Abstract
Reliable and efficient monitoring of animals in their natural habitat is crucial
to inform management and conservation decisions. Automation of the
classification of species is important since manually classifying animals is
expensive, monotonous, and time-consuming. Current studies have shown
that Deep Neural Networks (DNNs) are effective at animal identification.
However, there is little research on distinguishing species like antelopes from
each other. In this dissertation, an implementation of Deep Neural Networks
in the classification of antelopes is proposed. The model managed to achieve
a 93% Top-1 accuracy on the dataset which is higher than most of the models
discussed in the literature. This shows that DNNs can be used to classify antelopes
with high accuracy. This work also implemented out-of-distribution
modelling using softmax confidence and managed to achieve an accuracy of
74% after introducing an out-of-distribution sample to the test dataset. Further
work can be done to further classify species into categories like the lesser
Kudu or Greater Kudu which is more complex than the work of this study.