Tampere University of Technology

TUTCRIS Research Portal

Compressively Sensed Image Recognition

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

Details

Original languageEnglish
Title of host publication2018 7th European Workshop on Visual Information Processing (EUVIP)
PublisherIEEE
Number of pages6
ISBN (Electronic)978-1-5386-6897-9
ISBN (Print)978-1-5386-6898-6
DOIs
Publication statusPublished - Nov 2018
Publication typeA4 Article in a conference publication
EventEuropean Workshop on Visual Information Processing -
Duration: 1 Jan 1900 → …

Publication series

Name
ISSN (Electronic)2471-8963

Conference

ConferenceEuropean Workshop on Visual Information Processing
Period1/01/00 → …

Abstract

Compressive Sensing (CS) theory asserts that sparse signal reconstruction is possible from a small number of linear measurements. Although CS enables low-cost linear sampling, it requires non-linear and costly reconstruction. Recent literature works show that compressive image classification is possible in CS domain without reconstruction of the signal. In this work, we introduce a DCT base method that extracts binary discriminative features directly from CS measurements. These CS measurements can be obtained by using (i) a random or a pseudorandom measurement matrix, or (ii) a measurement matrix whose elements are learned from the training data to optimize the given classification task. We further introduce feature fusion by concatenating Bag of Words (BoW) representation of our binary features with one of the two state-of-the-art CNN-based feature vectors. We show that our fused feature outperforms the state-of-the-art in both cases.

Keywords

  • Image coding, Feature extraction, Image reconstruction, Discrete cosine transforms, Task analysis, Sensors, Microsoft Windows, Compressive Sensing, Compressive Learning, Inference on Measurement Domain, Learned Measurement Matrix, Compressive Classification, DCT-based Binary Descriptor

Publication forum classification

Field of science, Statistics Finland