The first sharpening issue we have to confront is the inevitable softening that occurs whenever we turn a continuous stream of photons into discrete square pixels. No matter how good our lenses, no matter how high-resolution our digital cameras or scanners, the digitization process always loses some sharpness.
Further softening occurs on output, when we translate image pixels to printer dots. Each output process has its own requirements for sharpeningwe need to apply very different sharpening to the same pixels printed at 300 ppi to a continuous-tone printer and at 300 ppi to a 150-line screen halftone printer.
Last but not least are the issues presented by the image content itself. A forest full of trees needs a different sharpening treatment from a head shot of a model. Sharpening that emphasizes the fine detail in the trees turns the model's skin into a moonscape, while sharpening that emphasizes the models eyes and lips may actually obscure the detail in the forest. Somehow, we have to reconcile the various and sometimes contradictary needs of the image source, the image content, and the output process.
I'll spend the rest of this chapter looking at these disparate demands in detail, because until they're understood and addressed, sharpening won't do justice to the image, and may actually harm it more than help it.