Unmanned Aerial Vehicles for early detection & warning of jellyfish & saeweed near coastal nuclear power plants
Lead Research Organisation:
CRANFIELD UNIVERSITY
Department Name: School of Water, Energy and Environment
Abstract
To couple UAV technology and current approaches in statistical high resolution image analysis with robust bloom metric estimation for the early detection and warning of jellyfish and seaweed presence.
To develop a set of algorithms for the automated detection of jellyfish and seaweed blooms from high resolution UAV aerial imagery.
To collate this detection capability with complementary data sets (e.g., satellite) to advance bloom behaviour understanding.
Extending work of 1) and 2) to develop a UAV based framework for early detection of jellyfish and seaweed blooms near nuclear power plants.
To develop a set of algorithms for the automated detection of jellyfish and seaweed blooms from high resolution UAV aerial imagery.
To collate this detection capability with complementary data sets (e.g., satellite) to advance bloom behaviour understanding.
Extending work of 1) and 2) to develop a UAV based framework for early detection of jellyfish and seaweed blooms near nuclear power plants.
Publications
Mcilwaine B
(2019)
Using 1st Derivative Reflectance Signatures within a Remote Sensing Framework to Identify Macroalgae in Marine Environments
in Remote Sensing
Mcilwaine B
(2021)
JellyNet: The convolutional neural network jellyfish bloom detector
in International Journal of Applied Earth Observation and Geoinformation
Description | - The wavelengths used to detect and species discriminate potentially damaging coastal macroalgaes have been identified. This will be used to optimise sensor selection for detection in a marine environment. - A jellyfish bloom detection model has been developed that can accept an input of a picture taken from a UAV/drone, and with ~97% accuracy it can inform as to whether a jellyfish bloom is present or not |
Exploitation Route | The current findings could be taken forward to help prevent or reduce the millions of dollars of annual damage that marine ingress events cause around the world. Deployment of the early warning detection system could be potentially deployed by any industry that suffers from jellyfish bloom or macroalgae occurrence. Coastal industries such as nuclear power stations, salmon and other farm fisheries, beach tourist attraction sites and desalination plants could all benefit from the outcomes of the award. Production of a full independent and automated detection system when combined with an autonomously launched and flown UAV system. |
Sectors | Energy Environment Leisure Activities including Sports Recreation and Tourism Security and Diplomacy Other |
Title | Jellyfish bloom detection model |
Description | A convolutional neural network to detect jellyfish blooms from aerial imagery |
Type Of Material | Computer model/algorithm |
Year Produced | 2020 |
Provided To Others? | No |
Impact | To help prevent damages and reduced revenues for coastal industries |