The SDO instruments are acquiring a frame every ten seconds. I don't know what frequency they used to make that video, but even if it's showing each frame, it's running at 30 frames per second, so that's at least 300 times real speed. I also don't know which combination of filters were used to make that video, but for the sake of example, let's say it's a composite of videos acquired from three filters, 171Å, 193Å, and 211Å. NASA colors the original data images blue, green, and red respectively.
The first thing you'd need is some video processing software which will save the movie, or a select group of frames, to a series of still frames. Most high-end video software probably does this in a batch, but even the cheap ones will probably let you save any particular frame. (Saving many frames one at a time wouldn't be fun.) Also, there are many common video formats to convert from, and many common image formats to convert to, so those things have to be considered. But, moving on...
Now we have a series of frames from a video saved as individual images. Each pixel in a RGB image contains a mathematical representation of the combination of varying amounts of those three colors. By reversing the math, in a manner of speaking, the original amounts of each color can be determined. This can be easily done with most high-end image processing software like PhotoShop. Once you (or the software) have the original combination you can remove selected amounts of selected colors. So the blue and green are removed entirely from a frame, leaving what started out as just the red. The blue and red are removed, leaving just the green. And the green and red are removed, leaving just the blue.
Save these three individual color copies as separate image files and voila! Barring the lossy compression and artifacts, the three output images are pretty much exactly the red, green, and blue images that we started with. If there are more color combinations in the mix it might become impossible to break them back into their component parts because most imaging software is working with individual pixels as mathematical combinations of red, green, and blue.
So taking a frame from that video and separating out the three colors, we get the three images representing the 211Å, 193Å, and 171Å filters as red, green, and blue respectively. I put those images on the left side of the composite below. Go ten frames later, because there is so little change between successive frames in that video that the changes would be almost imperceptible, and separate the colors, and you get the three images I put on the right side of the composite. Even at ten frames, which would be only a couple of minutes if the movie is using every original data frame, the changes are barely noticeable. (That's why for a running difference image or video they'll typically use frames from hours or days apart.)
The interesting thing about the results is, of course, you can look at the thermal characteristics as individual images. And that is actually what these SDO data images are all about. It's unlikely that anyone at NASA who is studying this stuff technically is terribly concerned with these composite videos. They're looking at the component parts. As has been mentioned several times already, this false color pretty picture stuff is made to impress the public.