Investigation of basic imaging properties in digital radiography. 2. Noise Wiener spectrum.

We investigated theoretically the effects of various digital parameters, such as sampling aperture, sampling distance, number of quantization levels, and display aperture, on the noise Wiener spectrum of digital radiographic imaging systems. We also measured Wiener spectra for our digital image simulation/processing system, and the results agreed well with the theoretical predictions. Aliasing, which is an artifact caused by undersampling, and the use of a limited number of quantization levels were found to increase the Wiener spectrum for digital systems. The effects of image processing, including unsharp mask filtering, integration, and subtraction, on the Wiener spectrum were also demonstrated. Since noise influences the detectability of radiologic objects and thus diagnostic accuracy, knowledge of the effects of the various digital parameters on the noise spectrum will be useful in the evaluation and design of future digital imaging systems.