Enhancement of image quality by utilization of a priori information

摘要

The quality of incomplete data Non-Destructive Evaluation and Computed Tomography images is improved by incorporating a priori information into the image reconstruction and image processing to supplement the available data. The a priori information is provided by electronic models of the part derived from a solid modeler, physics of the inspection process, and outputs of touch and other sensors. Methods of improving limited-angle X-ray CT images are given. Calculated projection data in the missing angular range is provided by calculating x-ray path lengths through a solid model of the part, and x-ray attenuation from known physical parameters of the part and source. The measured and calculated projection data are combined to reconstruct the CT image. In an iterative reconstruction approach, precise boundary information from a model and calculated attenuation are information to improve the limited angle image.