Share Email Print
cover

Proceedings Paper

Compact full-motion video hyperspectral cameras: development, image processing, and applications
Author(s): A. V. Kanaev
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Emergence of spectral pixel-level color filters has enabled development of hyper-spectral Full Motion Video (FMV) sensors operating in visible (EO) and infrared (IR) wavelengths. The new class of hyper-spectral cameras opens broad possibilities of its utilization for military and industry purposes. Indeed, such cameras are able to classify materials as well as detect and track spectral signatures continuously in real time while simultaneously providing an operator the benefit of enhanced-discrimination-color video. Supporting these extensive capabilities requires significant computational processing of the collected spectral data. In general, two processing streams are envisioned for mosaic array cameras. The first is spectral computation that provides essential spectral content analysis e.g. detection or classification. The second is presentation of the video to an operator that can offer the best display of the content depending on the performed task e.g. providing spatial resolution enhancement or color coding of the spectral analysis. These processing streams can be executed in parallel or they can utilize each other’s results. The spectral analysis algorithms have been developed extensively, however demosaicking of more than three equally-sampled spectral bands has been explored scarcely. We present unique approach to demosaicking based on multi-band super-resolution and show the trade-off between spatial resolution and spectral content. Using imagery collected with developed 9-band SWIR camera we demonstrate several of its concepts of operation including detection and tracking. We also compare the demosaicking results to the results of multi-frame super-resolution as well as to the combined multi-frame and multiband processing.

Paper Details

Date Published: 16 October 2015
PDF: 7 pages
Proc. SPIE 9649, Electro-Optical Remote Sensing, Photonic Technologies, and Applications IX, 96490R (16 October 2015); doi: 10.1117/12.2196718
Show Author Affiliations
A. V. Kanaev, U.S. Naval Research Lab. (United States)


Published in SPIE Proceedings Vol. 9649:
Electro-Optical Remote Sensing, Photonic Technologies, and Applications IX
Gary Kamerman; Ove Steinvall; Keith L. Lewis; John D. Gonglewski, Editor(s)

© SPIE. Terms of Use
Back to Top