Image Resolution

"Resolution: The act, operation, or process of resolving. Specifically: (a) The act of separating a compound into its elements or component parts."

What is image resolution?

Resolution is a measurement of density: how closely together pixels are displayed on a screen, or how many dots of ink are enclosed within a particular area of paper. For example, 300 DPI means there are 300 dots per inch (90,000 dots in a square inch).

Often, in discussions with publishers or printing suppliers, higher resolution may be taken to mean higher quality — which is likely to cause confusion and misunderstandings when preparing images for print. The origins of this issue stem from expressing exact dimensions for the final output (in which case the dot pitch will be a determining factor in quality). The quality of the original source image is a separate matter, and it is important to measure it independently.

What is image quality?

Quality comes from detail and accuracy in a photo (and from the content of the image - the skill of the photographer). In a digital image, each pixel of information defines part of the image. Modern professional equipment captures millions of pixels of information when a photograph is taken, so the measure used most commonly is "megapixels". As each pixel carries some detail, it is the sum of all this information which leads to the perception of quality by a viewer.

Separating resolution from quality

There is no link between the quality of an image and DPI. Quality is better measured by the total number of dots (pixels), not by how closely together they are printed. An image can be resampled to any value of DPI — any resolution — without changing its quality.

A 300 dpi image can readily be printed to half the physical size in both height and width, by printing at 600 dpi, since the dots are half the distance apart. That might affect the quality of the print but why would it affect the quality of the source image? It doesn't: that would be confusing the output quality and the input quality.

• Dots-per-inch is not a measure of quality - it is a measure of density
• The number of pixels (megapixels) is a good measure of quality.

Scanning Images

Having established the difference between resolution and quality, it is possible to consider scanning - a process where these two definitions are inter-connected. Very high levels of interpolation and poor optics are often in evidence in inexpensive scanners, usually with the purpose of marketing a scanner with a higher available DPI value. Remembering that quality and resolution are not the same, we can consider two examples:

Bad practice: take a poorly scanned image from a 35mm camera. This may come from a $200 scanner at nominally 9600 dpi. The scanner may have poor contrast and very basic colour calibration. The file can then be 'interpolated' to a much larger file size in PhotoShop, and saved as an uncompressed 16 bit TIFF file (perhaps 100MB). In this way, a poor quality image has been created, which far exceeds the underlying size and quality of the photograph. Sadly, this is all too common.

Good practice: using a high quality commercial scanner, extract a sharp scan of the 35mm image. A professional scanner will extract a full range of shadow and highlight details, with minimal colour bleeding and maximum contrast — as well as accurate, calibrated colours. Even though it may be a 4800 dpi (or less) scan, the information is closer to containing the information content of the source photography. After this point, make any adjustments and touch up the image. The image should be saved, either as a compressed TIFF or a JPEG at maximum quality, and not interpolated.

The file from the second example is very likely to be much smaller than the file from cheap scanner. It is, nonetheless, of better quality.

Preparing images for commercial use

Making adjustments to an image after scanning can improve the way it looks. Examples include changing the contrast and levels, removing spots of dust from scans or from digital photographs taken on bright days, and the 'unsharp' filter which improves the perceived sharpness of an image, especially on the web.

Some of these processes are genuinely helpful to commercial users (eg. removing dust, balancing levels), but some are better performed at the end of the page setting process. In particular, it is not sensible to sharpen an image heavily, since the perception of sharpness depends on the output media (print, screen, etc).

One way to look at post-production editing of this sort is to understand that very few filters add any information - many actually remove it. Taking care to provide a pristine image is important, but not if the filter reduces the quality of the output and cannot be reversed.

Save Money: Compress Your TIFF files

Often, high-resolution image data is exchanged using the TIFF (Tagged Image File Format). This file format is well-established, having been used in the industry for many years.

There is a worrying trend in image libraries to use uncompressed TIFF files, simply because the larger file size somehow implies higher quality - this logic is unfounded and is genuinely misleading image buyers. The larger file size simply implies inefficiency and waste, both in disk storage and bandwidth.

The main reason for using TIFF is that it is 'lossless', which is to say that there is no degradation of quality when an image is saved or repeatedly edited. They can be compressed for no loss of quality by using LZW or ZIP options available in most graphics packages when the files are being saved. This sort of compression usually saves 70% on storage and - not to be confused with JPEG - there is nothing lost.

Other file formats

TIFF is not always an efficient format to use. By using smarter compression algorithms, very significant reductions in file size are possible. JPEG compression is almost always indistinguishable from TIFF when the highest quality setting is used (ie. lowest compression). Anyone who plans to store images electronically should consider using JPEG instead of TIFF, provided that the JPEG file format is only used as the final stage of the workflow (thus avoiding compound losses).

More recently, JPEG2000 is being used as an improved format which is able to produce images that are simultaneously smaller and higher quality than JPEG. JPEG2000 can also provide lossless compression, like TIFF, but with smaller file sizes, even compared to LZW compressed TIFF images. Unfortunately, manipulating JPEG2000 files requires powerful computers, and saving files as JPEG2000 data is very slow. Despite this, as an emerging standard, JPEG2000 is undoubtedly the future of image storage.

IMS supports TIFF, JPEG and JPEG2000 as well as other formats. We recommend using JPEG on the highest quality setting, or LZW compressed TIFF images (never uncompressed TIFF files).