Currency Detector — Unattended-Cash Monitoring

What it does

A cash-handling security system: an open-vocabulary detector (OWL-ViT) finds banknotes in an RTSP camera stream, OCR reads the note to classify the currency, and each note is tracked over time. If money stays unattended longer than a configurable threshold (5 minutes by default), the system sends an email alert and logs the event to a web dashboard with date filtering and CSV export.

How it works

1. Step 1. OWL-ViT, an open-vocabulary detector, finds banknotes in the RTSP stream using text prompts such as “AED banknote” — no custom training required.
2. Step 2. OCR (English and Arabic) reads the note to classify it as AED, USD or EUR.
3. Step 3. Each note is tracked frame to frame (IoU-based), so the system knows how long it has been lying there.
4. Step 4. Past the threshold, an SMTP email alert goes out and the event — with duration, currency and confidence — appears on the dashboard.

Where it's used

• Cash counters and tills
• Bank branches
• Exchange houses
• Back-office cash rooms

Tech stack

• Python
• OWL-ViT (Transformers)
• OCR
• Flask dashboard
• RTSP

Build something like this

This project is part of our computer vision development and a fixed-price computer vision pilot. Tell us your use case and we will reply within one business day with a first step, a price and a date.