Renderização não fotorrealística: uma abordagem para abstração de imagens coloridas digitais.

Abstract

Non-photorealistic rendering is common to both Computer Graphics and Digital Image Processing, which encompasses a broad class of techniques for creating artistic images from scenes, geometric models or photographic images. In this context, digital image abstraction constitutes a paradigm that aims to transform complex photographic images into simplified artistic images in order to facilitate visual communication, while preserving its main content. In this dissertation it is presented a framework designed to generate cartoon-like abstracted images from algorithms of region smoothing and segmentation, edge detection, adjustment of brightness/saturation and color harmonization, which were investigated throughout this research. In order to validate these results, an experimental study involving human evaluators was conducted, in which it was compared the proposed approach with five other existing approaches for image cartoonization. Statistical analysis performed on experimental data from one-way ANOVA and Tukey-Kramer tests revealed significant differences in preference between results created by the proposed approach and those created by the other approaches considered in this research.