Image Preprocessing for Artistic Robotic Painting

Artistic robotic painting implies creating a picture on canvas according to brushstroke map preliminarily computed from source image. To make the look closer human artwork, image should be preprocessed render effects usually created by artists. In this paper, we consider three preprocessing effects: aerial perspective, gamut compression and coherence. We propose an algorithm for perspective amplification based principles of light scattering using depth map, nonlinear hue transformation gradient filtering obtaining well-coherent with reduced number brushstrokes, required practical painting. The described algorithms allow interactive correction final rendering manually painted artwork. illustrate our proposals, several test images computer paint monochromatic robot.