Intellectual Property India Publishes Patent Application for Machine Learning Based System And Method For Cotton Leaf Disease Detection And Classification Using Deep Residual Neural Network

MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202621052114 A) filed by Symbiosis International Deemed University on April 23, 2026, for Machine Learning Based System And Method For Cotton Leaf Disease Detection And Classification Using Deep Residual Neural Network.

Inventors include Sharwil Bhende; Shlok Anand; Samiksha Paliwal; Dr. Monali Gulhane; and Dr. Nitin Rakesh.

The application for the patent was published on June 19, 2026, under issue no. 25/2026.

Abstract: ABSTRACT MACHINE LEARNING BASED SYSTEM AND METHOD FOR COTTON LEAF DISEASE DETECTION AND CLASSIFICATION USING DEEP RESIDUAL NEURAL NETWORK The present invention discloses a machine learning based cotton leaf disease detection and classification system (100) utilizing deep residual neural network architecture. The system comprises an image acquisition module (110), a preprocessing module (120) with data augmentation and normalization capabilities, a deep learning classification module (130) implementing ResNet-50 architecture with pre-trained ImageNet weights, an evaluation module (140), and an output interface module (150). The preprocessing module standardizes images to 224x224 pixels and normalizes pixel intensities. The classification module employs frozen convolutional layers for feature extraction followed by custom dense layers with 256 neurons and ReLU activation, and 2-neuron output layer with softmax activation for binary classification. The system achieves 98.16% accuracy, 98.31% precision, and 97.64% recall for Cotton Leaf Curl Disease detection, significantly outperforming existing approaches. The invention enables integration with IoT systems for real-time agricultural monitoring and farmer accessibility through mobile interfaces. [

Disclaimer: Curated by HT Syndication.