filter overshoots and undershoots from a png

[0_o]

Hi,

I want to filter overshoots/undershoots from a png image or a 2d array.

However, LV has only 1d filters.

The implementation I'm working on is a Differential Interference Contrast Microscope where I take two interference images and run them through an algorithm based on a Kalman Filter.

The overshoots I'm dealing with appear around the edges of a scanned feature (nanotube for example).

If I run a profiler over such an edge I would see a noisy flat area before the edge, overshoot, undershoot and then another noisy flat area higher than the first one.

I want the image to have two noisy flat areas with a step between them.

Any ideas?