MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641097810 A) filed by New Horizon College Of Engineering on August 13, 2026, for An Integrated Hyperspectral Iot Edge-Ai System For Adaptive Spectral Drift Detection And Real- Time Micro-Anomaly Intelligence.
Inventors include Jenita Subash; Sriharshini Kaipa; Poornachandra Tejaswi; Gs Vetri; and Jatin P Hegde.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: An integrated hyperspectral IoT edge-AI system (101) configured for adaptive spectral drift detection and real-time micro-anomaly intelligence in continuous monitoring environments. The system utilizes a Hyperspectral IoT Sensor Node (102) to acquire multi-band spectral data along with environmental parameters, followed by spectral preprocessing and feature extraction (103) and comparison with a dynamically maintained Baseline Spectral Database (104). A Spectral Drift Detection Module (105) computes environmentally compensated deviation metrics, which are evaluated through an adaptive threshold mechanism (106) to validate spectral drift prior to anomaly inference. Upon confirmation of drift, an Edge AI Anomaly Detection Module (202) performs micro-anomaly analysis, and validated events are transmitted through IoT communication and cloud processing (203) for adaptive model updating and alert generation (204). The system further includes a User Dashboard (205) for real-time visualization and monitoring, completing the operational cycle (206). This structured drift-gated edge-cloud architecture enhances detection reliability, reduces false positives, and improves computational efficiency compared to conventional hyperspectral monitoring systems.
Disclaimer: Curated by HT Syndication.
