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

Algorithm 6

Generating an EVM-animation in the D-EVM from a frame sequence.

   Input: The path of the frames and time positions for initial and final frame.

(1)  Procedure generateAnimation  (String workingDirectory, integer initFrame, integer endFrame)

(2)      EVM currentFrame, prevFrame, diffFrame;

(3)      string framePath;

(4)      prevFrame ← initEVM ( );

(5)      currentFrame ← initEVM ( );

(6)      for  i = initFrame  to  endFrame  do

(7)        currentFrame ← initEVM ( );

(8)        framePath ← workingDirectory + “frame” + toString () + “.image”;

(9)        currentFrame ← loadImage (framePath);

(10)      diffFrame ← getCouplet (prevFrame, currentFrame);

(11)      saveEVM (diffFrame, );

(12)      prevFrame ← currentFrame;

(13)      saveEVM (prevFrame, endFrame + 1);