Rapid Imaging of Pulsed Terahertz Radiation with Spatial Light Modulators and Neural Networks

We report the first use of a convolutional neural network for terahertz (THz) imaging combined with a single-pixel camera to achieve high quality hyperspectral THz imaging 10x faster than the current commercial systems, with faster potential only limited by noise. Imaging with a spatial light modulator (SLM) relies on projecting a set of spatial patterns onto an object and recording the transmission with a single-pixel detector, and an optical delay unit (ODU) is used to obtain the terahertz time-of-flight information. A key breakthrough in this work is to synchronize the equipment to not need any time to instruct the ODU to move while projecting the patterns.

Automated summary

The research reports the first use of a convolutional neural network in terahertz (THz) imaging with a single-pixel camera, achieving high-quality hyperspectral THz imaging 10 times faster than current commercial systems, limited only by noise. The key breakthrough of synchronizing the equipment eliminates the need for additional time to instruct the ODU to move while projecting patterns.