Intellectual Property India Publishes Patent Application for A Deep Learning Framework With Spatially Modulated Continuous Layer For Plant Leaf Disease Detection

MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202621081512 A) filed by Symbiosis International Deemed University on July 01, 2026, for A Deep Learning Framework With Spatially Modulated Continuous Layer For Plant Leaf Disease Detection.

Inventors include Dr. Aniket K. Shahade; and Dr. Priyanka V. Deshmukh.

The application for the patent was published on August 21, 2026, under issue no. 34/2026.

Abstract: ABSTRACT A DEEP LEARNING FRAMEWORK WITH SPATIALLY MODULATED CONTINUOUS LAYER FOR PLANT LEAF DISEASE DETECTION The present invention discloses a deep learning framework (100) with a Continuous Layer (122) employing trainable Gaussian basis functions for spatially adaptive convolution in plant leaf disease detection. The framework comprises a Continuous Layer feature extraction module (120) with spatial modulation (124), a CNN backbone (130), a hybrid variational loss module (140) combining cross-entropy (142), kernel smoothness penalty (144), and basis function dispersion penalty (146), and a classification head (150). The Gaussian-modulated convolution adapts to disease-specific spatial patterns. The framework achieves 98.63 percent accuracy across four apple leaf disease categories, offering a parameter-efficient alternative to attention mechanisms. [

Disclaimer: Curated by HT Syndication.