Compressively Sensed Image Recognition

TL;DR

This paper presents a method for direct image classification from compressive sensing measurements using binary discriminative features, improving accuracy by fusing with CNN features.

Contribution

It introduces a DCT-based binary feature extraction method directly from CS measurements and combines it with CNN features for enhanced classification performance.

Findings

Fused features outperform state-of-the-art methods

Binary features extracted directly from CS measurements are effective

Measurement matrices can be learned from data

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 pseudo-random 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.