Object detection is an important application of computer vision that enables automated identification and localization of objects in images and videos. This study presents a deep learning-based Cat and Dog Detection System using YOLOv8 (You Only Look Once version 8). The main objective of the proposed system is to accurately detect and distinguish cats and dogs from input images using bounding boxes and class labels. A dataset containing images of cats and dogs is used for training and validation. The images are preprocessed and annotated before being provided to the YOLOv8 model for learning. The model is trained using Google Colab with GPU acceleration, allowing efficient deep learning model training. YOLOv8 is selected because it provides a suitable balance between detection accuracy and processing speed, making it useful for real-time object detection applications. The performance of the trained model is evaluated using important metrics such as Precision, Recall, mAP@50, and mAP@50–95. The experimental results show that the proposed model achieves an overall Precision of 92.3%, Recall of 84.0%, mAP@50 of 89.9%, and mAP@50–95 of 73.0%. The model demonstrates effective detection performance for both cat and dog classes under different image conditions. The proposed system can be applied in pet monitoring, animal recognition, smart surveillance, automated image analysis, and educational computer vision applications. The results demonstrate that YOLOv8 provides an efficient and reliable approach for automatic cat and dog detection, while future improvements can focus on increasing robustness under occlusion, poor lighting, and complex background.

Nishalini M, Shobhik R, Shiva Prakash M, Prathikshana E K, Dr. Amsaveni M,"YOLOv8 – Based Cat and Dog Detection Using Deep Learning" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(4): Page(492-496) July - August 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.