Wednesday, January 13, 2010

Resolution 101

A very important technical concept to understand when starting any digital imaging project is image resolution.

A pixel is the smallest measurable interval of a digital bitmap image. Resolution is a measurement of pixels. It determines how an image will appear on the screen and how the pixels are distributed in the output, or how the pixel information will be used on a printer. The original pixel dimensions of an image are determined by the capabilities of the hardware that was used to capture the image, usually a digital camera or scanner. More captured pixels in an image means more detail, better gradations and depth of color.

Like many other things in a digital workflow resolution can be thought of in two ways: input and output.

Input is the capturing of pixel information weather through a digital camera, scanner or created/rendered with bitmap image software.

Output is how that captured information is going to be used or distributed. On a monitor, projector, television or output through a printer to paper, canvas, wood, metal or any other substrate.

Even though they are often used interchangeably, pixels per inch (PPI) and dots per inch (DPI) are not the same thing. PPI refers to the number of pixels found in one inch of a digital image or a monitor. DPI refers to the amount of resolution a printer is capable of outputting. Printers use dots of ink to render an image; the more dots a printer can produce per square inch (DPI), the better its quality. The finer the dots and the pattern by which the dots of ink are laidd down also contribute to the quality of the printed image. But this is only half the equation. No matter how great the printer resolution is the quality of output will be determined by the resolution of the original capture. In an ideal situation you want the needed input resolution for the intended output.

Lets say you're working on a large trade show exhibit with multiple images from multiple sources and the main part of your exhibit is a large paneled wall printed about 8 feet by 10 feet. It would be best to set up and edit the images to output size before beginning any layout or design. This will give you an idea of how big you can go and possibly save you from having to rework anything later for lack of resolution. Viewing distance is a key factor in determining how big you can go. It is best to speak to a technician for information on resolution guidelines, but most likely this will be anywhere from 100 DPI to 300 DPI at actual print size for most printers.

To see how much image resolution you have for your intended print size open the image size dialog box in Photoshop under the image menu.



Here is a 10" x 15" image at 180 ppi resolution. Uncheck the resample box and enter a new document size.



This resizes the image and changes the image dimensions without changing the total pixel dimensions. Note how the pixel dimension is locked out and doesn't change. Here at 20" x 30" the resolution is 90 ppi. By doubling the document size the resolution is cut in half. Essentially the original pixels of this image are going to be distributed further apart to cover the new larger size. This also works in reverse when scaling down an image. Making the size smaller gives you more pixels per inch.

If you check the resample image box and change the dimension size the software uses interpolation to create new pixels, in effect adding pixels to the image. The interpolation process estimates the values of the pixels needed based on the existing pixels in the image. Interpolation can introduce a loss of quality even though it is adding pixel data. This depends on the quality of the image that is being resampled and the amount of resampling being done. The trick is to use it sparingly. It only really works if the original pixel data is of good quality, sharp, not grainy and not compressed. Then you have a shot at increasing your pixel dimension 15% to 25%. Another tip would be to size up in increments of 5% to 10% at time rather than all at once. This allows the software to make better estimates on the new pixel data.

Image resolution is a complex subject matter with many facets. Hopefully this will bring some light to the base concept and be of aid in your next large format printing project.

Paolo DeSanctis
Graphics Production

No comments:

Post a Comment