-, Preena Darshini (2024) Assessing the Efficacy of EfficientNet, Inception, and ResNet for Wildlife Species Identification. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (3MB) | Preview |
Abstract
In the current scenario, there is a lot of attention in the field of wildlife conservation to identify wildlife species in order to easily monitor biodiversity and aid ecologists in conservation efforts using deep learning techniques. This document discusses the use of EfficientNet, Inception, and ResNet to identify animal species on the iWildCam 2019 dataset. Challenges like class imbalance are handled. The models are trained and evaluated. The predictions from these models are combined with an ensemble to boost the performance. The ResNet model was fine-tuned to improve performance. The ensemble approach yielded an accuracy of 55.1%. The potential and limitations of these models are discussed with respect to the field of ecological research. Future work has also been suggested to improve the model accuracy which can handle complex imbalanced datasets.
Item Type: | Thesis (Masters) |
---|---|
Supervisors: | Name Email Haycock, Barry UNSPECIFIED |
Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science S Agriculture > SF Animal culture |
Divisions: | School of Computing > Master of Science in Data Analytics |
Depositing User: | Ciara O'Brien |
Date Deposited: | 06 Aug 2025 14:27 |
Last Modified: | 06 Aug 2025 14:27 |
URI: | https://norma.ncirl.ie/id/eprint/8446 |
Actions (login required)
![]() |
View Item |