I call the first step "crop the crap."

There's plenty of "crap" in these images already due to the fact that the RGB layers don't overlap perfectly (this may also be touched on in a future tutorial in an attempt to rescue some of the lost data there). Also, if we've been through the second tutorial we've added a lot of "crap" ourselves with our padding and distorting and pixel marking and whatnot. Basically anything at the edge of the picture that doesn't look like it belonged in the original picture can be cropped out.

You can use the cropping tool for this, or do as I do and select the good part with the rectangular marquee, then use Image>Crop. The results are the same.

Previous page | Next page