Automatic Representation and Segmentation of Video Sequences via a Novel Framework Based on the D-EVM and Kohonen Networks

Table 6

Size of the 3D-EVMs generated for every cluster associated with the traffic control video sequence under a SOM with 15 neurons and removal of the temporal component.

Class	Members	EVM size
0	9,111	8,310
1	5,063	4,658
2	147,723	61,178
3	2,839	1,324
4	83,716	30,748
5	3,131	2,930
6	249,764	62,728
7	143,663	55,420
8	141,288	36,546
9	139,151	18,918
10	2,404	2,192
11	100,026	45,394
12	250,749	50,160
13	924	400
14	25,528	15,888
Total	1,305,080	396,794