AI Smart Currency Detection System uses YOLOv8 running on Raspberry Pi 5 to identify and classify Indian currency notes in real time via Pi Camera Module. Image preprocessing—including resizing, noise removal, and brightness enhancement—feeds the trained YOLOv8 model, which performs feature extraction and currency classification. Recognition results are converted to speech via pyttsx3 and output through a speaker, enabling accessible currency identification for visually impaired users. Fork this diagram on Diagrams.so to customize the preprocessing pipeline, swap YOLOv8 for alternative models, or adapt it for other currency systems. The architecture demonstrates edge AI best practices: on-device inference eliminates cloud latency, reduces bandwidth, and ensures privacy while maintaining real-time performance on constrained hardware.