The resolution of a digital camera has a direct link with picture quality. The numbers of colors (bit depth) saved in the file, or captured by the camera, helps determine the overall quality of the image.
The higher the resolution of image, the greater the bit depth and the better is the image. High-resolution images with lots of colors are generally better quality than those with a limited color range and fewer pixels, but to understand how integral resolution is to making great digital images we must look a little deeper.
All digital photographs are constructed of a grid of colored pixels which when seen at a distance combine to form the appearance of a continuous color and tone picture.
Computers can only work with digital files. The world as we view it, and as we capture it in conventional camera photographs, is not in a digital format. Tones and colors merge gradually from one extreme to another. For the computer to be able to work with such images they must be changed from this 'analog' or continuous tone format to a digital one. Scanners and digital cameras make this change as part of the capturing process.
Resolution at this capturing stage refers to the frequency that samples are made of the image. Generally this measurement is represented as the number of samples taken in a one inch space; for this reason it is sometimes called Samples Per Inch or spi. Unfortunately most scanner software does not use this terminology but prefers to refer to this setting as Dots per inch (dpi). This is a hangover from language used in the printing industry and does more to confuse than clarify the situation.
If you are using a digital camera to capture an image then the resolution will be determined by the sensor of the camera, which has a specific number of CCDs set into a grid that is used to digitize the image. Scanner users, on the other hand, are able to control the sampling rate by changing the settings 'in the scanners dialog box'.
A high sampling rate will result in a higher quality image with a much greater file size. A low SPI will provide a smaller file of less quality. These facts lead a lot of new users to a situation where all images are scanned at the highest resolution possible. Do this and your hard drive will soon be completely used up. High-resolution scans require huge amounts of storage space.
Input resolutions should be decided on the basis of what the images final use will be. If the image is to be printed the size of a postage stamp then there is no point scanning at a resolution that will result in a file large enough to print an A2 poster.
How many pixels are sufficient?
Suitable for Elements – 6, 5.0, 4.0, 3.0, 2.0, 1.0 | Difficulty level – Intermediate
Related techniques – 2.02, 2.12, 2.13, 13.07
The pixels required is related to the end product you want to create. As an example, if you want to produce a 10 x 8 inch photographic quality print and you know that the lab you will use to output the image suggests a resolution of 250 dpi, then you have all the information to determine the number of pixels you will need to capture. Essentially the lab is saying that to produce photographic quality they need 250 pixels for every inch of the print. For the photograph to be 10 inches high then your file must contain a minimum of 2500 pixels for this dimension and to ensure the 8 inch width, you will need 2000 pixels. With this knowledge you can adjust the settings on you scanner so that you will end up with a picture file that contains the minimum pixel dimensions of 2500 x 2000.
For digital camera shooters understanding this concept will not only give you an indication of the maximum print size available from your camera's sensor, but will also allow you to accurately select the correct resolution, or more precisely the correct pixel dimension, setting on your camera for specific tasks. The table below will give you a good starting point.
| Chip pixel dimensions: | Chip resolution: (1 million = 1 megapixel) | Print size at 200 dpi: (e.g. photo print) | Image size at 72 dpi: (e.g. web use) | ||
| 640 x 480 pixels | 0.30 million | 3.2 x 2.4 inches | 8.8 x 6.6 inches | ||
| 1440 x 960 pixels | 1.38 million | 7.4 x 4.8 inches | 20.0 x 13.2 inches | ||
| 1600 x 1200 pixels | 1.90 million | 8.0 x 6.0 inches | 22.0 x 16.0 inches | ||
| 2048 x 1536 pixels | 3.15 million | 10.2 x .7.6 inches | 28.4 x 21.3 inches | ||
| 2304 x 1536 pixels | 3.40 million | 11.5 x 7.5 inches | 32.0 x 21.3 inches | ||
| 2560 x 1920 pixels | 4.92 million | 12.8 x 9.6 inches | 35.5 x 26.6 inches | ||
| 2000 x 3000 pixels | 6.0 million | 10 x 15 inches | 27.7 x 41.6 inches | ||
| 2336 x 3504 pixels | 8.2 million | 11.6 x 17.5 inches | 32.4 x 48.6 inches | ||
| 2592 x 3888 pixels | 10.2 million | 12.9 x 19.44 inches | 36 x 54 inches | ||
| Proposed use | Suggested image resolution | ||||
| Screen or web use | 72 dpi | ||||
| Draft quality inkjet | 150 dpi | ||||
| Photo-quality inkjet | 200-300 dpi | ||||
| Photo lab output (min) | 250 dpi | ||||
| Photo lab output (max) | 400 dpi | ||||
| Offset printing (good quality) | 300 dpi | ||||
The suggested output resolution changes for different end uses. For you to accurately capture enough pixels for the end result you desire, you will need to be aware of the resolution requirements for different end products. The table to the left indicates some of the resolution requirements for different uses.
Once you know the image resolution needed for the printer, the final size of the print and the size of your original you can easily calculate the scanning resolution and the total pixel dimensions you need for your digital file. Use the formula in the table below to give yourself an indication of the number of pixels you need for any size print job.
| 1. | Final image dimensions (pixels) | = | Original image dimensions (inches) | x | Scanning resolution (samples per inch) |
|
| 3000 x 2400 pixels | = | 10 x 8 inch print | x | 300 samples per inch |
|
| 6000 x 4000 pixels | = | 1.5 x 1 inch (135 mm film) | x | 4000 samples per inch |
| 2. | Print size (inches) | = | Image dimensions (pixels) | / | Image resolution (pixels per inch) |
|
| 15 x 12 inches | = | 3000 x 2400 pixels | / | 200 pixels per inch |
|
| 20 x 13.33 inches | = | 6000 x 4000 pixels | / | 300 pixels per inch |
