Developing an IoT and ML-driven platform for fruit ripeness evaluation and spoilage detection: A case study on bananas

Abstract

• This research aims to create a new Internet of Things and Machine Learning based framework for fruit ripeness detection is proposed to reduce food waste. • The proposed system relies on real-time data from IoT devices, such as temperature, humidity, and gas emission sensors, which are connected to ESP32 and Thingsperak cloud Platform. After collecting the data, we employed data analysis, visualization, and various machine learning algorithms to predict the fruit ripeness. • Machine learning algorithms, including CatBoost classifier, are used to detect ripeness stages with high accuracy. • The proposed system could automate quality assessment in agricultural supply chains. • The study contributes to the global efforts to reduce food waste and achieve sustainable development. • The framework has potential applications beyond fruit ripeness detection in the agricultural industry. Food waste is a significant global problem that demands immediate action to reduce it. This study presents a novel framework that merges Internet of Things (IoT) and machine learning (ML) technologies to detect fruit ripeness and spoilage, which is essential in minimizing losses in the cold chain process of fresh produce industry. The study employed temperature, humidity, and gas emission sensors along with an ESP32 microcontroller to establish a unique framework that achieved exceptional accuracy in predicting banana ripeness stages. This framework employed various machine learning algorithms to detect ripeness stages, with the CatBoost classifier exhibiting exceptional performance, demonstrating its dependability and effectiveness in assessing fruit quality. The benefits of this research extend beyond fruit ripeness detection and pave the way for future advancements in automating quality assessment in agricultural supply chains.

Topics & Concepts