In this tutorial we will take an oversized image and scan it in two passes. Once in PShop, we will stitch it together leaving it seamless.

Warning: I am using a drawing from my sketch pad. I am not an artist, thusly it is not an overly attractive drawing.

I am assuming scanner knowledge, if I use a word you don’t understand with regard to your scanner, consult your manual.

Step 1: Correct document placement

There are several sizes of paper in the US that are larger than a scanner, including, and this is my favorite, comic illustration board. It is important that we are on the same page, so-to-speak, with regard to what I am using for this. I have an oversized sketchbook that the drawing is coming out of, and it is about 60lb paper, so it is maleable. This will help with bending the paper onto the platen with the help of the lid.

I am going to put the paper on sideways to get 2 scans, there are several variations on this, depending on your paper size. You should ALWAYS have your paper on a straight edge or prefferably 2. I am starting like so:

The paper that I am using abutts the scanner on the bottom and one the left. There is no correct end to start on, just whatever you are comfortable with.

Step 2: Overscan

In this step, we need to look at our scanner dialog box and determine what is going to be our finished size. For the sake of swiftness let us assume 100% scale and 300dpi. Typically, it is a good idea to increase your resolution to 2x to 2.5x the finished size and then downsample. I am not going to do that, but I may cover super-sampling later as its own Tut. TURN OFF AUTO CORRECT AND/OR AUTO COLOR, and be sure that you are using the same settings for both. We have enough variables.

We must scan to the extents of the paper and over when choosing with the selection rectangle. This allows us some flexability in piecing the scans together. In my example, I can’t overscan the left or the bottom, so I am adding to the top and getting the entire width, even though the right is going to be messy because of the bend.

Step 3: Merging

Ahhh, the fun part. I can’t overstate the importance of good scans, but this is where the magic happens. Ok, maybe not magic. With both docs open, you must choose which you will merge into. You must not correct color or value yet, we have to have them to compare. I am choosing my first doc. So, since I am adding to the Right(I scanned upside down) I will open Image>Canvas Size like so:

What we want to do is make the document wider, and move the current art to the left making room for new. So we choose to anchor to the left and make the canvas 14 inches. Settings applied:

Note: I left the background color at the bottom. You can change this to whatever you like, some people like to use an incendary color to emphasize a semi-tranparent layer. I am not so hot on this idea, but to each his/her own.

Step 4: Copy/Paste

With our canvas larger, we need to select our other image with a click to activate it, and a Select All to select the extents of the doc. Now we Edit>Copy to pop it on the clipboard, and activate the other doc. Pasting it will result in a layers pallette like this:

We now use the Move tool to slide it to the right, and adjust our layer opacity to about 50%. This leaves us with a fuzzy image in the middle. But we are getting close to done.

Step 5: Alignment

Now we have 2 layers, and the top is 50% opaque, and the bottom is immovable, so lets move the top. Select the top layer with Select> Select All and we are going to move it to a specific point. As you can see from our illustration, we are very close to being aligned, so lets see where we need to be:

We use the nudge controls, which are the arrow keys to move the selected image until it lines up almost perfectly. Please note that you should not trust alignments that are made to close to either edge, as they are not always scanned linearly and will be slightly elongated.

OK, here we are closer:

This image is slightly askew, even with our attempts to keep it square. It won’t win however, we have a technique just for this. We bend it to our will using the Edit>Free Transform tool and setting the axial point on the aligned part.

Note the Axial point on the left, but not to close to the scan edge. The image is mostly aligned now, and the is an easy shortcut from here to finish up. I have read several scanner and pre-press books that discuss the laborious process of meticulous alignment. F-that.

Step 6: Layer Mask

Lets bring our opacity back up by selecting, in the layers palette, the top layer, and sliding opacity from 50% to 100 percent. While looking at out image, we notice that on the left we have a bit of an artifact from the scan. It is a gradient from where the paper didn’t fit and had to be folded. We can assume that this part of the image is elongated a bit as well so lets do away with it entirely. We click on the layer mask button in the layer pallette giving us this:

Notice that the layer mask is selected by default. This is good and will make our next task quicker.

Step 7: Gradient Mask

The gradient tool is what we will be using on the mask, so lets pick it and look at the options:

We must make sure that we are using Black and White in the gradient, and that it is linear. The mode is normal and the last 3 check boxes(not pictured) are UNchecked.

This yields a tool that will go from black to white in the gradient, which when used with the mask, will give us a transparency grad to the top image. Use it somewhat like this:

You should start the drag on the INSIDE of the scan edge, so that it will be totally invisible owing to the black part of the alpha mask.

Step 8: Correction:

When I zoom into my doc, I can see where my paper texture made it in the bottom scan, but not on the top, so I will have to use levels to correct this. Lets do it in an adjustment layer to ensure that we can change it or even add a mask to it. Select the bottom layer, this enables us to add the Adj. layer and it immediatly goes above the bottom, but below the top, so it doesn’t affect the top. Here is the Levels Adj Layer and the Dialog with the change:

Now my document is solid and looks seamless, but I still have a few more things to clean up. I scanned in color, mostly out of habit, but that ensures that I have maximum information to work with, just in case. With this drawing, I am going to 1-bit mode to save it, to keep the size down, and make it very sharp in print. But that is another tutorial. This is the finished product:

Thanks,

Mike Anderson

Ok, This is for my friend Alex, but I am posting it here for my homies. Take it or leave it.

Step 1: Scan - when scanning normal gray art, you should make sure you are using these settings:

100% scale
Grayscale
600-900dpi
full scanner bed(this is chosen visually usually)

I am abb. this assuming anyone wanting to downsample should know the basics of a scanner. I know… With adequate feedback, I will cover microtek scanners and perhaps HP

Step 2: In Pshop, the software of choice, you need to see what you have, so lets look at: Image>Image Size and it should look something like this:

In this dialog there are 3 key pieces of information that we need to concentrate on…

1. Resample needs to be ON to downsample

2. Constrain needs to be ON to downsample evenly

3. Resolution is the box that we are going to change

Step 3. Enter your target resolution, in this case we are D/S to 400dpi in grayscale mode. Like So:

It is important to note the pixel dimension change and that the document size in inches (or whatever) HASN’T changed.

This concludes my Tutorial on Downsampling. Please remember that no matter what anyone says, you SHOULD NEVER UPSAMPLE… EVER. If they tell you otherwise, they are trying to sell you something.