HyperVideo: Broadband, Multi- & Hyper-Spectral Agile Imager

Period of Performance: 07/23/2013 - 01/23/2014


Phase 1 SBIR

Recipient Firm

Opto-knowledge Systems, Inc.
19805 Hamilton Ave Array
Torrance, CA 90502
Principal Investigator


Present multi-/hyper-spectral imagers that function at video frame rates must sacrifice either spatial or spectral resolution in order to unwrap the scene and project the data cube onto the focal plane array. Warfighters, however, are accustomed to, and do not want to give up high-resolution video imagery for additional hyperspectral data. High-resolution video provides an intuitive picture of the scene of interest. Hyperspectral sensors provide data that must be analyzed electronically. We propose an agile sensor that produces full spatial resolution video just like a conventional camera, but at the same time, behind the scenes produces multi- or hyper-spectral data at a spectral resolution that is automatically selected based on the expected threat level in the environment. Threat detection algorithms can run in the background, in real-time, thereby providing useful target detection without interfering with the warfighter's conventional video display. Under Phase-I, we will develop several designs that we will test in the laboratory using components of existing OptoKnowledge hyperspectral sensors. We will also enhance our existing hyperspectral sensor spectro-radiometric performance model, and adapt to the concepts described in this proposal. Under the Phase-I Option, we will decide on a selected approach, conduct optical tolerance analysis, and embark on a conceptual mechanical design for a sensor to be built under Phase-II.