Manual MAP processing - a quick tutorial
Hi guys.
I was asked by a forum friend how I performed the manual MAP processing in Registax 3 (or 4). So here's a quick rundown/tutorial of the process I use.
I hope it helps some of you.
MAP processing will usually produce a much sharper image overall than aligning on a single point. This is due to types of seeing where one part of the image may be sharp, but another part may be blurry. Aligning on a single point will produce a final image where THAT part is sharp, but the rest of the image is not likely to be as pleasing.
These "waves" of fluctuating seeing are common, and so aligning on multiple points and combining the result ensures that each alignment point produces a sharp area of the image, and therefore the final image should be sharper overall.
Here's how I do it.
I run the avi through registax as normal, choosing a single alignment point, going through the normal alignment/optimise/stack/wavelets tabs, and save the image as a TIF file with some minor wavelets processing. I then run the same avi through registax again, aligned on a different point, same process. Repeat this for as many alignment points as you like. You'll end up with a TIF file for each alignment point.
Once i've done enough points, I open up all images in Photoshop CS2.
I choose one of the images to be the "base", and I copy each other image one at a time and paste it as a new layer on top of the base (remember to re-save as a PSD file).
I make all layers except the base invisible. I then, one layer at a time, working from the bottom up, make the layer visible, move it so it's aligned over the base layer, and look for the sharpest parts of both layers by using the visible/invisible toggle. I erase/delete the parts of the new layer where I want to reveal the layer beneath (where it's sharper).
I then make the next layer up visible, and repeat the same process.
Some layers may require some levels adjustment so that the joins are not evident. I also use a feather of 2px when selecting the areas to remove, so it blends better with the layer beneath.
By the end, all layers are visible and the sharpest parts of each are on top, while all other parts are deleted. I then save the combined PSD, and then re-save as a TIF (no layers), to take into other programs as required for sharpening or other post-processing.
I hope that helps!
Any questions, please ask!
|
