It’s a real drag when you think you’ve taken a great picture, only to load it up in photoshop and discover that your hands weren’t as steady as you thought they were. Depending on the magnitude of your error, chances are you can correct most small camera bumps or pans using a deconvolution filter. The particular technique used depends on which package you use, but they are all built around manipulating the image in the frequency domain to reduce the photo’s linear blurring.
Nathan Willis dissected three applications for removing the effects of camera movement from your photos. Two of them, Refocus and Iterative Refocus, are open source Gimp plugins. The third, Unshake, is a closed source Java application that is capable of producing high-quality results with little user effort (though your CPU will be hurting for a minute or two).
If you watch the movies, you have probably seen the impossibly accurate “computer enhancement” hand-waving that turns a blurry mess into a crystal clear mug shot or license plate for the hero to chase. Real-world image enhancement is not that good, but you may still be surprised at the level of quality a good Fast Fourier Transform and deconvolution can produce.
All three of these applications produce admirable results. Refocus is the fastest, and subjectively Unshake produces the cleanest results. It is unfortunate that among the three alternatives, one is not free software and the other two lack active maintainership. But since the math is well understood, maybe someone will pick up where the other programmers left off, and bring even better refocusing technology to the image editors of tomorrow.
The above photo is from the Unshake site. It seems to work well for predominately straight-line blurs over the range of 8 pixels or less. I haven’t tried the two Gimp plugins, but I have a feeling the Iterative Refocus package could produce the best results given enough tweaking of the setting.
It’s all Fast Fourier Transforms and way over my head, but it works (and frankly, if it was good enough for the Hubble, it’s good enough for me).