MV
M.B. Vorst
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
1 records found
1
This thesis presents the integration and evaluation of a 24-pixel silicon-germanium bipolar complementary metal-oxide-semiconductor terahertz camera in an active illumination transmission imaging setup. The camera had previously been characterized at detector level, but its performance as part of a complete imaging system still had to be evaluated. The main objective was therefore to determine whether the camera could produce repeatable and interpretable images when combined with the available quasi-optical measurement system. Two MATLAB-based graphical user interfaces were developed. The first interface predicts the expected image resolution for a selected camera configuration, frequency, and sample image. This allows for the expected visibility of object features to be estimated before measurement. The second interface controls the measurement workflow by configuring the source, detector bias supply, readout electronics, data-acquisition hardware, and mechanical positioning system. It provides hardware selection, connection testing, scan definition, live image preview, metadata storage, and image reconstruction from the measured pixel responses. Active imaging measurements were performed on an ivy leaf sample. The measured detector responses were mapped to spatial scan positions, converted to decibels, normalized, and then reconstructed into terahertz transmission images. The main result was obtained around 400 gigahertz, where the leaf outline and central vein were visible in the reconstructed image. Frequency averaging over five neighboring frequency points reduced background variations such as standing waves, while preserving the main leaf structure. The results show that the integrated system can produce interpretable active terahertz images and that the developed workflow supports controlled and reproducible measurements. However, the demonstrated performance is still limited by mechanical scanning time, optical alignment uncertainty, standing-wave artifacts, and pixel-dependent response differences.
...
This thesis presents the integration and evaluation of a 24-pixel silicon-germanium bipolar complementary metal-oxide-semiconductor terahertz camera in an active illumination transmission imaging setup. The camera had previously been characterized at detector level, but its performance as part of a complete imaging system still had to be evaluated. The main objective was therefore to determine whether the camera could produce repeatable and interpretable images when combined with the available quasi-optical measurement system. Two MATLAB-based graphical user interfaces were developed. The first interface predicts the expected image resolution for a selected camera configuration, frequency, and sample image. This allows for the expected visibility of object features to be estimated before measurement. The second interface controls the measurement workflow by configuring the source, detector bias supply, readout electronics, data-acquisition hardware, and mechanical positioning system. It provides hardware selection, connection testing, scan definition, live image preview, metadata storage, and image reconstruction from the measured pixel responses. Active imaging measurements were performed on an ivy leaf sample. The measured detector responses were mapped to spatial scan positions, converted to decibels, normalized, and then reconstructed into terahertz transmission images. The main result was obtained around 400 gigahertz, where the leaf outline and central vein were visible in the reconstructed image. Frequency averaging over five neighboring frequency points reduced background variations such as standing waves, while preserving the main leaf structure. The results show that the integrated system can produce interpretable active terahertz images and that the developed workflow supports controlled and reproducible measurements. However, the demonstrated performance is still limited by mechanical scanning time, optical alignment uncertainty, standing-wave artifacts, and pixel-dependent response differences.