Grayscale Image Colorization using Seeded Cellular Automaton

John Paul T. Yusiong, Nherie Lynn U. Celeste

Abstract


Image Colorization, or simply colorization, refers to the computer-assisted process of assigning colors to pixels of a grayscale image in order to convert monochromatic images into visually plausible colored images. Furthermore, adding color to a grayscale image is a very important technique to improve its visual quality. While the traditional method of colorization dealt with segmenting the images into different regions before flooding the segmented areas with colors, another approach in image colorization usually involves drawing colored scribbles on grayscale images. These scribbles will then act as markers or “seeds” that determine the color of a pixel in the image. Nevertheless, colorization is an expensive, tedious and time-consuming process even with current technologies. Thus, several researchers presented different approaches to solve these problems. This research work proposes a scribble-based method of image colorization with the use of the Seeded Cellular Automaton. Experiment results revealed that the Seeded Cellular Automaton successfully produced colored images from grayscale images with acceptable and satisfactory visual qualities.

 

Keywords: Colorization, Image Segmentation, Ford-Bellman Algorithm, Cellular Automaton, Seeded Cellular Automaton


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v6i1.2364

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science