as less safe than driving in Dublin City. The new findings have shown that the compliance of cyclists with the rules of the road increase their safety experience, while the reckless and careless attitudes of drivers are exceptionally detrimental to their perceived safety. The policy implications of the results of analysis are discussed with the intention of building on the reputation of cycling as a viable mode of transportation among all network users.
    Scopus© Citations 100  580
  • Publication
    Health impacts of cycling in Dublin on individual cyclists and on the local population
    There is an emerging consensus that personal and societal health benefits in cycling largely outweigh the risks. However, there exists limited research into the health impacts experienced by individuals who take up cycling or the marginal societal benefits resulting from incremental uptake of cycling. This paper models and estimates the health impacts of individuals in Dublin taking up cycling. The paper utilizes the 2011 census data of Ireland and a Burden of Disease (BOD) approach is used to estimate health impacts on the individuals taking up cycling for their regular commute and on the rest of the local population separately. The health impact to an individual changing from private car to cycling ranged from a benefit of 0.033 Disability Adjusted Life Years (DALYs)/year to a loss of 0.003 DALYs/year. The marginal health impact to the local population ranged from no change to a benefit of 0.006 DALYs/year. Increases in cycling have a consistently positive impact on the health of the local population, regardless of the current modal split. The net expected health impacts to the individual cyclists are also positive in most cases. However, for some individuals in the 20–29 age group, the expected health impact may be small to negative, mainly due to a higher traffic collision risk. Where total impacts of scenarios are modelled the potential negative health impacts to some individuals may be masked by the overall positive health benefits of cycling to the local population. When promoting cycling as an alternative to driving to improve population health impacts, the risks to some cyclists should be managed and mitigated through safe road systems approaches.
      343Scopus© Citations 14
  • Publication
    Short-term forecasting of bicycle traffic using structural time series models
    Short term forecasting algorithms are widely used for prediction of vehicular traffic flows for adaptive traffic management. However, despite the increasing interest in the promotion of cycling in cities, little research has been carried out into the use of traffic forecasting algorithms for bicycle traffic. Structural time series models allow the various components of a time series such as level, seasonal and regression effects to be modelled separately to allow analysis of previous trends and forecasting. In this paper, a case study at a segregated bicycle lane in Dublin, Ireland was performed to test the forecasting accuracy of structural time series models applied to continuous observations of cyclist traffic volumes. It has been shown that the proposed models can produce accurate peak period forecasts of cyclist traffic volumes at both 1 hour and fifteen minute resolution and that the percentage errors are lower for hourly forecasts. The inclusion of weather metrics as explanatory variables had varying effects on the forecasting accuracies of the models. These results directly aid the design of traffic signal control systems accommodating cyclists.
      310Scopus© Citations 8
  • Publication
    Health, environmental and travel cost impacts of urban cycling in Dublin, Ireland
    (ICE Publishing, 2019-03-18) ; ;
    Cycling as a mode of transport avoids the negative external costs of driving such as air pollution, carbon dioxide emissions and noise and can also reduce the public health costs associated with physical inactivity. However, increased cycling may also have disadvantages such as increased exposure to air pollution and risk of traffic collisions. A number of studies have developed methods to quantify these health and environmental impacts and have shown that the overall impact of increased cycling is positive. However, while minimising travel costs is traditionally the main objective of transport planners, these studies have not included travel cost impacts in their analyses. In this study, the impacts of a modal shift towards cycling are quantified, taking into account health, environmental and travel cost impacts. It was found that the health and environmental impacts of increased cycling in Dublin, Ireland, would be strongly positive, mainly due to the health benefits of physical activity. When travel costs are also included in the analysis, the central estimate of net impact remains positive but the uncertainty increases considerably. This underscores the importance of the transport and health sectors working together to maximise the social welfare resulting from transport projects.
      416Scopus© Citations 4
  • Publication
    Regionally enhanced multiphase segmentation technique for damaged surfaces
    Imaging‐based damage detection techniques are increasingly being utilized alongside traditional visual inspection methods to provide owners/operators of infrastructure with an efficient source of quantitative information for ensuring their continued safe and economic operation. However, there exists scope for significant development of improved damage detection algorithms that can characterize features of interest in challenging scenes with credibility. This article presents a new regionally enhanced multiphase segmentation (REMPS) technique that is designed to detect a broad range of damage forms on the surface of civil infrastructure. The technique is successfully applied to a corroding infrastructure component in a harbour facility. REMPS integrates spatial and pixel relationships to identify, classify, and quantify the area of damaged regions to a high degree of accuracy. The image of interest is preprocessed through a contrast enhancement and color reduction scheme. Features in the image are then identified using a Sobel edge detector, followed by subsequent classification using a clustering‐based filtering technique. Finally, support vector machines are used to classify pixels which are locally supplemented onto damaged regions to improve their size and shape characteristics. The performance of REMPS in different color spaces is investigated for best detection on the basis of receiver operating characteristics curves. The superiority of REMPS over existing segmentation approaches is demonstrated, in particular when considering high dynamic range imagery. It is shown that REMPS easily extends beyond the application presented and may be considered an effective and versatile standalone segmentation technique.
      536Scopus© Citations 49
  • Publication
    A Stereo-Matching Technique for Recovering 3D Information from Underwater Inspection Imagery
    Underwater inspections stand to gain from using stereo imaging systems to collect three-dimensional measurements. Although many stereo-matching algorithms have been devised to solve the correspondence problem, that is, find the same points in multiple images, these algorithms often perform poorly when applied to images of underwater scenes due to the poor visibility and the complex underwater light field. This article presents a new stereo-matching algorithm, called PaLPaBEL (Pyramidal Loopy Propagated BELief) that is designed to operate on challenging imagery. At its core, PaLPaBEL is a semiglobal method based on a loopy belief propagation message passing algorithm applied on a Markov random field. A pyramidal scheme is adopted that enables wide disparity ranges and high-resolution images to be handled efficiently. For performance evaluation, PaLPaBEL is applied to underwater stereo images captured under various visibility conditions in a laboratory setting, and to synthetic imagery created in a virtual underwater environment. The technique is also demonstrated on stereo images obtained from a real-world inspection. The successful results indicate that PaLPaBEL is well suited for underwater application and has value as a tool for the cost-effective inspection of marine structures.
    Scopus© Citations 20  743