Year: 2026
Type: article
Abstract: Wildlife monitoring and biodiversity conservation require accurate and timely identification of animal species in natural habitats. Traditional manual observation techniques are labor-intensive, time-consuming, and prone to human error, making them unsuitable for large-scale ecological analysis. This paper presents an Animal Species Detection System using YOLOv8, a state-of-the-art real-time object detection model. The proposed system automatically detects, classifies, and localizes multiple animal species from images and video streams with high accuracy and speed. The system is trained on annotated wildlife datasets covering diverse species in varying environmental conditions. Advanced preprocessing including mosaic augmentation, resizing, and normalization improves generalization. Experimental results demonstrate superior detection accuracy, better localization capability, and faster inference speed compared to conventional CNN-based methods, offering a scalable and intelligent solution for wildlife conservation and ecological management.
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
SDG: Life on land
Open Access status: diamond
OpenAlex ID: W7217461983