Multi-energy (ME) detectors are becoming a serious alternative to classical dualenergy sandwich (DE-S) detectors for X-ray applications such as medical imaging or explosive detection. They can use the full X-ray spectrum of irradiated materials, rather than disposing only of low and high energy measurements, which may be mixed. In this article, we intend to compare both simulated and real industrial detection systems, operating at a high count rate, independently of the dimensions of the measurements and independently of any signal processing methods. Simulations or prototypes of similar detectors have already been compared (see [1] for instance), but never independently of estimation methods and never with real detectors. We have simulated both an ME detector made of CdTe -based on the characteristics of the MultiX ME100 and -a DE-S detector -based on the characteristics of the Detection Technology's X-Card 1.5-64DE model. These detectors were compared to a perfect spectroscopic detector and an optimal DE-S detector.For comparison purposes, two approaches were investigated. The first approach addresses how to distinguise signals, while the second relates to identifying materials. Performance criteria were defined and comparisons were made over a range of material thicknesses and with different photon statistics. Experimental measurements in a specific configuration were acquired to checks simulations. Results showed good agreement between the ME simulation and the ME100 detector. Both criteria seem to be equivalent, and the ME detector performs 3.5 times better than the DE-S detector with same photon statistics based on simulations and experimental measurements. Regardless of the photon statistics ME detectors appeared more efficient than DE-S detectors for all material thicknesses between 1 and 9 cm when measuring plastics with an attenuation signature close that of explosive materials. This translates into an improved false detection rate (FDR): DE-S detectors have an FDR 2.87 ± 0.03-fold higher than ME detectors for 4 cm of POM with 20 000 incident photons, when identifications are screened against a two-material base.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.