An improved segmentation method for multi-material beam hardening correction in industrial x-ray computed tomography. (19th September 2019)

• Record Type: Journal Article
• Authors: Cao, Wenchao; Hawker, Sam; Fardell, Gemma; Price, Ben; Dewulf, Wim
• Is Part Of: Measurement science & technology. Volume 30:Number 12(2019:Dec.)
• Publication Date: 2019-09-19
• Subjects: algorithms -- multi-material -- beam hardening -- dimensional metrology -- x-ray CT; Physical measurements -- Periodicals; Scientific apparatus and instruments -- Periodicals; Equipment and Supplies -- Periodicals; Science -- instrumentation -- Periodicals; Technology -- instrumentation -- Periodicals; Mesures physiques -- Périodiques; Physical measurements; Scientific apparatus and instruments; Periodicals; 502.87
• DOI: 10.1088/1361-6501/ab30bb
• Languages: English
• ISSNs: 0957-0233

Abstract

Abstract: A new beam hardening correction (BHC) method for multi-material industrial computed tomography (CT) is presented. Our method is a statistical learning approach of the Joseph and Spital (JS) segmentation framework. It consists of four steps. First, a multi-dimensional surface and its corresponding hyperplane are simulated based on the x-ray transmission model. They represent the polychromatic and monochromatic forward projection data. A beam hardening correction regression model is trained with the simulated data afterwards. Then the propagation length of the high x-ray absorbing materials is traced from an initial reconstruction. Finally, beam hardening is corrected with the regression model and the corrected image is reconstructed by filtered-back projection (FBP). The accuracy of this method is investigated and compared to that of the conventional JS implementation. Results indicate that the proposed method can achieve comparable performance to the conventional JS method with only of its memory requirement. It is further evaluated on a four-material phantom simulated data set and two experimental dual-material data sets. Results prove that it can not only reduce the appearance of beam hardening, but also increase the dimensional measurement accuracy.