Abstract
Prior work indicated that view synthesis methods can reduce the hardware requirement of the kinetic depth effect X-ray (KDEX) imaging technique used for aviation luggage security screening. One of the greatest challenges in view synthesis is to solve the correspondence problem. Scale invariant feature transform (SIFT) has recently gained substantial attention in the computer vision community to address the problem. Due to fundamental differences between visible light images and X-ray images, this paper discusses the feasibility of applying SIFT to transmission X-ray images. Empirical results revealed that SIFT has promising potential to search for X-ray image correspondences. The transparency property of X-ray images provides additional information, which can also constrain the matching process under certain conditions. Future work will focus on improving the SIFT technique to synthesise colour KDEX images produced by materials discriminating dual-energy X-ray sensors.