A5090918085- Coral and Marine Ecosystems Studies
- Marine and fisheries research
- Environmental Monitoring and Data Management
- Machine Learning and Data Classification
- Seismic Imaging and Inversion Techniques
- Water Quality Monitoring Technologies
- Service-Oriented Architecture and Web Services
- Underwater Vehicles and Communication Systems
- Distributed and Parallel Computing Systems
- Remote-Sensing Image Classification
- Model Reduction and Neural Networks
- Calibration and Measurement Techniques
- Climate variability and models
- Marine Biology and Ecology Research
- Image and Signal Denoising Methods
- Ichthyology and Marine Biology
CSIRO Oceans and Atmosphere
2011-2023
University of Tasmania
2011-2013
Abstract Protecting deep‐sea coral‐based vulnerable marine ecosystems (VMEs) from human impacts, particularly bottom trawling, is a major conservation challenge in world oceans. Management processes for these are weakened by key uncertainties that could be substantially addressed having much greater volumes of quantitative image‐derived data detail the distribution and abundance coral reefs nature impacts upon them. Considerably available if resource costs image annotation reduced. In this...
Protection of vulnerable marine ecosystems (VME) is a critical goal for conservation. Yet, in many deep-sea settings, where quantitative data are typically sparse, it challenging to correctly identify the location and size VMEs. Here we assess sensitivity method coral reef VMEs based on bottom cover abundance stony Solenosmilia variabilis deep seamounts, using image from 2018 large survey off Tasmania, Australia. Whilst there was some detectable influence varying live heads, distribution not...
Imaging is increasingly used to capture information on the marine environment thanks improvements in imaging equipment, devices for carrying cameras and data storage recent years. In that context, biologists, geologists, computer specialists end-users must gather discuss methods procedures optimising quality quantity of collected from images. The 4 th Marine Workshop was organised 3-6 October 2022 Brest (France) a hybrid mode. More than hundred participants were welcomed person about 80...
A multiscale maximum entropy method (MEM) for image deconvolution is implemented and applied to MODIS (moderate resolution imaging spectroradiometer) data remove instrument point-spread function (PSF) effects. The implementation utilizes three efficient computational methods: a fast Fourier transform convolution, wavelet decomposition an algorithm gradient step-size estimation that together enable rapid deconvolution. Multiscale uses transforms implicitly include image's two-dimensional...
We successfully utilized the SQUIDLE+ online platform to crowdsource a dataset comprising 1.7 million annotations of seabed biota, in over 150,000 images from 325 expeditions gathered Australian waters. This facilitated training four competitive ML models, significantly reducing need for manual image annotation monitoring programs. Utilising application interface, we combined low-level different schemes into three high-level essential ocean variables (EOV) and one species-specific label,...
This paper describes a correction method for Fast Fourier Transform (FFT) convolution that limits boundary contamination artefacts resulting from padding methods. The proposed makes single data-driven condition assumption and only uses information contained within the original input signal to produce consistent results maintain data integrity. An analysis of algorithm shows it performs identically standard approach with discernible differences being resolved at level machine rounding errors....