Pet Rescue System Using Computer Vision

Abstract

Animal shelters face challenges in managing lost or abandoned pets due to outdated methods such as paper records or simple databases. These systems limit real-time tracking, accurate identification, and effective health monitoring. This study presents a smart shelter management system that integrates machine learning and computer vision techniques, including facial recognition and object detection. The system addresses three key areas: unique animal identification, health monitoring through image analysis, and integrated digital management of pet records. By reducing manual workload and improving record accuracy, the proposed approach enhances both efficiency and animal care. Unlike earlier models that focused mainly on adoption prediction, this system works in real time, is scalable, and adaptable to shelter growth. Future developments, such as video-based monitoring, can further support disease prevention and collaboration across multiple shelters.

Introduction

Animal shelters often face challenges in managing rescued animals due to outdated systems, manual records, and lack of efficient tracking, leading to errors and delays. To address these issues, the proposed Smart Pet Rescue Shelter Management System uses Artificial Intelligence, specifically Computer Vision and Machine Learning, to automate animal identification, health monitoring, and data management.

The system applies techniques like facial recognition, object detection, and image-based disease analysis to improve accuracy and enable real-time monitoring. It centralizes all animal-related data into a digital platform, reducing manual workload and enhancing decision-making.

Existing research highlights gaps such as lack of automation, limited datasets, and absence of integrated systems combining identification and health monitoring. This project fills that gap by offering a unified, scalable solution.

The methodology involves capturing animal images, preprocessing them, and using deep learning models (e.g., MobileNet, ResNet) for identification and disease prediction. Results are stored in a centralized database and presented through a user-friendly interface.

Overall, the system improves efficiency, reliability, and animal welfare by enabling smarter, technology-driven shelter management.

Conclusion

This project presents a smart shelter management system that utilizes computer vision techniques to enhance the identification, monitoring, and care of rescued and abandoned animals. By integrating facial recognition, object detection, and image-based health analysis, the system improves accuracy, reduces manual effort, and enables real-time tracking along with early health detection. The scalable design supports efficient shelter operations and contributes to improved animal welfare. The proposed system demonstrates the potential of intelligent technologies to transform traditional shelter management into a more efficient and data-driven process.

References

[1] H. Patil and N. Ansari, “Smart Surveillance and Animal Care System Using IoT and Deep Learning,” in Proc. 3rd Int. Conf. Advances in Science & Technology (ICAST), Mumbai, India, Apr. 2020, doi: 10.2139/ssrn.3565274. [2] Y. Wu, N. Singhal, and P. Krähenbühl, “Efficient Visual Recognition: A Survey on Recent Advances,” Frontiers of Computer Science, vol. 16, no. 1, pp. 1–22, 2022, doi: 10.1007/s11633-022-1340-5. [3] Y. Azizi and L. Zaman, “Deep Learning Pet Identification Using Face and Body,” Information, vol. 14, no. 5, Art. no. 278, May 2023, doi: 10.3390/info14050278. [4] H. Lu, “Pet Adoption Status Prediction Based on Multiple Machine Learning Models,” in Proc. Int. Conf. Modern Logistics and Supply Chain Management (MLSCM), 2024.

Copyright

Copyright © 2026 Ms. Tejal Vallabh, Ms. Bhargavi Patel. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.