The Prague Texture Segmentation Datagenerator and Benchmark

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
SWA
version def_par

f1: 0
f2: 1
f3: 0
f4: 0

E. Sharon & M. Galun & D. Sharon & R. Basri & A. Brandt: SWA algorithm

SWA algorithm segmentation by weighted aggregation, is derived from algebraic multigrid solvers for physical systems, and consists of fine-to-coarse pixel aggregation. Aggregates of various sizes, which may or may not overlap, are revealed as salient, without predetermining their number or scale.

List of uploaded results for 'SWA' 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]	SWA def_par	27.06	50.21	4.53	25.76	27.50	33.01	85.19	54.84	60.67	88.17	39.33	2.11	66.94	53.71	6.11	70.32	17.27	11.50	50.96	45.18	60.67	51.56	55.84	24.20	69.05	9.75	21.56	70.86	13.68	51.67	43.43	58.77	61.69	79.03	51.05	10.34
Grayscale [normal]	SWA def_par	13.43	38.22	5.05	40.06	41.97	42.12	77.01	47.75	55.64	83.86	44.36	2.85	61.18	46.73	7.25	65.00	23.45	16.88	57.63	51.05	55.64	45.68	49.07	28.93	63.03	11.64	27.13	64.46	16.26	43.91	36.68	52.61	55.17	70.51	45.59	11.73
BTF wood [exp.] [normal]	SWA	44.87	19.97	26.60	8.76	9.15	12.79	30.30	68.01	75.61	80.28	24.39	3.07	75.08	66.63	5.76	76.46	9.51	3.52	31.37	25.39	75.61	71.25	72.04	13.82	78.20	4.28	13.16	84.27	8.79	71.44	63.23	76.66	77.76	77.95	79.88	6.26