Automated visual quality inspection at line speed
Nara inspects every item at the inspection point for defects, colour, contaminants, size, presence, codes and printed text. Passing items are logged; failing items are rejected through your PLC, with an alert and an image for the record.
[ 01 ]The problem
Visual inspection by people is inconsistent and tiring, and it cannot keep up with line speed or cover every item.
- 01inputCamera at the inspection point, triggered by a part sensor
- 02processChecks for defects, colour, contaminants, size, presence, codes and text
- 03actionPass: log. Fail: reject through the PLC, switch the tower light, alert and log with an image
How it works
A part sensor triggers the camera as each item arrives. Nara runs the checks you configure: a custom defect or classification model trained in Reva, colour thresholds, code reading or OCR. If/switch logic decides pass or fail, writes the result to the PLC over Modbus, switches a tower light, and records the result and image in the database.
We train a model on your own products or defects in Reva, and Nara runs it on the line. Colour, code and text checks work out of the box. Defects specific to your product are trained in Reva.
What you need
- A Basler industrial or IP camera at the inspection point, with consistent lighting
- A part sensor to trigger capture
- A PLC or rejector reachable over Modbus or TCP/IP
- For product-specific defects: sample images to train a model in Reva
Nara building blocks
- Basler and IP camera nodes
- Custom detection or classification model (Reva)
- Colour detection and thresholds
- Code reader
- OCR
- If / switch
- Modbus write to PLC
- Tower light
- Database table
- AI model: OCR (text reading)
- AI model: Barcode and QR code reading
What it improves
- Escape rate (defects reaching customers)
- Inspection staff needed
- Scrap and rework
- An audit trail for every item
[ 03 ]Where it runs
Proof from real deployments.
FAQ
Questions, answered.
Do we need training data?
Not always. Colour, code and text checks use Nara’s built-in tools with no training. Defects specific to your product need a model trained in Reva on your own images.
Can it reject items automatically?
Yes. Nara writes the result to your PLC or rejector over Modbus or TCP/IP, so failing items are removed from the line.
Related
Related use cases
cameranara seesDefect detection and rejection
Defects must be removed from the line automatically, at line speed, without stopping production.
Foreign object detection
Contaminants in bulk product damage quality and customer trust, and manual sampling cannot cover the whole flow.
cameranara seesSorting and grading
Sorting products by type, grade or destination by hand is slow and inconsistent.
Talk to us about visual quality inspection.
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.