Multiscale random projections for compressive classification

Marco F. Duarte, Mark A. Davenport, Michael B. Wakin, Jason N. Laska, Dharmpal Takhar, Kevin F. Kelly, Richard G. Baraniuk

Research output: Chapter in Book/Report/Conference proceedingConference contribution

22 Scopus citations

Abstract

We propose a framework for exploiting dimension-reducing random projections in detection and classification problems. Our approach is based on the generalized likelihood ratio test; in the case of image classification, it exploits the fact that a set of images of a fixed scene under varying articulation parameters forms a low-dimensional, nonlinear manifold. Exploiting recent results showing that random projections stably embed a smooth manifold in a lower-dimensional space, we develop the multiscale smashed filter as a compressive analog of the familiar matched filter classifier. In a practical target classification problem using a single-pixel camera that directly acquires compressive image projections, we achieve high classification rates using many fewer measurements than the dimensionality of the images.

Original languageEnglish (US)
Title of host publication2007 IEEE International Conference on Image Processing, ICIP 2007 Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages161-164
Number of pages4
ISBN (Print)1424414377, 9781424414376
DOIs
StatePublished - 2006
Event14th IEEE International Conference on Image Processing, ICIP 2007 - San Antonio, TX, United States
Duration: Sep 16 2007Sep 19 2007

Publication series

NameProceedings - International Conference on Image Processing, ICIP
Volume6
ISSN (Print)1522-4880

Other

Other14th IEEE International Conference on Image Processing, ICIP 2007
CountryUnited States
CitySan Antonio, TX
Period9/16/079/19/07

Keywords

  • Data compression
  • Image classification
  • Image coding
  • Object recognition

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint

Dive into the research topics of 'Multiscale random projections for compressive classification'. Together they form a unique fingerprint.

Cite this