Image Reconstruction from Sparse-View Data Acquired with Portable X-Ray Devices

Portable X-ray imaging devices can be used for obtaining information of 3D tomographic structures within an object of interest, which can be of utility in inspection of suspicious packages and explosive materials. In a practical application of portable X-ray imaging device for obtaining 3D tomographic images, it is always desirable to collect data at sparse views for shortened scanning time and for accommodation of specific workflow. Standard reconstruction algorithms such as the FDK method generally require projection data collected at a substantial number (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$100 \sim 300$</tex>) of views for obtaining artifact-free images and thus yield images with artifacts when applied to sparse-view data. We have been developing a portable imaging system for inspection and analysis of potential threat objects, where imaging time and flexibility are crucial. In this work, we have investigated image reconstruction from sparse view data by use of the total variation (TV) algorithm tailored specific to our scanning conditions. The result of our study reveals that the TV algorithm can reconstruct 3D images of our application utility from sparse view data, thus enabling fast and flexible imaging specific to the application workflow needs. The work may provide insights into the development of fast and flexible scanning approach of potentially broad applications in field use, security screening, and emergency response scenarios.