Skip to content

junyuchen245/Active-Contour-Wavelet-Seg

Repository files navigation

Active Contour Without Edges on Wavelet Features

This is a simple implementation of my paper:

Chen, Junyu, et al. "Feature-Based Image Clustering and Segmentation Using Wavelets." arXiv preprint arXiv:1907.03591, 2019.

This method builds on the Active Contour Without Edges (ACWE) to incorporate Wavelet features, with a tunable parameter 'q' to allow to us control the type of information (low-frequency or high-frequency info.) we want to segment.

Data required to run the MATLAB program can be found here: required data.

Sample results for SPECT bone and bone lesion segmentation:

q = 0.75 (Bone Segmentation):

q = 1.7 (Lesion Segmentation):

Releases

No releases published

Packages

No packages published

Languages