A new approach for beam hardening correction based on the local spectrum distributions

Rasoulpour, Naser; Kamali-Asl, Alireza; Hemmati, Hamidreza

doi:10.1016/j.nima.2015.05.015

Cited by 6 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the monochromatic projections data were directly approximated and then beam hardening was corrected. Rasoulpour et al (2015) proposed a correction approach based on the attenuation coefficient of the local spectrum distribution where the corrected attenuation coefficient of mean energy can be calculated according to the expectation-maximization algorithm of the arbitrary thickness of known materials in the object. In this approach, a primary source spectrum is required to calculate the local spectrum in various depth.…”

Section: Introductionmentioning

confidence: 99%

A Review of Common Beam Hardening Correction Methods for Industrial X-ray Computed Tomography

Ahmed¹,

Song²

2018

JSM

View full text Add to dashboard Cite

X-ray computed tomography (XCT) became an important instrument for quality assurance in industry products as a non-destructive testing tool for inspection, evaluation, analysis and dimensional metrology. Thus, a high-quality image is required. Due to the polychromatic nature of X-ray energy in XCT, this leads to errors in attenuation coefficient which is generally known as beam hardening artifact. This leads to a distortion or blurring-like cupping and streak in the reconstruction images, where a significant decrease in imaging quality is observed. In this paper, recent research publications regarding common practical correction methods that were adopted to improve an imaging quality have been discussed. It was observed from the discussion and evaluation, that a problem behind beam hardening reduction for the multi-materials object, especially in the absence of prior information about X-ray spectrum and material characterizations would be a significant research contribution, if the correction could be achieved without the need to perform forward projections and multiple reconstructions.

show abstract

Section: Introductionmentioning

confidence: 99%