US167: IoT-Based Automated Fruit Freshness Detection For Supermarket Display Rack

Ho Yu Sheng Universiti Teknologi Malaysia

This project introduces a smart system designed to monitor and evaluate fruit (apple) freshness in supermarket environments. By using the YOLOv11 deep learning model, the system detects and classifies apples as "fresh" or "rotten" through real-time image analysis. Besides, a gas sensor is used to detect the gas level (ethylene) which will be highly produced by the rotten apples for the non-visible apples freshness detection. Additionally, environmental data such as temperature and humidity are captured through integrated sensors, providing insights into conditions that affect the fruit quality.

The data is visualized through a user-friendly mobile application built with Flutter which enables supermarket staff to access information for optimizing inventory management and reducing food waste. This innovative solution aims to reduce workload of the workers, enhance product quality assurance while promoting sustainability in food supply chains.