Post-production... it's a necessary evil. Since I scan all of my own artwork, I have to know Photoshop pretty well (or at least aspects of it) in order to create a file that will reproduce faithfully. I've learned a lot over the years, but it's been a process of trial and error.
The benefits of scanning your own artwork are well worth the time it takes. First and foremost, your originals never leave your hands and therefore are not at risk of being lost, damaged, or sent away to the wrong person. These things happen... sometimes to me, but not anymore. Additionally, you get a second chance at correcting any mistakes before you send it off. This also happens to be a disadvantage.
Sometimes I find myself — and I'm not alone here — spending way too much time agonizing over the most trivial details and color balances, when the end result is not of much importance. Often, I will let a piece sit around for a day before I scan or retouch it. This gives it a freshness to my mind's eye that lets me zero in on what's important.
But enough introduction. What do I do once I've got the image scanned? Actually, let's start with how I scan it. Someday I'll be able to afford the Epson 10,000 XL scanner, but for now my letter-sized scanner will have to make do. I scan the page, which is as wide as a letter-sized paper is long, in 3 sections. It works for the most part, but the edges warp and dim as they exceed the prescribed boundaries of the scanner. Luckily for me, the newest version of Photoshop, CS3, has an improved Photomerge function that has made my life much, much easier.
You can find this modern miracle under File> Automate> Photomerge. The previous version was almost worthless as it made no attempt at warping geometry to line up, let alone color matching. Thank you, Adobe.
Once in the computer, I make sure all my specs are correct, i.e. dimensions, resolution, and color gamut. I then create a separate layer for borders. The borders are created with a square brush tool, while holding down the shift key, which gives me a straight line. In addition to that, using the "Layer Style" palette, I add a black stroke to the outside of my borders, usually around 8 pixels wide. In this case, it makes no difference since the border and gutter are the same color, but it matters on other pages.
Then comes the hard part, which I've shown below. I use adjustment layers to tweak everything from value to saturation. Often, one all encompassing adjustment is not enough, so I use layer masks to isolate individual panels. I'm not going to go into the details of exactly what I do, but if anyone has any specific questions, feel free to ask. I'll gladly answer in the comments, or possibly get an idea for a future post.
One other tool I should mention is the "Liquefy" filter which I use to gently nudge pixels around. This is especially helpful on faces, where the slightest change can result in a different expression. I also utilize this in the beginning stage of a color study, which I'll show in a future post. With this page, I used it to correct distortion in the woman's face and reduce the size of Steve Rogers ear (it looked good in the pencil drawing, but paint can make cartooned exaggeration look awkward).
And so here it is, the final result. I've included the original scan to show you just what a dramatic difference Photoshop can make. There were several color corrections that I wanted to make in the painting, but I simply ran out of time. Luckily for me, I live in the 21st century and can use advanced tools to make myself look better.
I hope you've all enjoyed this in-depth look at the creation of a painted page. And again, if it's sparked more questions than it has answered, please feel free to ask.