This repository contains the Matlab implementation of the metrics described in [1]. This contour score has bean used to compare contours with each other, using a truncated distance-based approach.
[score, scoreStructure] = ContourScore(A, B, dmax, DT_A, DT_B)
This is the main function that returns score, the contour score between A and B, and a strcture scoreStructure which separate every part of the score. A and B should be binary images. You can use convertPolyLineIntoBinaryImage to convert a polyline into a binary image.
The demo.m file illustrates an example use with vizualization.
If you find the code useful, please cite the following paper:
[1]
@article{franccois2020detecting,
title={Detecting the occluding contours of the uterus to automatise augmented laparoscopy: score, loss, dataset, evaluation and user study},
author={Fran{\c{c}}ois, Tom and Calvet, Lilian and Zadeh, Sabrina Madad and Saboul, Damien and Gasparini, Simone and Samarakoon, Prasad and Bourdel, Nicolas and Bartoli, Adrien},
journal={International journal of computer assisted radiology and surgery},
volume={15},
number={7},
pages={1177--1186},
year={2020},
publisher={Springer}
}