MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641123493 A) filed by New Prince Shri Bhavani College Of Engineering And Technology; Ajitha G; Janani S; Kaviya E; and Monishajothi R on October 15, 2026, for Ai-Based Multimodal Cold Storage Monitoring And Remaining Shelf Life Prediction Using Cnn.
Inventors include Ajitha G; Janani S; Kaviya E; and Monishajothi R.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The importance of maintaining the suitable environmental conditions in cold storage facilities for preserving the quality of perishable products and reducing unnecessary •food wastage was emphasized. Conventional cold storage monitoring systems are mainly based on the observation of environmental parameters and only alert when specific limits are surpassed, but they do not directly estimate the period of remaining freshness for stored products. The proposed work suggests an AI- based multimodal cold storage monitoring and remaining shelf life prediction system, which fuses the environmental sensing and visual product analysis. The sensing unit is based on an ESP32 microcontroller and is equipped with a DHT22 sensor, which continuously monitors the temperature and humidity of the storage area, and a camera to capture images of the stored product for visual quality assessment. The collected data about the environment and the images of the product are processed using a python-based CNN and multimodal prediction framework. A CNN-based visual analysis model extracts relevant features such as colour, texture, surface condition, and visible deterioration from product images. These visual features are fused with temperature and humidity data to estimate the remaining shelf life and current freshness condition of the product. to estimate the remaining shelf life (days) and the current state of freshness of the product. The proposed approach integrates environmental data from the sensors with quality data from images, allowing for a more extensive evaluation of the quality than traditional threshold-based monitoring methods. All of its predictions are displayed on an interactive Stream-lit dashboard, along with the product• image, temperature, humidity, freshness status, and warning information. Another automated email notification system provides information to the user when an undesirable storage environment or critical freshness condition is detected. The proposed system is an intelligent and cost-effective system for proactive cold storage management, identification of the early signs of spoilage, and reduction of perishable product loss.
Disclaimer: Curated by HT Syndication.
