next up previous
Next: Template Matching as Testing Up: The Imaging Process Previous: Image resampling

Log-polar imaging

While the representation of image as rectangular arrays is the most widespread, alternative exists. Log-polar mapping, described in Section TM:2.4 is an example of space variant representation providing a spatial resolution decreasing from the center towards the boundary.

Log-polar mapping is usually obtained with specific sensors but can also be simulated via software to appreciate some of its characteristics. When we change the image representation from a (standard) rectangular lattice to a log-polar spatio-variant structure we are effectively performing a resampling task which is prone to aliasing if frequency content is not properly tuned. As log-polar sensor resolution is spatio-variant so must be the amount of smoothing applied before resampling. Usually, the two sensing structures are compared for a given maximum spatial resolution. In the case of a rectangular sensor the maximum resolution extends over the whole sensor while in the case of a log-polar sensor it is confined to the central (foveal) region.


53#53

Figure 2.9: Log-polar mapping, providing image representation with spatio-variant resolution, requires a complex smoothing pre-processing. The images in the bottom row present the resulting map (left) and its back-projection in the standard image representation.
[width=8cm]figures/lpOrig.jpg [width=8cm]figures/lpSmooth.jpg
[width=8cm]figures/lpMap.jpg [width=8cm]figures/lpEquiv.jpg


next up previous
Next: Template Matching as Testing Up: The Imaging Process Previous: Image resampling
Roberto Brunelli 2008-11-25