Veterinary clinics and animal shelters frequently receive inquiries about lost pets, yet despite microchipping and other identification methods, many lost cats remain unclaimed. This causes emotional distress for owners and burdens animal care services. However, very limited previous study has investigated the effectiveness of integrating computer vision with deep learning to support the public in reporting lost animals. This paper proposes an AI-driven framework that allows users to identify, and report lost cats simply by capturing a photo with their mobile phones. The framework integrates breed classification through a three-stage deep neural network, along with tag detection and text extraction. Empirical results show that the system can accurately capture visual features, including breed and tag information, improving the timeliness and effectiveness of lost pet reporting. The approach can also be extended to other animals.