MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641079258 A) filed by Vellore Institute Of Technology on June 26, 2026, for System For Hyperspectral Anomaly Detection Through Scene-Regime Classification, Detector Orchestration, And Adaptive Anomaly-Map Generation.
Inventors include Aju D; Pranav Kumar; and Cheralathan M.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present disclosure proposes a system (100) and a method for hyperspectral anomaly detection. The system (100) comprises a computing device (102) having a processor (104) and a memory (106) storing instructions executable by the processor (104). The processor (104) executes a spectral representation selection module (108), a scene profiling and classification module (110), a detector orchestration module (112), a hierarchical fusion module (114), and a threshold optimization module (116). The system (100) provides a fully unsupervised hyperspectral anomaly detection framework configured to detect anomalous targets within hyperspectral image data without requiring labelled training samples or prior target knowledge. By deriving detection parameters from scene-specific statistical characteristics, the system (100) adapts anomaly detection operations according to characteristics of a processed hyperspectral image cube and improves anomaly-background separation across diverse scene conditions.
Disclaimer: Curated by HT Syndication.
