An image-forming system (such as a digital camera) is never perfect, and will introduce artifacts (for example, noise, shading, blurring, aberrations) into any of its images.
The aim of filtering is to improve the image, in an attempt to retrieve the ideal (uncorrupted) image.
An image is enhanced when it is modified so that the information it contains is more clearly evident, more visually appealing or is better suited for further pattern recognition algorithms.
There are three types of operations on images:
Point Operations: A pixel value in the output image depends on a single pixel value in the input image.
Sample: Pseudocoloring a black-and-white image, by assigning arbitrary colors to the gray levels.
Local Operations: Several neighboring pixels in the input image determine the value of an output image pixel.
Samples: Average-, Sigma-, Fog-, Gauss-, Median-Filter
Global Operations: All of the input image pixels contribute to an output image pixel value.
Sample 1: Contrast enhancement by histogram stretching.
Sample 2: Discrete Cosine Transform Filter