Single-pixel terahertz imaging via compressed sensing

With the development of terahertz related technologies, the terahertz imaging technology will show its greater practical value in more areas. In this paper, we describe a terahertz imaging system that uses a single pixel detector in combination with a series of random masks to enable high-speed image acquisition. The image formation is based on the theory of compressed sensing (CS). When the scene under view is compressible by an algorithm like JPEG or JPEG2000, the CS theory enables us to stably reconstruct an image of the scene from fewer measurements than the number of reconstructed pixels. In this manner, we achieve sub-Nyquist image acquisition. CS theory mainly includes signal sparse representation, encoding measurement and reconstruction algorithm. CS combines sampling and compression into a single non-adaptive linear measurement process. Rather than measuring pixel samples of the scene under view, we measure inner products between the scene and a set of test functions. CS permits the reconstruction of a N-by-N pixel image using much fewer than N-2 measurements. This approach eliminates the need for raster scanning of the object or the terahertz beam, while maintaining the high sensitivity of a single-element detector. We demonstrate the concept using a backward wave oscillator (BWO) which is a continuous-wave terahertz source and get a preliminary test result.