TY - GEN
T1 - Object signature acquisition through compressive scanning
AU - Tamir, Jonathan I.
AU - Tamir, Dan E.
AU - Dolev, Shlomi
PY - 2013/9/9
Y1 - 2013/9/9
N2 - 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.
AB - 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.
KW - Compressive Sensing
KW - Digital Signal Processing
KW - Object Signature
KW - Optical Signal Processing
KW - Optical SuperComputing
KW - Shape Representation
UR - http://www.scopus.com/inward/record.url?scp=84883324310&partnerID=8YFLogxK
U2 - https://doi.org/10.1007/978-3-642-38250-5_11
DO - https://doi.org/10.1007/978-3-642-38250-5_11
M3 - Conference contribution
SN - 9783642382499
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 105
EP - 116
BT - Optical Supercomputing - 4th International Workshop, OSC 2012, in Memory of H. John Caulfield, Revised Selected Papers
T2 - 4th International Workshop on Optical SuperComputing, OSC 2012
Y2 - 19 July 2012 through 21 July 2012
ER -