Image Classification with A Deep Network Model based on Compressive Sensing

TL;DR

This paper introduces CSNet, a deep learning model utilizing cascaded compressive sensing for image classification, which simplifies network parameters and achieves high accuracy on MNIST.

Contribution

The paper presents a novel cascaded compressive sensing approach for feature extraction in deep networks, reducing parameters and maintaining high classification performance.

Findings

Higher accuracy on MNIST dataset

Effective feature extraction via compressive sensing

Simplified network architecture

Abstract

To simplify the parameter of the deep learning network, a cascaded compressive sensing model "CSNet" is implemented for image classification. Firstly, we use cascaded compressive sensing network to learn feature from the data. Secondly, CSNet generates the feature by binary hashing and block-wise histograms. Finally, a linear SVM classifier is used to classify these features. The experiments on the MNIST dataset indicate that higher classification accuracy can be obtained by this algorithm.