ABSTRACT
This chapter presents a repository driven approach to comprehensively map the effect of turbidity and lighting on the performance of image-based techniques. It describes the experimental set-up and the nature of the imagery in the repository and considers turbidity and lighting as these are the primary factors that influence the degree of visibility. The repository is called underwater lighting and turbidity image repository (ULTIR). It consists of images featuring various damage forms and material types that have been photographed under controlled lighting and turbidity levels, reflective of real-world operating conditions. The repository is partitioned into three categories according to the nature of the damage: cracks, surface damage, and 3D shape information. The chapter compares some of the main algorithmic approaches related to each category of ULTIR, namely: crack detection, surface damage detection, and 3D shape recovery.
