FRUDON: A Fruit Donation Platform with Convolutional Neural Network Based Fruit Shelf-life Prediction for Mitigating Starvation

Ashok  Kumar P. S.; Mohammed  Raiyan Ahmed; Mohammed  Salman M. R.; Saba  Samreen; Sabah  Khanam

Abstract

According to the United Nations Global Hunger Index, India's hunger situation is portrayed as alarming, as it ranks 101st out of 116 nations. The government predicts that 69% of children under the age of 5 years would die from malnutrition and hunger by the end of 2023, as starvation deaths have grown widespread in the nation. To tackle this concern, the mitigation of food wastage has been recognized as a prospective approach. This method introduces a two-part system, comprising a fruit shelf-life predictor and a mobile/web application that serves as a platform linking donors and NGOs. The fruit shelf-life predictor employs deep learning algorithms within a convolutional neural network (CNN) framework to examine and categorize fruit images, delivering an assessment of the remaining shelf life in terms of safety, nutrition, and taste. The mobile/web application serves as a platform for users to input fruit images, receive shelf-life predictions, and connect with NGOs for donating surplus fruits to reduce wastage. Similar to the human brain system, the CNN architecture employed in the fruit shelf-life prediction can recognize and arrange different elements in fruit photos. This system aims to leverage technology to reduce food wastage and combat hunger in India by providing an efficient and user-friendly platform for donors and NGOs to connect and share surplus fruits.

Abstract

Keywords