[Previous] | [Session 63] | [Next]
M. Karovska (Harvard-Smithsonian CfA)
Recent high-angular resolution ground- and space-based observations have produced unprecedented views of various astronomical objects at wavelengths ranging from Gamma Rays to Radio. These observations contain many complex components with different spatial scales and a wide range of contrast levels. Spatial characterization of these components is extremely important for understanding the physical characteristics of these sources. However, the multi-scale structures in these images cannot readily be studied directly, either because of the limited resolution and noise in the images, or because of the low contrast of the small-scale structures when compared to the large-scale features. Even the most spectacular high-angular resolution images are blurred by the telescope Point Spread Function (PSF), which often varies as a function of the off-axis angle and wavelength, and by instrumental effects including the limited detector pixel sizes. Image restoration including deconvolution techniques offers a powerful tool to improve the resolution in the images, and to extract information on the small-scale structure stored in the astronomical observations. I will review several deconvolution techniques including techniques that can be applied when the PSFs are well known. I will also address the accuracy of the restored images and the problem of assigning statistical significance and confidence levels to the results. To illustrate the current problem in accuracy estimation, I will show examples resulting from the application of several deconvolution techniques to a sample of astronomical images with different statistics.
This work was supported in part by NASA contract NAS8-39073 (Chandra X-ray Center).
[Previous] | [Session 63] | [Next]
Bulletin of the American Astronomical Society, 34, #4
© 2002. The American Astronomical Soceity.