Complex Analysis of Steerable Filters

Dizdaroglu B.

21st Signal Processing and Communications Applications Conference (SIU), CYPRUS, 24 - 26 Nisan 2013 identifier identifier


Extraction of image features in digital image processing approaches is one of the most important preprocessing steps. In this paper, steerable filters, developed on complex derivative approaches, have been investigated. Here, the steerable filters are able to estimate optimal image structures, even if the image contains artifacts such as noise. But it is necessary to generate the resulting image via steerable filters based on both the rotation and scaling operations and such processes are more advantageous in the complex plane. Therefore, characteristics of Gabor filter, a steerable filter, as an example has been examined depending on the different values of parameters. The visual results for the output of this filter are also given.