Increasing the Fine Structure Visibility of the Hinode SOT Ca ii H Filtergrams

We present an image processing technique, the improved OMC filter (also called Madmax), which selects maxima of convexities of intensity modulations of an image. The filter computes second derivatives of the image in multiple directions around each pixel. It is shown to be efficient for pattern recognition, and bright hair-like or small anisotropic features can be enhanced. The filter is tested on artificially generated images, and the effect of a different number of directions in which the second derivatives are calculated is evaluated. Ca ii H line images of spicules and prominences obtained with Hinode/SOT are also used to illustrate its efficiency on real data. It is shown how to make the choice of the sensitive parameters to be used in improving the image visibility.