fairfolio
Saved Projects

Democratizing Produce Waste Reduction Using Hyperspectral Imaging

AJAS · 2024

Dhruv Darbha · Washington Academy of Sciences

Overview

According to the United Nations, a third of the food globally (1.3 billion tons) is wasted every year. One of the UN Sustainable Development Goals is to, by 2030, reduce per capita global food waste by half at the retail and consumer levels which would also reduce the global emissions of greenhouse gases. This research aims to provide a solution to democratize (make more accessible) the reduction of food waste, specifically, of produce using hyperspectral imaging (HSI) to predict ripeness of vegetables and fruits. HSI measures intensity or reflectance at several (‘hyper’) wavelengths resulting in spectrally abundant information to identify and distinguish unique properties of objects. The author’s research last year showed that Hyperspectral Imaging was an effective method in identifying the ripeness of Tomato. However, it needed to use a stand-alone optical imaging device and this study proposes an economical alternative, iPhone with IR filter. This study answers a question – can spectral imaging in the IR wavelength spectrum be combined with RGB imaging to predict ripeness of fruits or vegetables and a challenge – can an easy-to-use App on an iPhone be developed to predict the firmness of a tomato with a simple click of a tomato with the iPhone’s camera. The study investigates the question using machine learning (ML) models with data collected on spectral IR images (using pass through IR filters) and RGB images from a smartphone camera, and ripeness measurement of tomatoes. Based on several measurements in the near IR spectrum, any measurements in the 750-1000nm were not producing accurate results and so the study focused on wavelengths less than 720 nm. On average, the ML models based on results from these images achieve a root-mean-squared-error (RMSE) on ripeness metric between 2.8 and 3.1 Newtons, which is a significant improvement over the previous research study. The challenge investigates building an App on an iPhone that returns a ripeness metric for the given image of a tomato. The ripeness metric returned by the App was within the RMSE of the actual measurement of the same tomato which provided confidence in the App. Further, the App guides the user to nearest food banks to donate produce if needed. This App directly helps with democratizing produce waste at the hands of an everyday consumer. This research demonstrates nondestructive, low cost and easy to use solutions to democratize the reduction of produce waste at the consumer, retail, and supplier level.

Competition history

  • AJAS 2024 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

Source: AAAS Annual Meeting (Confex) / American Junior Academy of Science

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google