Cameron Trotter
Modelling cetacean populations by photo-id using deep learning.
Modelling cetacean populations by photo-id using deep learning
Email: [email protected]
Supervisors
- School of Engineering
- School of Computing
- School of Natural and Environmental Sciences
Project description
Dolphins, whales and porpoises are cetaceans. Modelling cetacean population dynamics is paramount for effective conservation and management. We need robust data for the design and implementation of conservation strategies. We also need it to assess the risks presented by anthropogenic activity. Such activity includes offshore wind turbines and commercial fishing. Cetaceans make prime candidates for modelling ecosystem change. They reflect the current state of the ecosystem and respond to change across different spatial and temporal scale.
The global climate is changing and urbanisation of coastal areas is intensifying. It is imperative to develop methodologies for quick and effective assessment of the biological and ecological impact of rising sea temperatures, pollution, and habitat degradation. We can achieve this through modelling the population, behaviour, and health of large marine species.
Methodologies of cetacean research includes photo identification (photo-id). Photo-id involves collecting photographic data and identifying individuals based on unique permanent markings. It has been used for more than 40 years for modelling cetacean population dynamics and ecology.
Current identification techniques for cetaceans rely heavily on experts manually identifying individuals. This is often costly due to the number of person-hours required for identification. There is also a large potential for error due to issues such as observer fatigue. Further, individual identification of dolphins within a species is time consuming because of the nature of the task. Progressively more data is being collected during fieldwork through increased use of technology. Thus, there is an urgent need for an automatic system for quick identification with reduced error rates.
This project addresses these limitations by applying the techniques and computational power of deep learning to the field of marine biology. It brings together a multidisciplinary team from the Schools of Engineering and Computing, and the School of Natural and Environmental Science’s Marine MEGAfauna Lab.
Publications
- Trotter C, Atkinson G, Sharpe M, McGough AS, Wright N, Berggren P. The Northumberland Dolphin Dataset: A Multimedia Individual Cetacean Dataset for Fine-Grained Categorisation Presented at FGVC6: The Sixth Workshop on Fine-Grained Visual Categorization 2019. USA: Long Beach.
Interests
Computer vision, deep learning, marine conservation.
Qualifications
- MComp, Newcastle University (2017)
- PGDip Cloud Computing for Big Data, Newcastle University (2018)