Terahertz imaging with synthetic aperture arrays and adaptive image reconstruction

Buma, TakashiElectromagnetic terahertz (THz) pulses have received considerable interest for imaging applications in security, package inspection, and nondestructive testing. Conventional THz imaging systems employ a confocal geometry, where the illumination and detection are focused to the same location. High lateral resolution requires sharply focused THz beams, but this comes at the expense of an extremely short depth of focus. This leads to long acquisition times, which is a major drawback for practical applications. An alternative THz imaging approach uses phased arrays of transmitter and receiver elements. The generation and detection of coherent THz waves allows synthetic aperture techniques from a single transceiver. Image reconstruction involves combining the signals from a collection of transceiver positions. This thesis demonstrates, for the first time to our knowledge, an adaptive reconstruction method for THz imaging. Our approach improves image quality by computing a coherence factor (CF) based on the spatial coherence of the received THz signals. This CF is computed at every image point, where it suppresses undesirable side lobes for improved image contrast. This thesis also describes an adaptive reconstruction for sparse THz arrays. Sparse arrays employ widely separated elements to dramatically reduce array complexity, making them highly attractive for practical applications. However, sparse arrays suffer from grating lobe artifacts that degrade image quality. The key feature of the adaptive technique is to exploit both the spatial and temporal coherence of the single-cycle THz pulses used by the imaging system. Grating lobe artifacts are reduced by 15 dB without any prior knowledge of the object.University of Delaware, Department of Electrical and Computer EngineeringM.S