Material Identification Using Dual Energy X-ray Absorptiometry [report]

Dongwon Lee, Zachary Thompson
2021 unpublished
Two implementations of dual energy X-ray absorptiometry were studied to identify materials using X-ray attenuation data taken with the Digital Radiography and Computed Tomography (DRCT) systems that were developed for the Recovered Chemical Materiel Directorate (RCMD). Maitrejean et al.'s approach utilizes eigen effects through Principal Component Analysis, while Osipov et al.'s approach proposed a physics-based method. Both approaches approximate mass attenuation coefficients of materials as a
more » ... linear combination of basis functions (eigen effects) or physics-based equations. A set of coefficients {𝑎𝑎 1 , 𝑎𝑎 2 , 𝑎𝑎 3 } or {𝐵𝐵, 𝐷𝐷} were found by parameter optimization in EXCEL Solver. The identification parameters, { } or estimated effective atomic number 𝑍𝑍 ̂ from {𝐵𝐵, 𝐷𝐷}, were calculated to identify material of an aluminum 8 step wedge and a steel 8 step wedge in X-ray radiography images taken by a DRCT system. Maitrejean et al.'s approach was unable to provide reliable 𝑎𝑎 3 𝑎𝑎 1 ratio values for identification of materials. Osipov et al.'s approach was found to be more robust in identify materials with a semi-empirical formula derived from test results in this study. v
doi:10.2172/1825507 fatcat:42xunuyw3vcn5pm434wuvmf3ce