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

Algorithm 7

Algorithm for converting a frame to a EVM object.

   Input: The frame file path.

   Output: An EVM that is an extrusion for the input frame.

(1) Procedure loadImage  (string imagePath)

(2)       EVM // The resulting EVM.

(3)       ImageData imageData;

(4)       integer colorCount;

(5)       integer array hyperPrismBase, hyperPrismLengths;

(6)        ← InitEVM ( );

(7)       imageData ← readImageData (imagePath);

(8)       colorCount ← imageData.colors;

       / Initialize the hyper-prism base and lengths /

(9)       for    to  (colorCount − 1)  do

(10)         hyperPrismBase ;

(11)       hyperPrismLengths 0;

(12)       hyperPrismLengths ;

(13)       for    to  (imageData.height − 1)  do

(14)         hyperPrismBase ;

(15)         for    to  (imageData.width − 1)  do

(16)           hyperPrismBase ;

             / Get every color component for the current pixel. /

(17)         for to (colorCount − 1)  do

(18)             hyperPrismLengths  [] ← imageData.pixelData ().color ();

(19)           populateHyperPrism (, hyperPrismBase, 2 + colorCount, 0, hyperPrismLengths);

(20)       return