ONION GUARD: AN IOT-BASED ONION SPOILAGE RISK DETECTION SYSTEM

Manikandan, D and Karthik, S and Santha Kumari, S and Santhosh, R (2026) ONION GUARD: AN IOT-BASED ONION SPOILAGE RISK DETECTION SYSTEM. In: Avishkar 5.0 - the Intra-University Technical Project Expo, 02-04-2026, VISTAS.

[thumbnail of Manikandan Onion Paper.pdf] Text
Manikandan Onion Paper.pdf

Download (949kB)

Abstract

Post-harvest onion storage is vulnerable to spoilage caused by microbial activity and unfavourable
environmental conditions, resulting in significant agricultural losses. An Internet of Things (IoT)
based monitoring system was implemented to enable early spoilage detection by monitoring
volatile organic compound (VOC) emissions, temperature, and humidity within storage
environments. The system used an ESP32 microcontroller integrated with an MQ-135 gas sensor
for VOC detection and a DHT22 sensor for environmental monitoring. The collected data were
processed to compute an Onion Spoilage Risk Index (OSRI), representing the combined influence
of gas concentration, temperature, and humidity on storage stability. Sensor data were transmitted

to the cloud through MQTT communication using AWS IoT and visualized in real time via a Node-
RED dashboard. Experimental evaluation showed that the calculated OSRI, derived from VOC

concentration, temperature, and humidity, remained below 0.45 under normal storage conditions,
indicating safe storage. When environmental changes increased the OSRI beyond the threshold,
spoilage risk was detected, and a ventilation process was activated to reduce gas accumulation.
The monitoring system achieved an overall detection accuracy of approximately 88%. Future work
includes extending the system to support spoilage monitoring for multiple vegetables and
integrating improved sensing and analytical methods to further enhance detection accuracy and
enable intelligent agricultural storage management.

Item Type: Conference or Workshop Item (Paper)
Subjects: Agriculture > Environmental Sciences
Computer Science Engineering > Deep Learning
Computer Science Engineering > Machine Learning
Domains: Computer Science Engineering
Depositing User: Mr IR Admin
Last Modified: 18 May 2026 07:43
URI: https://ir.vistas.ac.in/id/eprint/20068

Actions (login required)

View Item
View Item