Quarktet's blind deconvolution algorithm, SeDDaRA, has been named as one of Advanced Imaging Magazine's Solution of the Year by unanimous decision! The winning solution is the removal of blur from an image of the background radiation of the universe. A portion of the image and its restoration appear below.

Fig1: Image of the background radiation of the universe as taken by the NASA Wilkinson Microwave Anistropy Probe (WMAP) satellite. The image has been false-colored to highlight the contrast. This image, together with four images in four other spectral bands, allowed NASA/WMAP scientists to more accurately define the age of the universe and the rate of expansion. Run your mouse over the image to reveal the restored figure. Fig 2: Using a fractal image as a model, a blur function, resembling a clover-leaf, was extracted from the image. A pseudo-inverse filter removed the blur to sharpen the image. The image is cleaner and reveals greater resolution. We hope the use of this processing technique will enable scientists to achieve even greater accuracy in their calculations. Image Courtesy of the NASA/WMAP Science Team.

About the Gallery....

Here and on the following pages, you will see more examples of blind deconvolutions performed using the SeDDaRA process. A great advantage of this technique is that it can be applied to any data type, such as consumer photographs, medical images, one dimensional signals, microscopy, and astronomy images. For most, the improvement is obvious. In some cases, the improvement more subtle. SeDDaRA removes the blur that is present in the image (or signal) by extracting the point spread function (or impulse function). Thus, the method returns only the physical aspects of the image, and does not create any new information. This is how the process differs from a standard edge-enhancing filters.

If you have viewed other deconvolution websites, you may have noticed that the images displayed here are significantly larger. When we apply SeDDaRA, we do not hesitate at processing larger images, and have had to crop many of the images on display here. This is a testament to the efficiency of the process. We do, however, apologize for the longer download times.

Fig1: Image of Saturn, taken by the Hubble Space Telescope before the optical correction was applied in December 1993. This image has become a standard image deconvolution example. Run your mouse over the image to reveal the restored figure. Fig 2: Restored Image. Not only is the planet much cleaner, but if you look close enough you will see stars behind the planet that were not apparent before.