With its latest sorting model, the optical sorting specialist brings greater precision and efficiency to vegetable processing.

With A-EYE, Optimum Sorting introduces a new generation of sorting technology for the vegetable processing industry. By leveraging Deep Learning, the system can detect and re-move defects in green beans and spinach that have long challenged conventional sorting technology. This includes stems, weeds and other plant material that closely resemble good product. As a result, processors benefit from higher sorting quality, increased automation and greater line capacity.

The launch of A-EYE is the result of years of research and practical expertise in sorting tech-nology. Since 2019, Optimum Sorting, headquartered in Belgium, has been investigating how Deep Learning can be used to tackle increasingly complex sorting challenges.
“We have been working with Deep Learning in various sorting applications for years, including the detection of cracks in whole carrots,” says Steve Raskin, CSO at Optimum Sorting. “To-day, the technology, computing power and training methodologies have advanced to a level where these models can be deployed reliably in industrial production environments. As a result, we can now address sorting challenges that were simply not feasible a few years ago.”

A sorting model that sees like a human

The name A-EYE combines “artificial intelligence” and “eye”. “While the model is powered by artificial intelligence, it is designed to assess objects much like the human eye does,” explains Jan De Jonghe, CTO at Optimum Sorting.

Unlike traditional sorting algorithms, which rely on predefined rules, A-EYE learns from re-al-world examples. As a result, the model can identify subtle differences that are difficult for conventional vision systems to detect. A-EYE does not rely on a single characteristic but evaluates objects based on a combination of shapes, textures and visual patterns.

“One of the most common misconceptions about AI is that it automatically knows what good product looks like and what doesn’t,” says Steve Raskin. “In reality, every model is built on the expertise of product specialists. Their knowledge shapes the training data and teaches the system what to look for when identifying defects. You could say we’re translating human expertise into a model that can apply the same judgement more consistently and thousands of times per second.”

“A-EYE FUNDAMENTALLY CHANGES THE WAY SORTING SYSTEMS EVALUATE PRODUCTS”

Setting a new standard in green bean and spinach sorting
Optimum Sorting chose green beans and spinach as the first applications for A-EYE. These product streams are particularly challenging to sort, as defects often closely resemble the product itself.

In spinach, weeds such as nettles and thistles can be difficult to distinguish from good prod-uct. In green beans, stem ends often resemble the tip of a bean, making accurate sorting a challenge for conventional vision systems. A-EYE also detects stalks, twigs and other plant material that enter the product stream during harvesting. Because these materials often closely resemble good product, they have traditionally been difficult to remove using conven-tional sorting technology.

According to Jan De Jonghe, purslane (Portulaca oleracea) is one of the clearest examples of the value A-EYE can deliver: “Purslane detection is one of the best examples of what Deep Learning can achieve,” says De Jonghe. “After harvesting, this plant shares so many visual characteristics with green beans that reliably distinguishing between the two becomes ex-tremely difficult using conventional technology. As a result, growers have seen entire batches rejected, even though a significant portion of the crop still met quality specifications. With A-EYE, we can identify purslane far more accurately, allowing more valuable product to be retained. The result is less food waste and significant savings throughout the season.”

Improving quality, yield and capacity
Complex defects in product streams such as green beans and spinach can now be detected with far greater accuracy and reliability. The impact is clear: detection rates for stem ends in green beans increased from 70% to 97%, resulting in a cleaner end product and higher overall product quality.

Beyond quality improvements, A-EYE also enables higher sorting line capacities. Thanks to its AI-powered recognition capabilities, the system continues to classify products accurately even when vegetables partially overlap on the conveyor. This allows processors to increase throughput without compromising sorting performance.

“WITH A-EYE, UP TO +50% MORE PRODUCT CAN BE PROCESSED WITHIN THE SAME SORTING TIME.”

French fry sorting also benefits from advanced algorithms
A-EYE is just one example of how artificial intelligence is pushing the boundaries of sorting technology. Optimum Sorting is also applying AI to new challenges within the potato proces-sing industry. One recent example is its collaboration with Belgian AI company Polysense on advanced French fry length measurement.

According to Tybo Mollet, CEO of Optimum Sorting, this collaboration demonstrates how artificial intelligence is creating new opportunities: “Our collaboration with Polysense per-fectly illustrates how AI is pushing the boundaries of traditional machine vision. Just as we can identify subtle differences between good beans and unwanted material, new algorithms now make it possible to accurately measure individual French fries within complex product streams. With the rollout of these new sorting models, we aim to set a new benchmark for what processors can expect from modern sorting technology.”