Compressively Sensed Image Recognition
This work addresses the computational bottleneck in compressive sensing for image recognition, offering an incremental improvement for applications requiring low-cost sampling.
The paper tackles the problem of costly image reconstruction in compressive sensing by introducing a DCT-based method to extract binary discriminative features directly from CS measurements, and shows that fusing these features with CNN-based vectors outperforms state-of-the-art methods.
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.