Multiscale random projections for compressive classification
| dc.citation.bibtexName | inproceedings | en_US |
| dc.citation.journalTitle | IEEE International Conference on Image Processing (ICIP) | en_US |
| dc.citation.location | San Antonio, Texas | en_US |
| dc.contributor.author | Duarte, Marco F. | en_US |
| dc.contributor.author | Davenport, Mark A. | en_US |
| dc.contributor.author | Wakin, Michael B. | en_US |
| dc.contributor.author | Laska, Jason N. | en_US |
| dc.contributor.author | Takhar, Dharmpal | en_US |
| dc.contributor.author | Kelly, Kevin F. | en_US |
| dc.contributor.author | Baraniuk, Richard G. | en_US |
| dc.date.accessioned | 2008-08-19T03:20:25Z | en_US |
| dc.date.available | 2008-08-19T03:20:25Z | en_US |
| dc.date.issued | 2007-09-01 | en_US |
| dc.description.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. | en_US |
| dc.description.sponsorship | Supported by NSF, ONR, AFOSR, DARPA and the Texas Instruments Leadership University Program. Thanks to Texas Instruments for providing the TI DMD developer’s kit and accessory light modulator package (ALP). | en_US |
| dc.identifier.citation | M. F. Duarte, M. A. Davenport, M. B. Wakin, J. N. Laska, D. Takhar, K. F. Kelly and R. G. Baraniuk, "Multiscale random projections for compressive classification," <i>IEEE International Conference on Image Processing (ICIP),</i> 2007. | en_US |
| dc.identifier.doi | http://dx.doi.org/10.1109/ICIP.2007.4379546 | en_US |
| dc.identifier.uri | https://hdl.handle.net/1911/21681 | en_US |
| dc.language.iso | eng | en_US |
| dc.subject | data compression | en_US |
| dc.subject | image coding | en_US |
| dc.subject | image classification | en_US |
| dc.subject | object recognition | en_US |
| dc.title | Multiscale random projections for compressive classification | en_US |
| dc.type | Journal article | en_US |
| dc.type.dcmi | Text | en_US |