On the design of a sparsifying dictionary for compressive image feature extraction
| Autor: | Marco Trevisi, Jorge Femandez-Berni, Ángel Rodríguez-Vázquez, Ricardo Carmona-Galan |
|---|---|
| Přispěvatelé: | Universidad de Sevilla. Departamento de Electrónica y Electromagnetismo |
| Rok vydání: | 2015 |
| Předmět: |
Compressive sampling
business.industry Computer science Feature extraction Pattern recognition Reconstruction algorithm Sparse approximation Image plane Image feature extraction Compressed sensing Sampling (signal processing) Feature (computer vision) Artificial intelligence business Sparse representation Feature detection (computer vision) |
| Zdroj: | idUS. Depósito de Investigación de la Universidad de Sevilla instname ICECS |
| ISSN: | 2012-3892 |
| Popis: | Compressive sensing is an alternative to Nyquist-rate sampling when the signal to be acquired is known to be sparse or compressible. A sparse signal has a small number of nonzero components compared to its total length. This property can either exist either in the sampling domain, i. e. time or space, or with respect to a transform basis. There is a parallel between representing a signal in a compressed domain and feature extraction. In both cases, there is an effort to reduce the amount of resources required to describe a large set of data. A given feature is often represented by a set of parameters, which only acquire a relevant value in a few points in the image plane. Although there are some works reported on feature extraction from compressed samples, none of them considers the implementation of the feature extractor as a part of the sensor itself. Our approach is to introduce a sparsifying dictionary, feasibly implementable at the focal plane, which describes the image in terms of features. This allows a standard reconstruction algorithm to directly recover the interesting image features, discarding the irrelevant information. In order to validate the approach, we have integrated a Harris-Stephens corner detector into the compressive sampling process. We have evaluated the accuracy of the reconstructed corners compared to applying the detector to a reconstructed image. Ministerio de Economía y Competitividad TEC2012-38921-C02, IPT-2011-1625-430000, IPC-20111009 Junta de Andalucía TIC 2338-2013 Office of Naval Research (USA) N000141410355 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|