Anomaly recovery from compressed spectral imagery via low-rank matrix minimization

Ana Ramirez; Henry Arguello; Gonzalo R. Arce

doi:10.1117/12.887495

3 June 2011 Anomaly recovery from compressed spectral imagery via low-rank matrix minimization

Ana Ramirez, Henry Arguello, Gonzalo R. Arce

Proceedings Volume 8058, Independent Component Analyses, Wavelets, Neural Networks, Biosystems, and Nanoengineering IX; 805806 (2011) https://doi.org/10.1117/12.887495
Event: SPIE Defense, Security, and Sensing, 2011, Orlando, Florida, United States

Abstract

This work describes a methodology for the recovery of anomalies and their spectral signatures from compressively sensed multi-spectral video using Principal Component Pursuit (PCP). In video surveillance, approaches based on PCP allow the anomaly detection in a cluttered background by modeling a sequence of video frames as a large data matrix composed by a low-rank matrix plus a sparse matrix. The low-rank matrix corresponds to the stationary background and the sparse matrix captures the anomalies in the foreground. The compressive spectral video frames are attained by the use of a Coded Aperture Snapshot Spectral Imaging (CASSI) system. The CASSI system allows the compressive measurement of spectrally rich video content by simply capturing a sequence of 2D coded aperture video frames. This paper describes improved procedures for the reconstruction of the video anomalies and their spectra based on the 2-D, aperture-coded, isolated anomalies.

Citation Download Citation

Ana Ramirez, Henry Arguello, and Gonzalo R. Arce "Anomaly recovery from compressed spectral imagery via low-rank matrix minimization", Proc. SPIE 8058, Independent Component Analyses, Wavelets, Neural Networks, Biosystems, and Nanoengineering IX, 805806 (3 June 2011); https://doi.org/10.1117/12.887495

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available