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

Lukman Aliyu, Umar Muhammad, Bikisu Ismail, Almustapha Wakili, Seid Yimam, Shamsuddeen Muhammad, Mustapha Abdullahi
DLI 2025 Research Track, PMLR 302:1-14, 2026.

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. Keywords:Image Classification, DenseNet, African wildlife, Computer Vision, Deep learning.

Cite this Paper

BibTeX
@InProceedings{pmlr-v302-aliyu26a, title = {Evaluating Deep Learning Models for African Wildlife Image Classification: From DenseNet to Vision Transformers}, author = {Aliyu, Lukman and Muhammad, Umar and Ismail, Bikisu and Wakili, Almustapha and Yimam, Seid and Muhammad, Shamsuddeen and Abdullahi, Mustapha}, booktitle = {DLI 2025 Research Track}, pages = {1--14}, year = {2026}, editor = {Haddad, Hatem and Kahira, Albert Njoroge and Bourhim, Sofia and Olatunji, Iyiola Emmanuel and Makhafola, Lesego and Mwase, Christine}, volume = {302}, series = {Proceedings of Machine Learning Research}, month = {17--22 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v302/main/assets/aliyu26a/aliyu26a.pdf}, url = {https://proceedings.mlr.press/v302/aliyu26a.html}, 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. Keywords:Image Classification, DenseNet, African wildlife, Computer Vision, Deep learning.} }
Endnote
%0 Conference Paper %T Evaluating Deep Learning Models for African Wildlife Image Classification: From DenseNet to Vision Transformers %A Lukman Aliyu %A Umar Muhammad %A Bikisu Ismail %A Almustapha Wakili %A Seid Yimam %A Shamsuddeen Muhammad %A Mustapha Abdullahi %B DLI 2025 Research Track %C Proceedings of Machine Learning Research %D 2026 %E Hatem Haddad %E Albert Njoroge Kahira %E Sofia Bourhim %E Iyiola Emmanuel Olatunji %E Lesego Makhafola %E Christine Mwase %F pmlr-v302-aliyu26a %I PMLR %P 1--14 %U https://proceedings.mlr.press/v302/aliyu26a.html %V 302 %X 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. Keywords:Image Classification, DenseNet, African wildlife, Computer Vision, Deep learning.
APA
Aliyu, L., Muhammad, U., Ismail, B., Wakili, A., Yimam, S., Muhammad, S. & Abdullahi, M.. (2026). Evaluating Deep Learning Models for African Wildlife Image Classification: From DenseNet to Vision Transformers. DLI 2025 Research Track, in Proceedings of Machine Learning Research 302:1-14 Available from https://proceedings.mlr.press/v302/aliyu26a.html.

Related Material