The Prague Texture Segmentation Datagenerator and Benchmark - Algorithms
optimized
for

The concise description of algorithms contains hyperlinks to further information (author, algorithms details, BIB entry, WWW external page).
The algorithm features are: f1 = classification (supervised segmentation), f2 = hiearchy result (manual selection), f3 = known number of regions, f4 = [reserved].
test's
EGBIS
BIB
WEB
SRC
DOC
f1: 0
f2: 0
f3: 0
f4: 0
Pedro F. Felzenszwalb and Daniel P. Huttenlocher:   Efficient Graph-Based Image Segmentation

We define a predicate for measuring the evidence for a boundary between two regions using a graph-based representation of the image. We then develop an efficient segmentation algorithm based on this predicate, and show that although this algorithm makes greedy decisions it produces segmentations that satisfy global properties. We apply the algorithm to image segmentation using two different kinds of local neighborhoods in constructing the graph, and illustrate the results with both real and synthetic images. The algorithm runs in time nearly linear in the number of graph edges and is also fast in practice. An important characteristic of the method is its ability to preserve detail in low-variability image regions while ignoring detail in high-variability regions.

List of uploaded results for 'EGBIS' algorithm
   benchmark label version CS OS US ME NE O C CA CO CC I II EA MS RM CI GCE LCE BCE GBCE BGM SC SSC VD L AVI NVI NMI M ARI JC DC FMI WI WII NBDE
Colour [normal] EGBIS 28.78 19.69 39.15 20.42 21.54 44.35 82.87 51.10 64.12 72.73 35.88 7.59 59.88 49.03 8.38 63.11 16.64 8.97 46.67 39.00 64.12 58.75 56.28 21.29 62.16 8.58 21.24 66.50 19.72 49.98 46.39 62.19 64.19 55.71 76.97 13.05
BTF wood [exp.] [normal] EGBIS 45.41 34.19 45.90 1.13 2.81 35.79 96.43 57.47 68.06 78.42 31.94 8.22 63.05 54.77 6.83 66.14 8.46 2.86 39.06 33.46 68.06 64.88 63.41 16.81 72.91 6.13 13.95 72.82 20.33 57.05 56.19 68.06 71.09 61.15 88.12 10.43