AI pallet and product counting with your existing cameras
Nara counts items, parts, products and pallets from IP or industrial cameras, and object tracking makes sure each one is counted exactly once. Totals appear live on a dashboard and can go straight to your MES or ERP, all processed on your own hardware.
[ 01 ]The problem
Manual counts of output, pallets, parts or batches are slow, error-prone and never real-time.
- 01inputIP or industrial camera over the line, dock or storage area
- 02processObject detection and tracking with a counting line or zone, so each item is counted once
- 03actionCounts logged and shown live, sent to MES or ERP, with batch-complete or short-batch alerts
How it works
Draw a counting line or zone on the camera view. Nara detects each item, tracks it across frames and counts it once as it crosses the line or enters the zone. Counts are stored in Nara’s built-in database, shown on dashboard counters and charts, and published over MQTT or HTTP to your production systems. A count-to node can raise a batch-complete or short-batch alert.
Works with Nara’s built-in models. You set up the zones, rules and outputs. Unusual items can be trained in Reva.
What you need
- An IP camera (RTSP) or Basler industrial camera with a clear view of the items
- A PC, industrial PC or NVIDIA Jetson running Nara
- Optional: an MQTT broker or HTTP endpoint for your MES or ERP
Nara building blocks
- Camera nodes
- Object detection
- Tracker count line and count zones
- Count-to
- Database table
- Dashboard counters and charts
- MQTT and HTTP output
- AI model: General object detection
- AI model: Object tracking and analytics
What it improves
- Count accuracy
- Labour hours saved
- Real-time production visibility
FAQ
Questions, answered.
Can it count our own products?
Common objects are detected out of the box. If your items are unusual, we train a model on them in Reva and Nara uses it in the same flow.
Will the same item be counted twice?
No. Nara tracks each item across frames and counts it once, when it crosses the line or enters the zone.
Can we use the cameras we already have?
Usually, yes. Nara reads standard IP cameras over RTSP, as well as Basler industrial cameras, USB webcams and video files.
Related
Related use cases
cameranara seesBarcode and label inspection
Manual scanning is slow and misses items, and a wrong lot number, expiry date or allergen label means recalls and penalties.
cameranara seesSorting and grading
Sorting products by type, grade or destination by hand is slow and inconsistent.
Robot and AMR orchestration
Robots and AMRs follow fixed programs and do not react to what is actually on the line.
Talk to us about item and pallet counting.
Tell us about your site, cameras and machines, and we’ll show you how it runs on your own hardware, or download Nara and try it yourself.