Java optimization package development

Sahar Hanna

Master Semester Project

Management of Technology and Entrepreneurship, EPFL

July 2016

Abstract

Pattern recognition continues to grow into one of the most important areas of image analysis and data processing. Networks of polygonal cells are in fact commonly encountered in nature. Their appearance in various and diverse biological entities, and the necessity of quantifying and understanding their properties make it important to develop corresponding junction detectors. For this semester project, I base my work upon the paper published on this topic by Puspoki et al, explore the results achieved for symmetric junctions, as well as the possible generalization anticipated for asymmetric junctions. The aim of this project is to develop the existing framework to detect all types of junctions, by detecting their locations, scale, orientation as well as the angles defining the junction.