(A) novel approach to empirical dual-energy calibration (EDEC) for material decomposition in x-ray imaging

Abstract

Dual-energy x-ray imaging provides information for material identification of an object, which is useful for the industrial x-ray inspection system. However, conventional dual-energy calibration techniques for material decomposition have the limitation of being time-consuming and labor-intensive. Empirical dual-energy calibration (EDEC) method using tomographic image reconstructed by projection images acquired for each angle overcomes the limitations of conventional EDEC method and enables fast and efficient calibration. However, EDEC also has a limitation that there is a risk of obtaining inconsistent projection data due to the case that the object is beyond the calibrated length and the instability of the x-ray beam. To overcome these two limitations, this study proposes a novel empirical calibration method using piecewise linear decomposition functions. This calibration method enables fast and efficient dual-energy calibration while providing stable image decomposition for objects of various sizes and x-ray instability. We performed Monte Carlo simulations using x-rays of 6MV and 9MV to compare the proposed EDEC method with the conventional EDEC method. By evaluating the quality of the material decomposition images obtained for various phantoms, it is demonstrated that the proposed EDEC method improves two limitations of the conventional EDEC method.