A multi-modal, non-linear, multi-scale-space model for precise mapping and monitoring of estuary health

Managing coastal regions is crucial for ensuring the continued function of coastal ecosystems that support life on the planet. Effective monitoring and mapping are logistically difficult tasks due to budget, labour and time constraints. The result is infrequent monitoring and sparser data.

Modern autonomous vehicles such as UAVs and AUVs offer the flexibility for data acquisition that is missing in current monitoring approaches but require new approaches to process the different data they can obtain.

This project aims to provide scientific knowledge and tools to enhance monitoring efforts in scale, density, precision and cost with imaging sensors and autonomous vehicle platforms. More specifically, developing new models and algorithms to capture the multiscale relationships of different features when imaged by optical sensors at different distances using the the latest advances in computer vision and machine learning.

This project will specifically look at monitoring mussel bed restoration efforts using AUVs. The end goal is to take a step closer to the autonomous monitoring of key ecological sites around New Zealand and providing marine ecologists, regional councils and the public with new tools and data to assist in preserving New Zealand’s beautiful coasts.