How Self-Checkout Cameras Recognise Fruit
A camera compares the colour, shape and texture of what is on the scale with thousands of labelled examples and picks the best match.
The robot repeats this loop many times a second. Tap a step to pause on it.
The scale camera sees a grid of colour numbers. A trained model has learned from many images of apples, bananas and tomatoes what each one tends to look like.
It scores each possible fruit and shows you the top few, so you only need to tap the right one. The weight then gives the price.
Shoppers do not need to look up codes for loose produce, so checkout is faster and fewer items get mis-scanned.
You would search a menu or key in a number for every fruit, and staff would have to help.
Look-alike items, plastic bags or odd lighting can confuse the camera, so it shows several choices rather than deciding alone.
Why does the screen offer a few choices?
Go deeper in the Atlas
How Robots Sense
- Ultrasonic Sensor
- Computer Vision
- How a Smart Speaker Hears Its Wake Word
- Gyroscope and IMU
- How Self-Checkout Cameras Recognise Fruit
- How LiDAR Works
- SLAM
- Sensor Fusion
- SLAM Loop Closure
- Weeding Robots