Use this URL to cite or link to this record in EThOS:
Title: On-belt Tomosynthesis : 3D imaging of baggage for security inspection
Author: Kolokytha, S.
ISNI:       0000 0004 8503 6212
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This thesis describes the design, testing and evaluation of 'On-belt Tomosynthesis' (ObT): a cost-effective baggage screening system based on limited angle digital X-ray tomosynthesis and close-range photogrammetry. It is designed to be retrofitted to existing airport conveyor-belt systems and to overcome the limitations of current systems creating a pseudo-3D imaging system by combining X-ray and optical imaging to form digital tomograms. The ObT design and set-up consists of a configuration of two X-ray sources illuminating 12 strip detectors around a conveyor belt curve forming an 180° arc. Investigating the acquired ObT X-ray images' noise sources and distortions, improvements were demonstrated using developed image correction methods. An increase of 45% in image uniformity was shown as a result, in the postcorrection images. Simulation image reconstruction of objects with lower attenuation coefficients showed the potential of ObT to clearly distinguish between them. Reconstruction of real data showed that objects of bigger attenuation differences (copper versus perspex, rather than air versus perspex) could be observed better. The main conclusion from the reconstruction results was that the current imaging method needed further refinements, regarding the geometry registration and the image reconstruction. The simulation results confirmed that advancing the experimental method could produce better results than the ones which can currently be achieved. For the current state of ObT, a standard deviation of 2mm in (a) the source coordinates, and in (b) the detector angles does not affect the image reconstruction results. Therefore, a low-cost single camera coordination and tracking solution was developed to replace the previously used manual measurements. Results obtained by the developed solution showed that the necessary prerequisites for the ObT image reconstruction could be addressed. The resulting standard deviation was of an average of 0.4 mm and 1 degree for (a) and (b) respectively.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available