AI Article Synopsis
- The paper discusses issues with reconstructing 3D object distributions from tomographic images, highlighting the problem of noise amplification in certain frequency components.
- It introduces a generalized matrix inversion technique aimed at controlling this noise amplification to match the spatial resolution of the imaging system.
- Results from applying this method to a simulated positron camera show a notable enhancement in background elimination.
Article Abstract
A major problem with the reconstruction of three-dimensional object distributions from focused tomographic images using Fourier transforms is the amplification of statistical noise in certain frequency components. This paper describes the use of a generalised matrix inversion technique to limit noise amplification to a level related to the spatial resolution of the imaging system. The reconstruction method is applied to a simulated positron camera, and results are presented on the imaging of an extended, three-dimensional object distribution. A significant improvement in the elimination of the background is achieved.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1088/0031-9155/23/2/002 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!