MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202621067093 A) filed by Nichat, Mangesh K.; Yedey, Sanjay; Kute, V. B.; P R Pote Patil College Of Engineering And Management, Amravati; Sagane, S. S.; and Nemade, P. G. on May 28, 2026, for A Computer-Implemented Plant Disease Diagnosis System For Malvaceae Family Crops.
Inventors include Nichat, Mangesh K.; Yedey, Sanjay; Kute, V. B.; Sagane, S. S.; and Nemade, P. G..
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention is related to a computer-implemented plant disease diagnosis system for malvaceae family crops including cotton (1), okra (2), and hibiscus (3). The system comprises an image acquisition module, followed by an image preprocessing pipeline incorporating resizing (4), normalization (5), grayscale conversion (6), and noise removal (7). Dual feature extraction is performed using a Color Layout Descriptor (CLD) module (8) generating CLD prediction output (9), and an Edge Histogram Descriptor (EHD) module (10) generating EHD prediction output (11). A deep learning model creation module (12) and training module (13) utilize extracted CLD and EHD feature vectors instead of raw pixels, thereby reducing computational complexity. A prediction fusion mechanism (14) integrates CLD, EHD, and deep learning outputs through majority voting or weighted averaging to produce a final disease classification output (15). The hybrid approach enhances accuracy, robustness, and efficiency, enabling deployment on edge and mobile devices.
Disclaimer: Curated by HT Syndication.
