Experimental Techniques and Methods for the Automatic Analysis of Image Colors and their Applications to Web Design

I. R. Akhmadeeva¹, Programmer, e-mail: ah.irishka@gmail.com, Yu. A. Zagorulko^{1, 2} Head of Laboratory, e-mail: zagor@iis.nsk.su, A. S. Sery^{1, 2}, Junior Researcher, e-mail: Alexey.Seryj@iis.nsk.su, V. K. Shestakov^{1, 2}, Junior Researcher, e-mail: shestakov@iis.nsk.su
¹ Novosibirsk State University,
² A. P. Ershov Institute of Informatics Systems Russian Academy of Sciences Siberian Branch, Novosibirsk

When designing a website, choosing appropriate decoration colors may not be an easy task, especially if the designer is not professional but an ordinary internet user. This paper proposes methods for automating some tasks involving working with images and color schemes. Such methods may prove useful for both professional designers and beginners. Some of these methods, based on various color quantization algorithms, help to automatically pick up colors that blend well with each other and with the images the designer wants to place on his/her website. The other method is intended for searching for appropriate images within a specified collection on a given color scheme. It is based on an analysis of the image histogram and its result depends on several parameters, which were optimized with genetic algorithm. The paper also presents a brief overview of various tools designed for constructing color palettes on a basis of images and searching for images on color palettes.

Keywords: color quantization, image color analysis, image search, web design

pp. 19–27