Object signature acquisition through compressive scanning

Jonathan I. Tamir, Dan E. Tamir, Shlomi Dolev

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper we explore the utility of compressive sensing for object signature generation in the optical domain. We use laser scanning in the data acquisition stage to obtain a small (sub-Nyquist) number of points of an object's boundary. This can be used to construct the signature, thereby enabling object identification, reconstruction, and, image data compression. We refer to this framework as compressive scanning of objects' signatures. The main contributions of the paper are the following: 1) we use this framework to replace parts of the digital processing with optical processing, 2) the use of compressive scanning reduces laser data obtained and maintains high reconstruction accuracy, and 3) we show that using compressive sensing can lead to a reduction in the amount of stored data without significantly affecting the utility of this data for image recognition and image compression.

Original languageAmerican English
Title of host publicationOptical Supercomputing - 4th International Workshop, OSC 2012, in Memory of H. John Caulfield, Revised Selected Papers
Pages105-116
Number of pages12
DOIs
StatePublished - 9 Sep 2013
Event4th International Workshop on Optical SuperComputing, OSC 2012 - Bertinoro, Italy
Duration: 19 Jul 201221 Jul 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7715 LNCS

Conference

Conference4th International Workshop on Optical SuperComputing, OSC 2012
Country/TerritoryItaly
CityBertinoro
Period19/07/1221/07/12

Keywords

  • Compressive Sensing
  • Digital Signal Processing
  • Object Signature
  • Optical Signal Processing
  • Optical SuperComputing
  • Shape Representation

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Object signature acquisition through compressive scanning'. Together they form a unique fingerprint.

Cite this