Measuring optical polarization requires a loss of resolution and/or signal to noise ratio. We are working on methods that allow this resolution loss to be controlled by creating hybrid modulation schemes and controlling the reconstruction methods. We show modeling and experimental results with space-time modulation.
The need for a reliable and cost-efficient gas detection and identification system is of prime importance, especially when security threatening situations occur. In this communication we will discuss the Telops newly designed UAV-based Compact TIR Hyperspectral Imaging solution for Real-time Gas Detection, Identification and Quantification.
Efforts to apply machine learning to HSI have produced mixed results as numerous challenges exist in applying these methods. This work describes these challenges along with on-going and future work aimed at reaping the potential benefits offered by machine learning in improving HSI data analysis.
We introduce an architecture for snapshot spectral imaging that combines pixelated interference filters and stacked photodiode imagers. Simple design decisions permit target detection with improved signal-to-noise ratios or spectral reconstruction with enhanced accuracy while maintaining the compactness and robustness required for human-borne and vehicle-mounted sensors.