Chuck Gardner's Photography Class
Part Three - Digital Photography Basics
by Chuck Gardner
Increasing sharpness - Unsharp Masking

As noted above digital image data is recorded using three separate RGB receptors then combined into a single data point denoted as a pixel. The image is broken into a chessboard pattern of pixels. The net result of this slicing and dicing of image is a lack of sharpness. Because the image data is digital it is possible to mathematically manipulate it to increase sharpness. This is done by evaluating adjacent pixels and increasing the brightness of those which are adjacent to dark areas. This process is known as unsharp masking, a term which is a carryover from conventional color separation. A more accurate description in the digital realm would be "edge enhancement effect". If done discretely unsharp masking will increase the apparent sharpness of the image. If overdone there will be a noticeable light "halo" around the dark areas of the photo.

All consumer digital camera cameras apply some degree of unsharp masking or "edge enhancement" to the files. Some of them allow the user to control the amount of sharpness which is applied in the camera, which is a desirable feature since the amount of unsharp masking needed varies, depending on whether the file is displayed on screen at 72 or 96 pixels per inch (ppi) resolution on a web page, printed on a ink jet printer at 180 ppi, or used for offset printing at 270 - 300 ppi.

Other factors affecting image quality

Camera optics: Good digital imaging requires the CCD sensor to be illuminated evenly, in focus, edge-to-edge. Unless the image reaching the CCD is free from optical distortion of vertical and horizontal lines, and chromatic aberration (i.e., different wavelengths of light focusing differently, resulting in color halos) the image quality will be poor, regardless of how many megapixels the CCD array has. Not all camera lenses are created equal, and the most expensive one is not always the best. Ultimately it is the combination lens system, CCD, and in-camera image processing which is the true measure of which camera produces the "best" images. Links to sources for on-line comparison of cameras, including standard test charts for color fidelity and image resolution, are listed below.

Electronic "noise" and artifacts: Variations in the sensitivity of the individual sensing cells in digital camera CCD can result in apparent differences in even areas of tone in a photograph, especially in dark areas. The ability of the camera to detect and correct for these "artifacts" varies with camera brand and model. Design and usage of the camera can also influence the amount of noise and artifacts in the image. There is a known relationship between the temperature of the CCD chip and the amount of noise created. The amount of noise doubles with each 6 degree C rise in temperature. A camera design which places a heat generating LCD display in close proximity to the CCD sensor will be more prone to heat generated noise than one which has a LCD which can be moved away from the camera or turned off.

Goto Next Page >

Goto < Previous Page

Goto Class Outline

Goto super.nova.org my home page.