Share Email Print
cover

Proceedings Paper

Spectral dynamic scenes reconstruction based in compressive sensing using optical color filters
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

Spectral-temporal compressive imaging is a technique that allows to sense spatio-spectro-temporal information, known as spectral video from a single low-framerate monochrome measurement. Several optical architectures have been recently developed to capture the spectral features of a dynamic scene or spectral video based on the compressive sensing framework. These spectral-temporal compressive architectures are principally composed by four elements: a main lens, a coded aperture, a dispersive element and an FPA detector. Traditionally, these acquisition systems use block-unblock coded apertures that either block the light rays in the optical path or allow them to pass through. However, the modulation produced by the block-unblock coded apertures is wavelength independent, thus ignoring the highly correlated structure of the spectral information in a dynamic scene. In this work, the block-unblock coded apertures are replaced by colored coded apertures, whose pixels represent a set of specific optical filters such as low, high or band pass filters that modulate particular wavelengths of the scene. An analysis of the variations in the colored coded aperture pattern that allows to obtain improvements in PSNR compared with the block-unblock coded apertures is presented. Simulation results show an improvement up to 2 dB in the reconstruction quality with respect to the block-unblock coded apertures.

Paper Details

Date Published: 10 May 2016
PDF: 8 pages
Proc. SPIE 9860, Hyperspectral Imaging Sensors: Innovative Applications and Sensor Standards 2016, 98600D (10 May 2016); doi: 10.1117/12.2224330
Show Author Affiliations
Kareth M. León, Univ. Industrial de Santander (Colombia)
Laura Galvis, Univ. of Delaware (United States)
Henry Arguello, Univ. Industrial de Santander (Colombia)


Published in SPIE Proceedings Vol. 9860:
Hyperspectral Imaging Sensors: Innovative Applications and Sensor Standards 2016
David P. Bannon, Editor(s)

© SPIE. Terms of Use
Back to Top