Reducing Label Dependency for Underwater Scene Understanding: A Survey of Datasets, Techniques and Applications

TL;DR

This survey focuses on approaches which reduce dependency on human expert input, while reviewing the prior and related approaches to position these works in the wider field of underwater perception and examines available datasets and platforms to identify gaps, barriers, and opportunities for automating underwater surveys.

Abstract

Underwater surveys provide long-term data for informing management strategies, monitoring coral reef health, and estimating blue carbon stocks. Advances in broad-scale survey methods, such as robotic underwater vehicles, have increased the range of marine surveys but generate large volumes of imagery requiring analysis. Computer vision methods such as semantic segmentation aid automated image analysis, but typically rely on fully supervised training with extensive labelled data. While ground truth label masks for tasks like street scene segmentation can be quickly and affordably generated by non-experts through crowdsourcing services like Amazon Mechanical Turk, ecology presents greater challenges. The complexity of underwater images, coupled with the specialist expertise needed to accurately identify species at the pixel level, makes this process costly, time-consuming, and heavily dependent on domain experts. In recent years, some works have performed automated analysis of underwater imagery, and a smaller number of studies have focused on weakly supervised approaches which aim to reduce the expert-provided labelled data required. This survey focuses on approaches which reduce dependency on human expert input, while reviewing the prior and related approaches to position these works in the wider field of underwater perception. Further, we offer an overview of coastal ecosystems and the challenges of underwater imagery. We provide background on weakly and self-supervised deep learning and integrate these elements into a taxonomy that centres on the intersection of underwater monitoring, computer vision, and deep learning, while motivating approaches for weakly supervised deep learning with reduced dependency on domain expert data annotations. Lastly, the survey examines available datasets and platforms, and identifies gaps, barriers, and opportunities for automating underwater surveys.

Figures (20)

• Figure 1: Examples of variation in underwater imagery. Images show differences in camera quality, lighting, turbidity, camera viewpoint and resolution.
• Figure 2: An example image labelled with sparse random points in the Coral Point Count with Excel extensions (CPCe) software kohler2006coral.
• Figure 3: Weakly Supervised Underwater Image Analysis aims to reduce dependency on domain expert labelling, and operates at the intersection of underwater environmental monitoring, computer vision and deep learning. This survey reviews relevant prior approaches in other fields as well as specialised implementations for analysis of underwater imagery. This figure depicts the main sections of this survey and how the different facets are connected.
• Figure 5: Examples of two important coastal ecosystems: seagrass meadows. Seagrass meadows (a) are communities of marine flowering plants which play a role in stabilising sediment, supporting marine life and sequestering blue carbon. Coral reefs (b) are living structures that support a rich diversity of marine species and act as natural protective barriers for coastlines.
• Figure 6: Examples of underwater monitoring tasks, including: a) quantifying fish populations, b) detecting the Stone Coral Tissue Loss Disease (SCTLD), c) mapping the Crown-of-Thorns Starfish (COTS), and d) performing re-identification of manta rays.