Plate Dispenser Using Face Recognition to Prevent Food Wastage in Hostel Mess

Abstract

Food wastage in hostel mess facilities is a persistent and critical issue faced by educational institutions, primarily due to unauthorized access by day scholars and repeated food collection by the same individuals. Conventional food distribution systems rely heavily on manual supervision, identity cards, or verbal verification, all of which are inefficient, error-prone, and susceptible to misuse. This paper proposes a fully automated plate dispensing system based on face recognition technology to ensure controlled and fair food distribution in hostel mess environments. The proposed system integrates computer vision techniques with embedded hardware components such as a Raspberry Pi, Arduino microcontroller, relay module, and solenoid lock to automate the entire plate allocation process. Real-time facial images are captured through a camera and processed using OpenCV-based face detection algorithms. The detected face is verified against a pre-trained dataset consisting of hostel students and non-hostelers. Only authorized hostel members are allowed to receive a plate, and the system ensures that each individual is served only once per meal cycle. A dataset of 800 individuals is used for training and validation, with live image capture during testing. Experimental observations demonstrate that the proposed system effectively reduces food wastage, prevents unauthorized access, and improves transparency, accountability, and operational efficiency in hostel mess management. The system provides a scalable and cost-effective solution that can be further enhanced using cloud integration and mobile-based monitoring.