MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202621082758 A) filed by Symbiosis International Deemed University on July 03, 2026, for Hybrid Spatial-Contextual Deep Learning System For Integrated Leaf Quality Assessment.
Inventors include Chetan Bangale; Dr. Aniket K. Shahade; and Priyanka V. Deshmukh.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: ABSTRACT HYBRID SPATIAL-CONTEXTUAL DEEP LEARNING SYSTEM FOR INTEGRATED LEAF QUALITY ASSESSMENT The present invention relates to a hybrid spatial-contextual deep learning system (100) for integrated freshness and disease detection in leafy vegetables. The system comprises an image acquisition and preprocessing module (110), a spatial feature encoder (120) utilizing a densely connected convolutional neural network backbone for hierarchical feature extraction, a contextual dependency module (130) employing a bidirectional gated recurrent unit for modeling long-range spatial relationships among extracted features, a feature aggregation module (140) for compressing contextual representations through global average pooling, and a classification head (150) producing probability distributions across multiple quality classes via softmax activation. The system enables simultaneous multi-attribute classification encompassing variety identification, freshness assessment, and disease detection from leaf images, achieving superior accuracy compared to standalone convolutional architectures while maintaining computational efficiency for precision agriculture deployment. [
Disclaimer: Curated by HT Syndication.
