Brazilian Researchers Test the Use of Artificial Intelligence to Sort Coffee Beans

Artificial intelligence is becoming more and more prevalent in our daily lives. From automated algorithms that control our social media feeds to the cars around us on the road, AI is seeping into every aspect of modern life.

Recently, researchers in Brazil have been trialing the use of artificial intelligence to sort coffee beans. Using MSI (multispectral imaging) and machine learning, the Brazilian team are classifying coffee beans as either standard or speciality, depending on what they score on the Specialty Coffee Association’s 100-point scale.

How is Speciality Coffee Sorted?

“Normally, trained coffee cuppers will use the SCA standards to manually grade and sort coffee. However, this is a very time-consuming task,” says Rob Hodge, a Director at Rave Coffee. “While the idea of using AI to sort the specialty coffee beans is a fantastic one, it’s important that the quality of these beans doesn’t decrease.

Typically, cuppers in Brazil and elsewhere will be trained to identify a specialty bean and a standard bean based on a range of different metrics. However, manual testing requires inspecting, roasting and tasting coffee samples, which can be an incredible time-consuming process.

If AI can be used to sort specialist green coffee beans from standard coffee beans, that would not only speed up the sorting process, but also ensure that the quality of all organic, green coffee is consistent.”

How Does AI Inspect & Assign A Score to Green Coffee Beans?

As part of the study, four different machine-learning algorithms review images of the coffee beans. By looking for the presence and concentration of certain molecules, including caffeine, chlorogenic acids and catechin, the AI system is able to understand what is considered a specialty bean and what is a standard coffee bean.

“In order to be classified as a specialty green coffee bean, it must score higher than 80 on the SCA scale,” Rob continued, “that ensures a consistent quality across specialty coffee beans, which goes to create a consistent, high-quality taste in the coffee you drink at home.”

The AI System is currently able to assign a number to each coffee bean, differentiating between specialty and standard beans.

What’s most impressive,” says Rob, “is the speed at which these machines are able to identify a quality bean from a standard one. Not only does it eliminate the need for roasting and tasting, it can handle a large number of beans at once, speeding up the entire tasting process, while still adhering to the SCA’s guidelines.”

Does Rave Coffee Use AI In Coffee Bean Sorting?

Currently, Rave Coffee doesn’t use AI sorted coffee beans. While AI does look like it will, one day, be a core component of coffee bean production and sorting, Rave believes in the skill of highly-trained growers and coffee sorters.

All of our producers and partners across the world sort their green coffee beans by hand. That means every specialty coffee we provide has been tested and evaluated by a trained professional. Whether that’s part of our coffee subscriptions service, or even our Nespresso-compatible coffee pods.

While we never say never, at Rave, we’re dedicated to delivering the best coffee to our clients, without compromising our eco-friendly approach and long-term support for independent coffee growers across the world.

Saying that, we’re excited to see the long-term results of this technology, and hope to see it prove useful for our producers, whether they’re based in Brazil, Papua New Guinea or anywhere else.”

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.