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

Figure 8

Number of subanimations for each cluster associated with the traffic control video sequence under a SOM with 5 neurons.