ABSTRACT
The concept of image restoration differs substantially from the idea of image enhancement. The rich body of optimization theory and iterative algorithms combined with the signal-theoretical concepts thus provide a powerful tool that often enables the recovering of useful information even from heavily distorted images. The measured data may be considered heavily distorted observed images, on the basis of which the proper image slice or three-dimensional image data are obtained by inverting the transform that has yielded the data. The measured intensity at an individual optical-image pixel depends, besides on a measured parameter such as reflectivity, on the illumination at the pixel position and on the sensitivity of the particular sensor that measures the reflected or transmitted light at that position. Image noise, as a random two-dimensional signal generated by a stochastic field, has to be described by its statistical properties; its concrete values are naturally unknown in restoration. Most advanced methods of signal and image restoration are formulated as optimization problems.
