MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641070934 A) filed by Srinivasa Ramanujan Institute Of Technology; Dr. P. Veera Prakash; Mr. K. Lokeshnath; Dr. S. Sunitha; Mr. C. Sudheer Kumar; Mr. A. Yerriswamy; Y. Ruhinaaz; R. Siri; and P. Janardhan on June 08, 2026, for Image Based Classification Of Rice Grains Into Raw, Semi Cooked And Fully Cooked Categories Using Deep Learning.
Inventors include Srinivasa Ramanujan Institute Of Technology; Dr. P. Veera Prakash; Mr. K. Lokeshnath; Dr. S. Sunitha; Mr. C. Sudheer Kumar; Mr. A. Yerriswamy; Y. Ruhinaaz; R. Siri; and P. Janardhan.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: The primary aim of this project is to design and develop an system capable of accurately classifying the cooking state of rice into three categories: raw, semi-cooked, and fully cooked. The proposed system utilizes deep learning and computer vision techniques to overcome the limitations of traditional cooking methods, which rely heavily on manual observation, estimation, and fixed time-based approaches. This project focuses on creating a cost-effective, efficient, and automated solution for real- time monitoring of rice cooking status. A custom dataset is developed by capturing rice images under controlled conditions, and advanced preprocessing techniques such as resizing, normalization, Gaussian blur, and data augmentation are applied to enhance model performance and generalization. Transfer learning models including ResNet50, VGG16, and MobileNetV2 are employed for classification. Experimental results demonstrate that ResNet50 and VGG16 achieved a high accuracy of 99%, while MobileNetV2 achieved an accuracy of 84%. These results indicate that deeper architectures such as ResNet50 and VGG16 are more effective for this specific classification task due to their superior feature extraction capabilities. The proposed system analyzes visual features of rice grains and predicts their cooking stage with high precision and reliability. This approach significantly reduces human effort, improves consistency in cooking, and minimizes errors associated with manual inspection. Furthermore, the system is deployed using a Flask-based web application that allows users to upload images and obtain instant predictions along with confidence scores. Overall, this project contributes to the development of intelligent food quality assessment systems and supports advancements in smart kitchen technology by providing a scalable, accurate, and real-time solution for rice cooking state classification.
Disclaimer: Curated by HT Syndication.
