Resolution-Enhanced Polarimetric Terahertz Imaging

• Published in: IEEE transactions on terahertz science and technology Vol. 14; no. 5; pp. 675 - 690
• Main Authors: Hossain, Maruf Md. Sajjad, Nahar, Niru K., Sertel, Kubilay
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01-09-2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Anisotropic; Biomedical imaging; Composite materials; Horn antennas; Image quality; Image resolution; Imaging; Lenses; Linear polarization; Medical imaging; Millimeter wave communication; Millimeter waves; millimeter-wave; Polarimetry; Reflection; Reflectometry; reflectrometry; Sensors; Signal generation; Statistical analysis; terahertz; Terahertz communications; Terahertz frequencies; Terahertz wave imaging
• ISSN: 2156-342X; 2156-3446
• DOI: 10.1109/TTHZ.2024.3430040

We present a novel millimeter-wave and terahertz-band imaging system capable of sensing the full polarization matrix of the signals that are locally reflected by a sample under test. A high-resistivity extended-hemispherical lens further provides 3.4× improved pixel resolution, leading to a powerful fully polarimetric imaging sensor. The generation and sensing of millimeter-wave/terahertz signals are based on commercially available VNA frequency extension modules interfaced with diagonal horn antenna. As such, the transmitted and received signals can be controlled to exhibit pure linear polarization. With a precisely aligned quasi-optical link consisting of 90° off-axis parabolic mirror, a wire-grid polarizer and the high-permittivity lens, the beam can be tightly focused at the sample plane, resulting in an unprecedented resolution as fine as 50 <inline-formula><tex-math notation="LaTeX">\mu</tex-math></inline-formula>m for the WR-1.0 band (750-1100 GHz). To demonstrate the utility of the proposed fully polarimetric THz reflectrometry, copolarized and cross-polarized reflection images of a diverse set of FFPE brain tissue samples are captured and the image quality, as well as statistical properties are discussed. The effectiveness of THz fully polarimetric imaging in capturing the anisotropic features in a sample by discriminating edges based on the polarization is demonstrated. This new approach can find wide application areas in nondestructive imaging, such as crack and delamination detection in multilayered composite materials, and diagnosis of degenerative diseases.