Abstract
Ring artifacts in computed tomography (CT) images are introduced by detector element's response nonuniformity. Ring artifact degrades the quality of CT image, influences the determination of defect, and hampers post processing and prohibiting quantitative analysis. Two algorithms for ring artifact correction are proposed based on feature recognition and B-form of the spline curve fitting. The former is for the situation that the response of a single detector element or few continuous detector elements is obviously nonuniform. According to the algorithm, the position of the detector element whose response is nonuniform is determined. And a local median filter is executed at the position to depress the vertical streaking in the sinogram. The latter is for the situation that many continuous detector elements' response has small nonuniformity. In this method, the normalized gain factors of the detector elements are gotten through B-form of the spline curve fitting. So detector element's response nonuniformity can be corrected. The two algorithms have their own characteristics which supply each other and compose together a complete ring artifact correction system. The feasibility of the two methods is validated by the experimental data.
| Original language | English |
|---|---|
| Pages (from-to) | 1849-1853 |
| Number of pages | 5 |
| Journal | Guangxue Xuebao/Acta Optica Sinica |
| Volume | 29 |
| Issue number | 7 |
| DOIs | |
| State | Published - Jul 2009 |
Keywords
- B-form of the spline curve fitting
- Computed tomography
- Feature recognition
- Ring artifact
- Sinogram
- X-ray optics
Fingerprint
Dive into the research topics of 'Two ring artifact correction methods for computed tomagraphy'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver