Lighting

The concept is simple, but the implementation is not always so. The concept is that you want even light on your piece.

In the example to the left, of a framed photograph (with a gray card above it), the light is pretty even, but the light source is causing a reflection on the glass. If your art will not generate any reflections, as might be the case with some types of paper, a camera's built-in flash may work just fine. But anything with dimensionality, like oil paint, anything with glass, or anything with even the slightest reflectivity will bounce direct flash right back at the camera.

And, as in this example, a highly reflective surface like glass will reflect not only the light, but objects. You might see yourself standing there, holding the camera. That wouldn't be so good.

In this example I've moved the light source and camera to an angle that would prevent their reflections from showing. I've solved the reflection problem but I've introduced two new problems. One is that the light is uneven. The other is that the picture is no longer rectangular in the image.

While it is sometimes possible to even out the lighting in a computer program like Photoshop, unless you're very proficient at such manipulation, it's much easier to make the light even in the first place.

However, it is fairly easy to make the picture rectangular in Photoshop, so we'll accept the angle but correct the light.

How do you do that? Light loses energy with distance, which is why a spotlight is brighter on a nearby object than a distant one. In fact, light loses energy very quickly, so if you're shining one light on your piece from the side, as in this example, the other side will be noticably darker.

One solution is to use sunlight, because the difference between the distance from the sun to the left side of your piece and from the sun to the right side of your piece is pretty darned small. With sunlight, be cautious of reflections though.

Overcast light can be ideal. But if the "overcast" is caused by being on a porch or other shaded area, the light may not be even.

Another solution is to use two lights, one from each side. If you have access to a copy stand (see illustration), and if your piece will fit on it, this is a great solution, but it will not solve the reflection problem. If you can remove the glass, a copy stand will provide even light and a correctly rectangular image.

But if you don't have a copy stand, can't remove the glass, or your piece is too big for a copy stand, you'll need another solution.

Humans are not particularly adept at seeing small changes in light levels, so making the light even can be tricky. If you have two flood lights, both the same distance away and at the same angle, you'll probably get it pretty close. (This is what a copy stand has.)

I have to ask a question here: How particular are you? If you want the photograph to really accurately represent your piece, you want the light to be really even. If you're using a copy stand or perfect overcast light, you're probably ok, but otherwise you may want to look into a thing called an incident light meter. (Most light meters have a little bubble you can move over the sensor to make it take incident readings.) What this does is allow you to read the amount of light falling on a particular part of your subject. You can move the meter around over the surface (being careful not to change the distance from the surface) and watch the reading. If the light is even, the reading won't change.

This example was made with the camera slightly off to one side to avoid reflection and with two lights, one on either side.

Now that we have our basic digital capture, we need to make some corrections to it. One correction will be the rectangularity, mentioned above, but we'll do that later. The other correction is color. Look at the (supposedly) white mat board in this photo. Notice that it has an orangish cast to it.

Here, again, I have to ask you: how particular are you? Do you care if the colors in your photograph accurately represent your piece? If you don't care, you can skip the whole next section. If you do, read on.

The reason the color is off is because different light sources emit light of different colors. Flourescent is greenish, incandescent is orangish. Sunset is pinkish. Clear sky is bluish. The human eye tends to compensate for the color of the light, so we don't usually notice it, but cameras aren't as good as eyes at this task, so we will correct the color in Photoshop.

Color

We're going to use Adobe Photoshop to correct the color in your image, but before we look at that, I need to touch on a little color theory. I will be brief.

Being an artist, you know that red, blue, and yellow are the primary colors. But, that's only true for paint. (And it's really not entirely true even there, but that's a different discussion.) With light, the primary colors are red, green, and blue. If you were to shine a red light, a green light, and a blue light so that their beams intersected on a white surface, you'd get something like this illustration. The intersection of all three is white light.

Photoshop represents color in three "channels", red, green, and blue. Each channel, by itself, is shades of gray, but when the picture is displayed, the red channel is (metaphorically) shown through a red light, the green channel is shown through a green light, and the blue channel through blue. If a particular spot on the image has the same shade of gray in all three channels, that point will have no color--it will be gray.

The reason I want you to understand this about Photoshop (and light) is that we're going to use this characteristic to correct the color in our piece.

If something is white or gray or black, it has the same amount of red and green and blue. If it has some color tint, it does not have the same amount of these three colors. So, we're going to force our white, black and gray reference points to have equal amounts of red, green, and blue. That will make our photograph have the correct color.

We're going to use one particular adjustment in Photoshop to color-correct our photograph: Levels. With your image loaded into Photoshop, on the Menu Bar, click on Image, Adjustments (or, in older versions, Adjust), Levels.

This will bring up a window like the one in this illustration. The main part of the window shows a histogram, or bar graph, indicating the quantities of values from black to white in the image. (The values of the three channels are combined to produce this.) Black is on the left and white on the right. Notice the large spike on the right side. This shows that there is a lot of near-white in this image.

The white triangle, identified with the red arrow, is at pure white. Notice that in this image there is no pure white. The black spike in the chart drops before reaching the white triangle. The black triangle, marked with a blue arrow, is pure black. If you look carefully, you'll see that there is some pure black in the image.

We will be adjusting our "white point" and "black point" (along with the gray point) in a minute, but first, we have to do a little fine-tuning.

Slightly advanced step: photographs tend to lose detail at the two extremes of white and black. Because of that, it is a good idea to make the lightest thing in your photo not quite white and the darkest thing not quite black. So we're going to tell Photoshop exactly what color to make our white point and our black point.

The tools we'll use to set the white and black points are the eyedropper tools, identified in the illustration above, respectively, with the red and blue lines on the lower right.

Double-click on the white eyedropper. This illustration is a detail from the resulting window. Change the R, G, and B numbers from 255 to 249, as shown. Notice in the large rectangle near the top that the previous and current colors are shown. You have now told Photoshop that what you click on with that white eyedropper you want to have Photoshop adjust so that all three channels have a value of 249, which will make that spot the color you see in the top of that rectangle. Click OK.

Do the same thing with the black eyedropper, setting all three channels to 8.

Now you'll preserve detail in your lightest and darkest areas, but with color accuracy.

This part's pretty cool: You're going to find the the lightest and darkest things in the image and make them white and black (or not-quite, if you did the previous step).

Assuming you're still in the Levels window, first click in the title bar (blue strip) on the top of the Levels window and move the window somewhere so you can still see most of your image. Then click on the white triangle, identified in the above illustration with the red arrow. While holding the mouse button down, click: Mac: Option; PC: Alt. While holding that key and the mouse button down, slowly drag the mouse to the left. The image will turn black and as you drag the triangle, you'll start to see things turn colors then turn white. Find the spot that first turns white. That's the lightest thing in your image. When you've found it, drag the triangle all the way to the right again. You only did this to find that spot. Click on the white eyedropper (red underline in the above illustration), and click on the spot you just identified.

Do the same thing with the black triangle, only dragging it to the right. The image turns white, then black begins to appear. Use the black eyedropper just as you did the white one.

Finally, click the gray eyedropper (between the black and white ones), and click on the gray card. (You did include a gray card, didn't you? If you didn't include a gray card, skip this step.)

Click OK in the Levels window. Your image should now have a full range of values from black to white and the colors should be correct. Congratulations!

Squareness

Now you're going to trim off the excess fat and square up the image.

In the Toolbox, select the Crop tool . Click and drag it around your piece. You do not have to be very accurate here. In the Tool Options Bar (under the Menu Bar), click the check box next to "Perspective".

Zoom in on a corner by pressing the space bar and Mac: Command; PC: Ctrl and drag around the area you want to see. Zoom right in there to see the corner.

