Applications of a Wavelet-based Filtering and Deconvolution Technique

Previous abstract Next abstract

Session 40 -- Numerical Techniques and Models
Display presentation, Tuesday, 10, 1995, 9:20am - 6:30pm

[40.08] Applications of a Wavelet-based Filtering and Deconvolution Technique

S.E. Vance (Andrews University), J.E. Grindlay (CfA)

A wavelet-based filtering technique is applied to background images of the Energetic X-Ray Imaging Telescope Experiment (EXITE). The resulting images preserve both low spatial frequency and statistically significant high spatial frequency components. Analysis of EXITE Crab Nebula data using filtered background images shows gains in signal-to-noise ratio (SNR) and reductions in the ratio of the root-mean-square noise to the expected Poisson fluctuations when compared with analysis using non-filtered background images. A description of this wavelet based filtering algorithm is given. In addition, examples of filtered and non-filtered background images are shown.

Likewise, a wavelet-based (or regularized) Lucy-Richardson deconvolution technique is applied to both a ROSAT HRI observation and a ROSAT PSPC observation of the eastern lobe and jet of SS433. The resulting deconvolved images reveal faint, large-scale structure, not clearly visible in the original images, and maintain small-scale statistically significant details. A description of this regularized Lucy-Richardson deconvolution algorithm is given. Original images and examples of deconvolved images are shown and compared with a published Einstein observation.

Tuesday program listing