Single-Pixel Image Reconstruction Based on Block Compressive Sensing and Deep Learning

TL;DR

This paper introduces a deep learning-based method for single-pixel image reconstruction using block compressive sensing, demonstrating improved performance and flexibility in real SPI setups, even with pretrained models on natural images.

Contribution

A novel CNN-based reconstruction model that outperforms existing algorithms and enables size-independent image reconstruction in SPI using block compressive sensing.

Findings

The proposed model surpasses other CS reconstruction algorithms in accuracy.

It can reconstruct images of any size above a minimum threshold.

Pretrained models on natural images effectively reconstruct SPI images from different domains.

Abstract

Single-pixel imaging (SPI) is a novel imaging technique whose working principle is based on the compressive sensing (CS) theory. In SPI, data is obtained through a series of compressive measurements and the corresponding image is reconstructed. Typically, the reconstruction algorithm such as basis pursuit relies on the sparsity assumption in images. However, recent advances in deep learning have found its uses in reconstructing CS images. Despite showing a promising result in simulations, it is often unclear how such an algorithm can be implemented in an actual SPI setup. In this paper, we demonstrate the use of deep learning on the reconstruction of SPI images in conjunction with block compressive sensing (BCS). We also proposed a novel reconstruction model based on convolutional neural networks that outperforms other competitive CS reconstruction algorithms. Besides, by incorporating BCS in our deep learning model, we were able to reconstruct images of any size above a certain smallest image size. In addition, we show that our model is capable of reconstructing images obtained from an SPI setup while being priorly trained on natural images, which can be vastly different from the SPI images. This opens up opportunity for the feasibility of pretrained deep learning models for CS reconstructions of images from various domain areas.