Inverse halftoning by means of self-organizing maps

Abstract

Halftoning is the process used to convert a grayscale image into another binary image such that the binary image appears to be similar to grayscale when observed from a certain distance. This process is useful for many printers which are binary in nature, once it allows the printer to deposit the ink as series of dots of constant darkness to print grayscale images. Inverse Halftoning is the reconstruction of grayscale image from its halftoned version. This process can be used in several applications when some image processing operation requires the original grayscale image and only its binary version is available. In this paper we present a method for inverse halftoning using Self-Organizing Maps that is able to reconstruct grayscale images from their halftoned versions generated by dispersed-dot ordered dithering and error diffusion algorithms. Obtained results demonstrate that the proposed method is a good alternative for the investigated purpose. © 2013 Springer-Verlag.