Steady-state image processing

This paper presents a new approach to the application mode of image processing operators, the so-called steady-state image processing. The approach reminds a steady-state genetic processing of images by considering each pixel of the image as an individual. So, some pixels are selected, processed and copied back into the image. This differs from the standard approach, where all image pixels are processed at once. The proposed approach offers many choices for variation, and allows for the assignment of dynamic measures to images. This will serve new families of soft computing methods as, e.g. immune-based algorithms, which need images as non-static objects in order to fulfill reasonable tasks. This paper also introduces some basic steady-state operators and exemplifies the analysis of an image by means of a small example. Also, it is shown how steady-state image processing can be applied in the context of texture segmentation. Steady-state image processing can be considered a way of processing images, which is deeply inspired by genetic algorithms.