Evaluating Deep Learning Models for African Wildlife Image Classification: From DenseNet to Vision Transformers

2025-07-28

AI Summary

This study addresses automatic classification of African wildlife images to support biodiversity monitoring. We systematically evaluate four deep learning architectures—DenseNet-201, ResNet-152, EfficientNet-B4, and ViT-H/14—via transfer learning on a localized African dataset comprising buffalo, elephant, rhinoceros, and zebra. To our knowledge, this is the first comparative study of CNNs and vision transformers in an African wildlife context. Results show ViT-H/14 achieves the highest accuracy (99%) but incurs prohibitive computational cost; DenseNet-201 attains 67% accuracy—the best among lightweight CNNs. Based on this trade-off analysis, we propose a model selection strategy balancing accuracy and deployment feasibility for resource-constrained field settings. Furthermore, we integrate the optimal CNN into a Hugging Face Gradio web application, enabling real-time, on-the-fly wildlife identification in野外 environments.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Deep Neural Architectures and Foundation Models
• Natural Language Processing: Text Classification & Sentiment Analysis

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Wildlife populations in Africa face severe threats, with vertebrate numbers declining by over 65% in the past five decades. In response, image classification using deep learning has emerged as a promising tool for biodiversity monitoring and conservation. This paper presents a comparative study of deep learning models for automatically classifying African wildlife images, focusing on transfer learning with frozen feature extractors. Using a public dataset of four species: buffalo, elephant, rhinoceros, and zebra; we evaluate the performance of DenseNet-201, ResNet-152, EfficientNet-B4, and Vision Transformer ViT-H/14. DenseNet-201 achieved the best performance among convolutional networks (67% accuracy), while ViT-H/14 achieved the highest overall accuracy (99%), but with significantly higher computational cost, raising deployment concerns. Our experiments highlight the trade-offs between accuracy, resource requirements, and deployability. The best-performing CNN (DenseNet-201) was integrated into a Hugging Face Gradio Space for real-time field use, demonstrating the feasibility of deploying lightweight models in conservation settings. This work contributes to African-grounded AI research by offering practical insights into model selection, dataset preparation, and responsible deployment of deep learning tools for wildlife conservation.

Problem

• Evaluate deep learning models for African wildlife image classification
• Compare performance and computational costs of various models
• Assess deployability of models in conservation settings

Innovation

• Transfer learning with frozen feature extractors
• Comparative study of DenseNet ResNet EfficientNet ViT
• Hugging Face Gradio for real-time field deployment