Research Information System
Journal of Infrared, Millimeter, and Terahertz Waves, vol.45, no.9-10, pp.841-867, 2024 (SCI-Expanded)
Passive millimeter-wave (MMW) and TeraHertz (THz) imaging systems have become increasingly popular in recent years due to their cost-effectiveness and non-invasive characteristics compared to active systems, prompting a surge in research interest. Evaluating the quality of reconstructed images used in these systems is essential for revealing the fine details. General image quality metrics such as the structural similarity index (SSIM) and the peak signal-to-noise ratio (PSNR) require a reference image in order to compare the reconstructed image. However, there is a notable gap in the literature regarding the evaluation of reconstruction or deconvolution algorithms with a reference image in the passive MMW/THz bands. This study proposes a reference image generation technique for passive MMW/THz imaging systems using an infrared imaging system that shares a similar physical background. Then, passive MMW/THz images were evaluated using the reference images at varying target distances and spatial resolutions. Besides these, the assessment of passive MMW/THz images with the SSIM and PSNR metrics after the reconstruction algorithms were performed. The metrics SSIM and PSNR, are inadequate in the evaluation of reconstruction algorithms alone in terms of concealed object (CO) detection. Because of this reason, the contrast level (CL) method was proposed to address the application-based shortcomings of PSNR and SSIM metrics. Hence, the image quality metric, CL, indicates that the Richardson–Lucy (RL) algorithm yielded superior results in variable optical configurations and target distances with the aid of CL metric. Finally, contrast enhancement techniques were developed in order to increase the contrast level of the CO. As a result, the introduction of these novel methods—the reference image generation technique using an infrared imaging system in passive MMW/THz bands, the evaluation of the reconstructed images with the application-based CL metric, and contrast enhancement techniques for single-band or multi-band imaging methods—holds the potential for the development of innovative techniques. These advancements may contribute to the creation of new applications within the passive MMW/THz bands, particularly focusing on the improvement of detection methods in the future.