Research project supported by FAPESP during October 2020 - September 2021.
Developer: André Luís Mendes Fakhoury
Coordinator: João do Espirito Santo Batista Neto
The main goal of this research project is to extract robust features in R2 to reconstruct curves obtained from images. It aims to analyse image preprocessing algorithms, contour extraction of objects, analysis of important points from curves and the respective reconstruction of the original curve.
The development steps of the project can be seen in the following diagram:
The preprocessing aims to eliminate noise on the contour, allowing the extraction of a curvature that better fits it. The identification of important points is done calculating the discrete curvature of the contour. The curve reconstruction is based on a method describe by Sorkine (2006) using a few points (called anchor) and connectivity information, by a discretization of the Laplace-Beltrami operator. The validation is done by comparing the reconstructed curve and the original curve, using the Euclidian distance.
There is a python .ipynb notebook containing part of the code developed. There is also a latex folder containing some reports and presentations (mostly in Portuguese). The img directory contains some images extracted from ImageCLEF 2011 database.