International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 186 - Number 11 |
Year of Publication: 2024 |
Authors: Shailendra Singh Kathait, Vaibhav Singh, Ashish Kumar |
![]() |
Shailendra Singh Kathait, Vaibhav Singh, Ashish Kumar . Individual Tiger Identification using Transfer Learning. International Journal of Computer Applications. 186, 11 ( Mar 2024), 46-50. DOI=10.5120/ijca2024923469
This paper describes the methods to classify images of tigers into their respective classes using deep learning models. The classes here represent the individual tigers themselves. With the use of motion activated cameras it has been possible to get a huge number of images of animals in their natural habitat but to make any use of this data in unique animal identification has been a challenge, which is aimed to be solved in this paper. There are techniques available to identify species of animals from the images, but classifying a particular animal specie into its individuals yet remains a tough task. The method used has a pipeline that involves the YOLOv8 model and EfficientNetB3 model with transfer learning to classify an image of tiger while working with images of 98 unique tigers. Out of 192 tigers whose images were available, the paper included only 98 in the model after setting a threshold of availability of 15 images at least for each tiger.