larger the radius the more blurred the image becomes until the pixels
are homogeneous (by default <math>n=16</math>).
* '''Sobel filter''': calculates the [http://en.wikipedia.org/wiki/Sobel_filter gradient at each pixel]. [http://en.wikipedia.org/wiki/Gaussian_blur Gaussian blurs] with <math>\sigma=1, 2, 4... 2^{n}</math> are performed prior tothe filter.
* '''Hessian''': Calculates a [http://en.wikipedia.org/wiki/Hessian_matrix Hessian matrix] <math>H</math> at each pixel. Prior to the application of any filters, a [http://en.wikipedia.org/wiki/Gaussian_blur Gaussian blur] with <math>\sigma=1, 2,
4... 2^{n}</math> is performed. The final features used for pixel classification, given the Hessian matrix <math>\left(\begin{array}{cc}