MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202531091734 A) filed by Narula Institute Of Technology, Kolkata, West Bengal, on Sept. 25, 2025, for 'automated rice grain analysis and classification using ml.'
Inventor(s) include Mr. Suman Kumar Bhattacharyya; Dr. Suchismita Maiti; Dr. Neepa Biswas; Mr. Dipayan Das; Mrs. Mahe Parah; Mrs. Tania Roy; Mr. Sarthak Sampad Roy; and Mr. Premjeet Kumar Singh.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "In this present invention a comprehensive computer vision-based system has been proposed for rice grain classification, size measurement, and broken grain detection using OpenCV and watershed segmentation. The methodology begins with preprocessing steps such as grayscale conversion, Gaussian blurring, and thresholding to reduce noise and enhance contrast. A distance transform and marker-based watershed algorithm are then applied to accurately segment overlapping rice grains. Morphological features such as area, perimeter, aspect ratio, and roundness are extracted from each segmented grain to classify them into categories: short grain, medium grain, and long grain. Additionally, a broken grain detection module classifies grains as broken if their area falls below 75% of the average grain size. The system calculates and displays metrics including grain count, average area, total area, and broken grain percentage. The pipeline includes real-world size estimation by calibrating pixel size, ensuring physical measurement accuracy. Visual annotations are generated on the original image for intuitive analysis. The model's modular design and extensibility make it suitable for integration into industrial quality control systems, offering a low-cost, automated solution for rice quality assessment."
Disclaimer: Curated by HT Syndication.
