A novel local intensity order transformation for robust curvilinear object segmentation(TIP)
-
Updated
Oct 12, 2022 - Python
A novel local intensity order transformation for robust curvilinear object segmentation(TIP)
This repository contains the implementation for our work "Topology-Aware Uncertainty for Image Segmentation", accepted to NeurIPS 2023.
In this repository, two types of robust multi-scale anisotropy features, namely PCmPA and PCmFFA, that considering the local contextual information of curvilinear structures are extracted from phase congruency maps for the effective discrimination of isotropic structures and curvilinear structures.
This code is an implementation of the Step filtering and the Polynomial filtering method proposed in [1] providing an enhancement on partial curvilinear structures.
Bayesian Wavelet Topology Optimization for Curvilinear Pattern Recognition
To associate your repository with the curvilinear-structures topic, visit your repo's landing page and select "manage topics."