Intellectual Property India Publishes Patent Application for Performance Analysis Of Tea Leaf Disease Detection By Utilizing Hybrid Resnet And Bilstm

MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641101046 A) filed by Sri Eshwar College Of Engineering on August 20, 2026, for Performance Analysis Of Tea Leaf Disease Detection By Utilizing Hybrid Resnet And Bilstm.

Inventor includes Dr. N. Muthukumaran.

The application for the patent was published on August 28, 2026, under issue no. 35/2026.

Abstract: Tea leaf disease detection is a crucial task in the agriculture industry, where timely and accurate identification of leaf conditions can lead to better crop management and yield. Traditional methods of leaf disease detection are often time-consuming and subjective, relying on manual inspections that are not scalable in large tea plantations and are prone to human error. With advancements in deep learning, this study proposes a hybrid model combining ResNet and Bidirectional Long Short-Term Memory (BiLSTM) networks for enhanced performance in tea leaf detection. The ResNet component extracts robust spatial features from images, while BiLSTM captures sequential dependencies within these features to refine the classification accuracy and model interpretability. Through extensive experimentation and testing, we evaluated the effectiveness of the ResNet and BiLSTM combination on a dataset of tea leaf images encompassing a wide range of disease types, severity levels, and environmental variations. Performance metrics such as accuracy, precision, recall, and F1 score were analyzed in depth to benchmark the hybrid model against standalone ResNet and BiLSTM implementations, revealing distinct advantages in prediction stability and generalizability across diverse data. The results indicate that the integrated ResNet-BiLSTM model not only outperforms individual models but also shows greater adaptability to unseen leaf samples, demonstrating superior classification accuracy and robustness in challenging conditions. This research highlights the potential of deep learning frameworks in transforming agricultural diagnostics, presenting a scalable, automated solution for tea leaf disease detection that could contribute significantly to sustainable agricultural practices and more efficient crop management strategies.

Disclaimer: Curated by HT Syndication.