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
TBES
version 1.0
WEB
DOC
f1: 0
f2: 0
f3: 0
f4: 0
Shankar Rao, Hossein Mobahi, Allen Yang, Shankar Sastry and Yi Ma:   Natural Image Segmentation with Adaptive Texture and Boundary Encoding

Algorithm for unsupervised segmentation of natural images that harnesses the principle of minimum description length (MDL). The method is based on observations that a homogeneously textured region of a natural image can be well modeled by a Gaussian distribution and the region boundary can be effectively coded by an adaptive chain code. The optimal segmentation of an image is the one that gives the shortest coding length for encoding all textures and boundaries in the image, and is obtained via an agglomerative clustering process applied to a hierarchy of decreasing window sizes. The optimal segmentation also provides an accurate estimate of the overall coding length and hence the true entropy of the image.

List of uploaded results for 'TBES' 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] TBES 1.0 37.72 59.77 1.25 7.24 8.22 27.55 95.08 63.30 65.66 96.20 34.34 0.60 74.19 63.73 5.27 77.29 7.57 6.02 44.81 43.26 65.66 56.32 63.34 19.40 75.45 8.76 18.00 75.24 10.93 60.46 51.10 66.22 69.21 91.55 53.39 8.10
#Colour [large] TBES 1.0 43.43 56.34 2.19 5.38 6.45 27.28 80.06 65.64 68.15 95.50 31.85 0.88 75.64 66.02 4.53 78.45 7.51 6.01 41.75 40.25 68.15 59.33 65.93 18.00 76.18 8.36 16.92 76.67 9.82 64.42 55.54 69.52 72.19 91.42 58.90 7.12