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

Algorithm 2

Computing the Regularized Boolean Operations for two nD-EVMs.

   Input: Two D-EVMs and , the Boolean Operation op, and the dimensionality .

   Output: A resulting D-EVM.

(1)  Procedure booleanOperation  (D-EVM , D-EVM , booleanOperator op, integer )

(2)      EVM pSection, qSection;

(3)      EVM couplet;

(4)      boolean , ;

(5)      real coord;

(6)      EVM result, rPrevSection, rCurrSection;

(7)      if  then

(8)        return booleanOperation1D ();

(9)       else

(10)        pSection ← initEVM ( );

(11)        qSection ← initEVM ( );

(12)        rCurrSection ← initEVM ( );

(13)        nextObject ();

(14)        while  !(endEVM (p))  and  !(endEVM ())  do

(15)        if  fromP  then

(16)          couplet ← readCouplet ();

(17)          pSection ← getSection ();

(18)        if  fromQ  then

(19)          couplet ← readCouplet ();

(20)          qSection ← getSection ();

(21)        rPrevSection ← rCurrSection

(22)        rCurrSection ← booleanOperation ();

(23)        couplet ← getCouplet (rPrevSection, rCurrSection);

(24)        setCoord (couplet, coord);

(25)        putCouplet (couplet, result);

(26)        nextObject ();

(27)        while  !endEVM ()  do

(28)        couplet ← readCouplet ();

(29)        putBool (couplet, result, op);

(30)        while  !endEVM ()  do

(31)        couplet ← readCouplet ();

(32)        putBool (couplet, result, op);

(33)     return result