MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202521132704 A) filed by Sandip Institute Of Technology And Research Centre; Sandip Institute Of Engineering And Management; Sandip Polytechnic; Sandip Institute Of Pharmaceutical Sciences; Sandip University Nashik; and Sandip University Sijjoul on December 27, 2025, for An Automated Soybean Leaf Disease Detection System.
Inventors include Karale, Ankita; Gaidhani, Abhay; Kale, Sunil; Thouttam, Naresh; and Patil, Pramod.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: The present invention relates to an automated soybean leaf disease detection system using image processing and deep learning techniques to improve accuracy, efficiency, and timeliness of disease diagnosis in agricultural fields. The system captures high-resolution images of soybean leaves using a digital imaging device and performs image pre-processing, background removal, and leaf segmentation to isolate relevant leaf regions from complex field environments. Disease-affected lesions are identified through advanced segmentation methods based on color space transformation and thresholding techniques. A deep learning–based classification module employing continuous convolution layers and sparse Maxout convolutional neural networks extracts discriminative features and accurately classifies soybean leaf diseases, including brown spot disease. The system further includes a database for storing images, trained models, and results, along with user and administrator modules for secure access, data management, and system updates. The invention enables rapid, cost-effective, and reliable soybean disease diagnosis, supports precision agriculture practices, reduces yield losses, and assists farmers and agricultural stakeholders in effective disease management.
Disclaimer: Curated by HT Syndication.