Click on the little square in the corner of your dotted line, indicating where the image is to be cropped, and reposition it exactly to the corner of your piece.

Hold the space bar down and click and drag to reposition your view of your image. Go to another corner and reposition that square. Do the same for all four corners.

Press Mac: Return; PC: Enter to finish the crop.

Proportions

If you were able to photograph your piece exactly head-on, not from an angle, you can skip this step.

If you photographed it from an angle, you have distorted the length-to-width ratio, so now we will correct that distortion. You must know the actual dimensions of your piece to do this step, so grab a tape measure and go measure it. I'll wait.

You're back. Good. Now, in the Photoshop menu bar click Image and Image Size. On the bottom of the window is a check box labeled Resample Image. Uncheck that. Then, in the Document Size section's Width box, put the actual width of the piece in.

Click OK.

Now, open the same Image Size window again. This time, check the Resample Image box, and UNcheck the Constrain Proportions box.

In the Document Size section's Height box, put in the actual height of the piece.

Click OK.

Save the file by clicking File and Save As. In the Format box, specify TIFF. Select an appropriate file name and location and click OK.

Congratulations. You've done it!

File Size

How will you be using this digital file? Will it be for displaying on the Web or for print? Will it be distributed on a CD? Will it be incorporated in a presentation of some kind?

There are two fundamental classes of digital images: those intended for display on a computer and those intended to be printed. For print you need a much higher-resolution file.

For comuter display, you only need it to contain as many pixels as the screen on which it will be displayed, no matter how large that screen is. A common resolution for computer screens these days is 1024 pixels across by 768 pixels high. (Note that computer screens are always in horizontal format.) If you're going to be giving a lecture in a giant hall with an enormous projection system connected to a computer, chances are that the resolution of the projector is still no more than 1024 by 768. But, find out!

Ok, here's a test, to find out if you've been paying attention:

We want to save the picture we've been working on for display on a computer screen. Which of the following should we make it? (dimensions are width x height)

a 1024 x 768
b 768 x 1024
c 654 x 768

The answer is c, 654 x 768, because the picture is vertically oriented, so it won't fill the entire horizontal area of the screen, and the maximum height on the screen is 768 pixels. So we make the height fit and the width is what it has to be to maintain the correct proportions. Don't worry: you don't have to do the math.

If you're saving the file for some specific application, find out what the requirements are. If you just want to have a high-quality digital image, stick with the original size.

File Format

How will you be using this image?

You always want a TIFF (or PSD) version of your file, but if you're going to display the image on a computer, also save in the JPEG format, for that application. If you're saving for print, stick with TIFF. These formats are options that you can select when you click File and Save As.

A note about JPEG: This format compresses the data, and you lose some quality in the process. If you may have to make any changes and resave the file, do not use JPEG, or be sure to save a TIFF version as well. Each time you resave in JPEG format, you lose more information. Note: merely copying the file from one location to another does not degrade the file; only opening it and clicking File and Save or Save As.

... and resize the image using the Pixel Dimensions boxes at the top of the window. Remember the limitations of your computer display (discussed above), or specifications for your particular application. (Notice that when you change either the width or height, Photoshop will automatically change the other, because you have told Photoshop to constrain the proportions.)

Then go to File and Save As. Be sure to do Save AS, and change the format to JPEG.

This illustration shows changing from TIFF to JPEG format. The extension on the file name will change from ".tif" to ".jpg" when you do that.

Presentation

You can put your work into a slide show in a program like Microsoft PowerPoint, or you can put it in a gallery on a Web site. Or you can put it in an HTML-based presentation on a CD. Or you can just put the files on a CD (probably in JPEG format, for on-screen viewing).

If you're preparing for a specific purpose, find out what's expected. If you want the most flexibility, just put the files on a CD in JPEG format, in screen resolution, and keep your full-resolution, TIFF files elsewhere.

Again, if you are preparing for a specific purpose, find out what's expected!

End