Search
Close this search box.
Search
Close this search box.

Foreign Object Detection in Cheese

Cheese on the perClass Mira Stage, being scanned by the MV.C NIR
The MV.Scan NIR package scanning cheese for foreign objects.

Using the MV.C NIR and the perClass Mira Stage and Software

Recalls are costly and damaging to a brand’s reputation. Even a small packaging defect can force a recall of tens of thousands of packages of product.1 For example, a large chain of grocery stores issued a recall of several types of cheese for small shards of plastic that went undetected into plastic-wrapped blocks and grated pieces of cheese in plastic pouches.2

When large blocks of cheese are unwrapped, cut, and repackaged, processors must inspect the cheese for any remaining debris. Clear or even tinted plastic is notoriously difficulty to spot by eye or conventional inspection methods. If any pieces manage to get under the surface, it is likely that the contaminated cheese may be wrapped and sold to an unsuspecting consumer.

Once wrapped, conventional inspection methods have difficulty detecting debris underneath. However, hyperspectral imaging can provide a non-contact, real time detection method as the cheese—wrapped or unwrapped—travels along a conveyor to be sorted, packed, and shipped.

Headwall Group company perClass worked with GROBA, a maker of cheese-processing equipment, to determine how well hyperspectral imaging works for detecting contamination from plastics.

Per object performances using confusion matrix

THE EXPERIMENT

Two different types of cheese with several different colorful or transparent plastics were used. Scans were recorded on the perClass Mira Stage using the MV.C NIR camera from Headwall.

Our objective was to detect foreign objects (colorful and transparent plastics) on top of the cheese using the perClass Mira Stage and Software in the NIR range, and compute the confusion matrix per object for the entire test set.

All 28 foreign objects were detected. There was no false detection.

Headwall’s MV.C NIR and the perClass Mira Software provide a solution for detecting foreign objects commonly found in cheese manufacturing and packaging. Both colorful and transparent plastics can be detected on several types of cheese. Coordinates of foreign bodies can be provided to the actuators for foreign object removal.

Hyperspectral imaging is a great tool for quality control in production lines. It continues to provide time- and money-saving solutions for various industries, giving a real-time, non-contact solution for detecting the unseen. It not only works on cheese, but also in the:

  • Meat industry, detecting plastics, labels, and bones
  • Fish industry, detecting nets or seaweed in mussels
  • Vegetable industry to detect natural objects like rocks in potatoes or spinach, and more!

WANT TO KNOW MORE?

Our Headwall Applications Team will work with you to explore how HSI can deliver value to your food processing plant or farm.

Contact us to learn more!

GROBA ALPHA 400-R unpacking machine uses an integrated robot to hygienically unwrap large cheeses. Typically the cheeses are unwrapped, sliced, and wrapped for sale to the retailer or consumer.
Foreign objects detected in cheese by the MV.C NIR
A false-RGB image of a scan of cheese with plastic foreign objects of varying colors and transparencies. The perClass Mira model successfully identified and placed bounding boxes around all of the pieces of plastic, even those that are invisible to the naked eye and traditional imaging solutions.

More Application Notes

An RGB image of a mixture of herbs and spices alongside a hyperspectral false-color scan of the same pile.
Cayenne and Oregano

Spectral classification via the perClass Mira Software makes the task of assessing herb and spice quality less challenging by providing a clear indication of the

Read More »

Explore more content

Hyperspectral 101

Webinar / Videos

Published Research