Time-domain terahertz compressive imaging

TL;DR

This paper introduces a novel single-pixel terahertz imaging system that uses compressive sensing to efficiently reconstruct temporal, spatial, and spectral information without mechanical scanning, reducing complexity and acquisition time.

Contribution

It integrates single-pixel imaging with compressive sensing into THz time-domain spectroscopy, enabling faster, less complex THz imaging with spectral and temporal capabilities.

Findings

Successful reconstruction of THz waveforms without raster-scanning

Demonstration of time-of-flight imaging using temporal data

Reconstruction of spectral images with less than 50% measurements

Abstract

We present an implementation of the single-pixel imaging approach into a terahertz (THz) time-domain spectroscopy (TDS) system. We demonstrate the indirect coherent reconstruction of THz temporal waveforms at each spatial position of an object, without the need of mechanical raster-scanning. First, we exploit such temporal information to realize (far-field) time-of-flight images. In addition, as a proof of concept, we apply a typical compressive sensing algorithm to demonstrate image reconstruction with less than 50% of the total required measurements. Finally, the access to frequency domain is also demonstrated by reconstructing spectral images of an object featuring an absorption line in the THz range. The combination of single-pixel imaging with compressive sensing algorithms allows to reduce both complexity and acquisition time of current THz-TDS imaging systems.