P-median and P-center applied to Color Quantization and Pixel Mapping

The use of digital images has becoming increasingly popular, mainly due to the drop in media equipament's prices, such as, cameras, scanners, video cameras, computers, among others. However, even for the current storage patterns [3,4,5], the digital images consume a large quantity of the resources. In this context, this paper presents a single and intuitive manner to reduce the size of digital images. Instead of sophisticated mathematic theories, such as Fourier transformation, Z theory, among others, our purpose is to reduce the number of colors, treating the problem through facility location optimization model approach, more specifically using p-median and p-center models [1]. The results, obtained by multi-spectrum images indexing like the GIF approach, is promising, in spite of its incipient phase.