Transfer Learning for Wildlife Classification: Evaluating YOLOv8 against DenseNet, ResNet, and VGGNet on a Custom Dataset

TL;DR

This paper compares deep learning models for wildlife classification, finding YOLOv8 superior in accuracy and efficiency on a custom endangered species dataset, aiding conservation efforts.

Contribution

It evaluates transfer learning performance of YOLOv8 against DenseNet, ResNet, and VGGNet for wildlife classification, demonstrating YOLOv8's superior accuracy and efficiency.

Findings

YOLOv8 achieved 97.39% training accuracy.

Validation F1-score was 96.50%.

YOLOv8 outperformed other models in this task.

Abstract

This study evaluates the performance of various deep learning models, specifically DenseNet, ResNet, VGGNet, and YOLOv8, for wildlife species classification on a custom dataset. The dataset comprises 575 images of 23 endangered species sourced from reputable online repositories. The study utilizes transfer learning to fine-tune pre-trained models on the dataset, focusing on reducing training time and enhancing classification accuracy. The results demonstrate that YOLOv8 outperforms other models, achieving a training accuracy of 97.39% and a validation F1-score of 96.50%. These findings suggest that YOLOv8, with its advanced architecture and efficient feature extraction capabilities, holds great promise for automating wildlife monitoring and conservation efforts.