Algorithm that generates accurate superpixel segmentation for noisy images (particularly useful for biological images such as electron microscopic images etc.)
Matlab
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
APD.m
ComputeCannyEdgeWeights.m
ComputeEdgeWeights.m
Generate_Features.m
LICENSE
NLmeansfilter.m
README.md
Region_Merging.m
Salient_Watershed.m
UPD.m
adjustImageContrast.m
intersect_sorted.m
kmeans_color.m
merge_regions.m
partition_dist.m
plotGreenEdges.m
plotRedEdges.m
plotRedGreenEdges.m
post_process.m
reg_distance.m
region_adj_graph.m
remove_block_zeros.m
remove_boundaries.m
remove_small_regions.m
validation.m

README.md

Date: 06/05/2013, v0.1

Salient_Watershed.m -- Performs the salient watershed over-segmentation for noisy images. Region_Merging.m -- Takes the salient watershed output and performs region merging.

Dependencies: 1. BSD (Berkeley segmentation dataset and code utilities). 2. FastEMD (http://www.cs.huji.ac.il/~ofirpele/FastEMD/ 3. Texture (http://www.robots.ox.ac.uk/~vgg/research/texclass/)

Cite: S. Navlakha (), P. Ahammad (), E.W. Myers. Unsupervised Segmentation of Noisy Electron Microscopy Images Using Salient Watersheds and Region Merging. BMC Bioinformatics 2013. (*) = Equal co-authors

Contact: S. Navlakha -- navlakha@cs.cmu.edu P. Ahammad -- parvez@ieee.org