Researchers think they're on the cusp of a better way to identify ripe tomatoes. Their novel method involves innovative tech, including hyperspectral imaging, machine learning, and artificial intelligence.
The hope of the Hebrew University of Jerusalem-led team is ultimately to expand the scope beyond tomatoes and save farmers from continuing to leverage the conventional methods of evaluating produce quality described in a recent study as "time-consuming, expensive, and limited in scope." The research was published in the journal Computers and Electronics in Agriculture, and the team detailed their findings in a news release.
Handheld hyperspectral imaging allows users to assess crops like tomatoes based on how they reflect light. That offers the advantage of being non-destructive, fast, and cost-effective compared to current methods.
After capturing the images of 567 tomatoes over five cultivars, the scientists tested out machine learning algorithms that could project tomato quality in seven critical parameters. Those were weight, firmness, total soluble solids (TSS), citric acid, ascorbic acid, lycopene, and pH.
The best of their algorithms delivered an R² of 0.94 for tomato weight and 0.89 for firmness in the Random Forest Algorithm, with all but one of the scores in the seven measurements over 0.60. That translates to high accuracy that can save farmers time and money.
The scientists say their research could equip farmers with a portable, low-cost device down the road, one that could measure more crops in all sorts of conditions. Another selling point is the tech's capability to be used as crops ripen, which could improve harvest timing and quality.
"Our research aims to bridge the gap between advanced imaging technology, AI, and practical agricultural applications," said research leader Dr. David Helman.
The team's work joins a slew of efforts to use AI to optimize farming globally. A key area in which scientists are deploying AI is the aim of greatly reducing herbicide use in agriculture. AI-powered cameras allow farmers to more sparingly use chemicals, reducing environmentally damaging "drift," saving farmers money, and protecting consumers.
Another promising use case for machine learning and AI is as part of a multinational effort to aid wheat farmers in the fight against harvest-threatening rust. AI is also being deployed in autonomous tractors and fungus-blasting robots — and these are just some of the innovations that need to be seen to be believed.
|
What is the biggest reason you don't grow food at home? Click your choice to see results and speak your mind. |
Not enough time ⏳ As the global food supply faces increasing challenges posed by the changing climate, technology could be a powerful tool to help farmers. All that computing power does come with a cost, and it's important that companies use clean energy to power AI while finding ways to make it more efficient.
TCD Picks » Upway Spotlight
💡Upway makes it easy to find discounts of up to 60% on premium e-bike brands
Dr. Helman has been encouraged by his team's results but is already looking to expand the work to address all sorts of fruits while bringing the cost down.
"This work has the potential to revolutionize quality monitoring not only in tomatoes but also in other crops," Helman said. "Our next step is to build a low-cost device (ToMAI-SENS) based on our model that will be used across the fruit value chain, from farms to consumers."
Join our free newsletter for weekly updates on the latest innovations improving our lives and shaping our future, and don't miss this cool list of easy ways to help yourself while helping the planet.
Cool Picks
