We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
A Survey of Seafloor Characterization and Mapping Techniques
Summary
This survey reviews the main techniques used to map and characterize the deep seafloor, including optical and acoustic sensing methods. The paper covers approaches ranging from traditional statistics to deep learning, highlighting current limitations in underwater imaging and pointing to future research directions for better understanding seabed ecosystems.
The deep seabed is composed of heterogeneous ecosystems, containing diverse habitats for marine life. Consequently, understanding the geological and ecological characteristics of the seabed’s features is a key step for many applications. The majority of approaches commonly use optical and acoustic sensors to address these tasks; however, each sensor has limitations associated with the underwater environment. This paper presents a survey of the main techniques and trends related to seabed characterization, highlighting approaches in three tasks: classification, detection, and segmentation. The bibliography is categorized into four approaches: statistics-based, classical machine learning, deep learning, and object-based image analysis. The differences between the techniques are presented, and the main challenges for deep sea research and potential directions of study are outlined.
Sign in to start a discussion.
More Papers Like This
A Review of Underwater Image Enhancement and Restoration Techniques Based on Gan
This review examines underwater image enhancement and restoration technologies based on generative adversarial networks (GANs), assessing challenges arising from the underwater imaging environment and evaluating GAN-based methods for improving image quality in support of marine operations and seabed resource development.
Machine learning in marine ecology: an overview of techniques and applications
This overview examines how machine learning techniques are being applied across marine ecology, covering data types from satellite imagery and acoustics to underwater images and genomic data. Researchers built a database of roughly 1,000 publications to map which techniques work best for different marine research questions. The study highlights that growing data volumes and computing power are making machine learning an increasingly essential tool for understanding ocean ecosystems.
Robust and Fair Undersea Target Detection with Automated Underwater Vehicles for Biodiversity Data Collection
Researchers developed a robust and fair underwater target detection system for automated underwater vehicles (AUVs) to support marine biodiversity data collection, addressing unique challenges of subsea imaging compared to above-ground remote sensing.
Deep-Sea Debris Identification Using Deep Convolutional Neural Networks
Researchers developed a deep convolutional neural network classifier to identify and distinguish deep-sea debris from seafloor imagery, demonstrating that automated AI-based detection can support submersible clean-up operations targeting marine debris in deep-sea environments.
A Characterisation of Benthic Currents from Seabed Bathymetry: An Object-Based Image Analysis of Cold-Water Coral Mounds
Researchers used object-based image analysis of multibeam sonar data to characterize seafloor bedforms and infer benthic current directions and speeds around cold-water coral mounds. The approach automatically classifies seafloor features and links them to oceanographic conditions. Better mapping of deep-sea currents helps explain how marine organisms and particles — including potential plastic debris — are transported along the seafloor.