摘要
Diffraction-enhanced imaging （DEI） is a powerful phase-sensitive technique that provides higher spatial resolution and supercontrast of weakly absorbing objects than conventional radiography. It derives contrast from the X-ray absorption, refraction, and ultra-small-angle X-ray scattering （USAXS） properties of an object. The separation of different-contrast contributions from images is an important issue for the potential application of DEI. In this paper, an improved DEI （IDEI） method is proposed based on the Gaussian curve fitting of the rocking curve （RC）. Utilizing only three input images, the IDEI method can accurately separate the absorption, refraction, and USAXS contrasts produced by the object. The IDEI method can therefore be viewed as an improvement to the extended DEI （EDEI） method. In contrast, the IDEI method can circumvent the limitations of the EDEI method well since it does not impose a Taylor approximation on the RC. Additionally, analysis of the IDEI model errors is performed to further investigate the factors that lead to the image artifacts, and finally validation studies are conducted using computer simulation and synchrotron experimental data.
Diffraction-enhanced imaging （DEI） is a powerful phase-sensitive technique that provides higher spatial resolution and supercontrast of weakly absorbing objects than conventional radiography. It derives contrast from the X-ray absorption, refraction, and ultra-small-angle X-ray scattering （USAXS） properties of an object. The separation of different-contrast contributions from images is an important issue for the potential application of DEI. In this paper, an improved DEI （IDEI） method is proposed based on the Gaussian curve fitting of the rocking curve （RC）. Utilizing only three input images, the IDEI method can accurately separate the absorption, refraction, and USAXS contrasts produced by the object. The IDEI method can therefore be viewed as an improvement to the extended DEI （EDEI） method. In contrast, the IDEI method can circumvent the limitations of the EDEI method well since it does not impose a Taylor approximation on the RC. Additionally, analysis of the IDEI model errors is performed to further investigate the factors that lead to the image artifacts, and finally validation studies are conducted using computer simulation and synchrotron experimental data.
基金
Project supported by the National Natural Science Foundation of China(Grant Nos.30900333 and 81371549)
the WBE Liver Fibrosis Foundation,China(Grant No.CFHPC20131033)
