A research project received $640,000 in funding to track household food waste using artificial intelligence (AI) applications.
The project will be led by Patrick Donnelly, Assistant Professor of Computer Science at Oregon State University’s School of Engineering. With more than a third of all food produced in the U.S. going to waste, this project could solve billions of dollars a year.
“Waste is tracked, measured and quantified at every other stage of the agricultural supply chain,” explains Donnelly.
“However, approaches to measuring post-consumer waste are costly, time-consuming, prone to human error, and impractical at scale.”
Using AI applications to measure the amount of food we throw away
Donnelly and his colleagues are building a kitchen compost bin that automatically measures how much food your home throws away.
he said:
“When a user throws food waste and non-edible food waste into the trash, our device uses an AI application to prompt the user to describe the item thrown in. The user’s notes are automated voice recognition. transcribed in and associated with the item’s weight measurement.”
Additionally, the device collects 3D images and sensor readings of the waste, providing “a whole new set of data to enable and encourage future researchers to use computer vision to tackle the problem of food waste measurement.” bring the set.
Food waste is a global problem
This research was supported by the Food and Agriculture Research Foundation and Kroger. Zero Waste Foundation.
According to the Foundation for Food & Agriculture Research, approximately 37% of food waste in the United States occurs at home. This equates to a staggering $400 billion worth of food waste each year.
Aside from the economic inefficiency and moral issues of throwing away food, this waste produces large amounts of greenhouse gases such as carbon dioxide and methane.
“There’s a familiar adage: You can’t manage what you don’t measure,” says Donnelly. “Our goal is to drive future waste reduction by specifically quantifying, measuring and tracking the amount of food composted by household consumers.”
Dataset collection
As part of the project, researchers will conduct a small pilot study. This could be done in spring 2024.
Mr Donnelly said: The purpose of the feasibility study is to test the technology and collect measurements and images of the waste dataset. “
Currently, this type of study is conducted by asking small groups of participants to manually weigh food waste and record the measurements in a journal. However, Donnelly believes the proposed solution is more efficient.
“Our solution fully automates this process, allowing researchers to collect this data more accurately and efficiently,” he said.
“Our work is the first step towards developing a fully autonomous computer vision solution that will allow households to track the amount and type of food compost waste they generate. Through driven interventions, we hope to encourage consumers to think about wasted food over time and change their behavior.”
