Using Ultralytics’ YOLO models, Specialvideo has developed a reliable, high-precision quality control system for pizzas that identifies, counts, and verifies toppings on the fly.
Pizza makers often need to produce visually appealing, high-quality products, but visually inspecting the toppings on each pizza may not be compatible with fast production lines that bake a new pizza every 600 milliseconds! Specialvideo uses an artificial intelligence-based approach to detect and count ingredients in real time, enabling manufacturers to meet recipe standards, reduce waste, and maintain smooth operations.
Dedicated to this vision system, Ultralytics published the article Ultralytics YOLO powers Specialvideo’s food inspection tool.
A promising result regarding the extension of our pizza inspection system was the recognition of ingredients even on food products never used during training, such as salad and pasta. This gives us a great advantage on new products with characteristics common to products already seen, and the dataset construction process will be significantly accelerated. For example, we achieved recognition on salad and pasta of various ingredients learned in the context of stuffed pizzas.
