Offshore and marine engineering have developed rapidly in recent decades. Advances in construction materials and methods have paved the way for new forms of structures, such as wave energy converters, to be installed in deeper and more hostile conditions. Inevitably, however, newly installed structures will join the existing infrastructure stock, and the focus will shift toward monitoring and maintenance activities to ensure that these structures reach, or even surpass, their target lifespans (Wenzel, 2003).
The harsh and unrelenting marine environment means that offshore and marine structures are prone to aesthetic, functional, or structural degradation, which, over time, typically leads to a loss of serviceability at either a component or global level. Owners/managers must, therefore, inspect structures to ensure that they are safe, fit for service, and so that they can make more informed decisions when allocating resources toward the correction of deficiencies. This latter aspect has attracted a growing interest in recent times as the importance of life cycle optimization and related financial benefits continue to be recognized, especially in relation to marine structures (Schoefs et al., 2012). Given that many significant decisions are made based on inspection findings, it is important that inspections strive to provide accurate information that is reflective of the true condition of the structure.
This book highlights the value that image-processing can bring to the inspection process, especially in relation to, but not confined to, underwater inspections. The versatility of image-processing is conveyed through the wide range of applications presented in this book.
1.1Aim of This Book
The goal of this book is to serve as a comprehensive guide for structural engineers and inspectors on how to effectively integrate image-processing techniques into the inspection regime. Image-based inspections are comprised of several distinct, yet inexorably linked, stages, as illustrated in Figure 1.1. The success at each stage has a significant bearing on the overall success; failure to ensure that each stage is given due consideration will almost invariably result in a reduced quality of the inspection results. In this book, a holistic view of image-based inspections is adopted, and each stage is comprehensively addressed.
Figure 1.1Elements of image-based underwater inspections.
The three principal stages of imaging in underwater inspections are covered. These are:
On-site image acquisition procedures
Image-processing algorithms
Using the results from image-processing for further analysis with specialized engineering software
A firm grasp of these stages provides the building blocks necessary for readers to become conversant in using imaging in inspections. The first pillar sets out best practice guidelines for obtaining imagery that is well-suited for subsequent quantitative analysis.
For the second stage, a concise and relevant overview of the fundamentals of image-processing is presented. Following this, we delve into more advanced damage assessment algorithms. These algorithms are described in a step-by-step manner as the goal here is to give readers a thorough understanding of the algorithmic design choices and to equip readers with the skills necessary to modify algorithms for their purposes. MATLAB® scripts are provided, and readers have access to a large underwater image repository so that techniques can be run and tested on a collection of realistic image samples.
For the third and final stage, we look at how the information obtained from image analysis can be used in the broader context of Structural Health Monitoring (SHM). The material in this book is explained with the help of real-world case studies. It is hoped that these studies will give readers a glimpse of the potential application areas of image-processing.
This book is principally geared toward the inspection of offshore and marine structures; however, many of the concepts and techniques presented in this book can be applied to terrestrial/top-side inspections.
1.2Imaging in Inspections
Image-processing in underwater inspections is still a young field. Work supporting the development and implementation of image-processing methods has only begun to emerge relatively recently, and its position within the inspection framework is not well defined. The role of cameras is typically limited to taking photographs or video of instances of damage without any agreed protocol of image collection and subsequent interpretation within the inspection framework (Phares et al., 2004). The collected imagery is usually archived and is rarely ever used in a quantitative sense. However, there exists significant scope for further development in the domain of underwater SHM as, currently, the full potential of cameras is not being realized.
Image-processing is well-suited to complementing traditional visual inspection methods as opposed to completely replacing them. Visual inspections are the most common means of collecting data about the state of marine structures; however, they have some inherent limitations. They are affected by the ability of inspectors to observe and objectively record details of defects, and they are prone to considerations such as boredom, lapses in concentration, subjectivity, and fatigue, all of which contribute to greater variability and reduced accuracy (Dirksen et al., 2013; Estes & Frangopol, 2003). Image-processing provides a way of overcoming some of these shortcomings by making visual data a part of quantitative assessment.
1.2.1Advantages of image-processing as an inspection tool
Assessing the submerged part of marine structures introduces new challenges for inspectors. Many damage diagnostic tools that can be used on dry land cannot be readily adapted for underwater deployment. Additionally, only a limited amount of time can be spent underwater, especially when the inspection is being carried out by a diver rather than a remotely operated vehicle (ROV). This puts an emphasis on adopting expeditious data collection practices. Unlike other non-destructive testing (NDT) tools, cameras can acquire data in an efficient manner, they are easily adapted for underwater application, and they require only minimal training in their operation. Moreover, vision-based NDT tools are often the only practical way to detect certain damage indicators such as cracks, corrosion, or surface breaking defects.
Cameras are already used in almost all visual inspections. Typically, photographs are captured to include in the inspection report to accompany the inspector’s comments; however, the photographs are rarely exploited to their fullest potential in either a qualitative or a quantitative fashion. Adopting effective image-based techniques can provide accurate quantitative information with minimal human supervision to supplement visual inspection techniques and increase reliability. The quantitative nature of the data obtained from image analysis naturally lends itself to numerous applications—it is helpful for developing new damage models or strengthening existing ones, which are used to forecast the rate of propagation of damage as the structure continues to operate.
1.2.2Limitations
While image-based methods undoubtedly have the potential to be a convenient and useful underwater NDT tool, there are some limitations as well. First, the technology is only appropriate for assessing visible damage forms and surface breaking defects. It is not possible to assess internal defects or damage that is masked by bio-fouling without first cleaning the structure. Second, the poor underwater visibility conditions diminish the ability of cameras, and subsequent image-processing techniques, to effectively identify instances of damage. Underwater images are affected by light attenuation, scattering, color absorption, suspended particles, and air bubbles, as identified by Massot-Campos and Oliver-Codina (2015), who carried out a survey on optical sensors and methods for 3D reconstruction in underwater environments....