A new AI tool developed by Oregon State University researchers uses smartphone images to accurately predict the ripeness of avocados, potentially reducing food waste and helping consumers enjoy perfectly ripe fruit.
Researchers from Oregon State University and Florida State University have developed an artificial intelligence system that uses smartphone images to accurately predict the ripeness and internal quality of avocados.
This innovative tool, details of which are published in the journal Current Research in Food Science, could significantly reduce food waste and help consumers and retailers make smarter choices about when to use or sell avocados.
“Avocados are among the most wasted fruits globally due to overripeness,” corresponding author Luyao Ma, an assistant professor in the Department of Food Science and Technology in Oregon State’s College of Agricultural Sciences, said in a news release. “Our goal was to create a tool that helps consumers and retailers make smarter decisions about when to use or sell avocados.”
The interdisciplinary team trained their AI model using over 1,400 iPhone images of Hass avocados.
The AI system demonstrated remarkable accuracy, predicting the firmness of avocados — a vital indicator of ripeness — with nearly 92% precision, and distinguishing between fresh and rotten fruit with over 84% accuracy.
These results hint at the robustness of the model, which the researchers believe can be further refined as more images are added. They envision the technology extending its reach to other types of food, leveraging AI to assess ripeness and overall quality.
Looking ahead, the team hopes to further develop this tool so that consumers can utilize it at home, ensuring they enjoy their avocados at peak ripeness. This could help avoid the common disappointment of discovering brown spots after cutting into an avocado.
Beyond home use, the technology has promising applications in avocado processing facilities, where it could optimize sorting and grading processes. For instance, batches identified as more ripe could be shipped to nearer retailers, minimizing waste during transport. Retailers, too, could benefit by prioritizing the sale of avocados based on their ripeness.
The research builds on previous studies that employed machine learning techniques to assess food quality.
“To overcome these limitations, we used deep learning approaches that automatically capture a broader range of information, including shape, texture, and spatial patterns to enhance the accuracy and robustness of avocado quality predictions,” added first author In-Hwan Lee, a doctoral student in the Department of Food Science and Technology.
Ma’s focus on avocados stems from their high market value and significant waste rates. She also noted a personal connection, as an avid consumer of avocado toast often frustrated by the unpredictability of the fruit’s ripeness.
This research addresses a pressing global issue: food waste. Approximately 30% of the world’s food production is wasted. In response, the U.S. Department of Agriculture and the Environmental Protection Agency have set a national goal to cut food waste by half by 2030.
“Avocados are just the beginning,” Ma added. “This technology could be applied much more broadly, helping consumers, retailers, and distributors make smarter decisions and reduce waste.”
Source: Oregon State University