The first step is to decide what kind of optimization the image requires for both source and content. Source optimization for a specific digital camera is essentially constant, though at high ISO or on severely underexposed shots some noise reduction may be needed first. Source optimization for film scans depends on the scanner, the resolution at which the film is scanned, and the film format itself. For digital captures, the goal of source optimization is to counteract the effect of the antialiasing filter, while with film scans, the focus may be more on noise reduction, especially with 35mm format. Tip: Scanning to Reduce Noise Some scanner drivers let you scan the film multiple times to reduce scanner noise. You can often get a much cleaner scan using four passes than you can using one. Some scanner drivers let you scan as many as 16 passes, which in my view is overkill. This technique suppresses scanner noise, but doesn't address film grain. A technique that does address film grain is to scan at a higher resolution than is needed, then downsamplethis mitigates both scanner noise and film grain. Next, you need to decide whether you want to treat the image as low-frequency, high-frequency, or something in between. Most images have both low-frequency and high-frequency components, but there's usually a dominant tendency you wish to emphasizesee "Optimizing for Image Content" in Chapter 3, Sharpening Strategies. With source and content in mind, let's look at some real examples of optimizing for source and content, and the techniques used to do so. |