Procedure DTW_Most_Matching_Learning (); |
Initialize the values of thresholds and to be constants; |
/*DTW recognition process*/ |
Perform DTW template matching; |
Record DTW-distances of recognition results of Top-1 to Top-10; |
/* Decide if starting learning by the unsupervised method */ |
If (DTW score of Top-1 < Threshold ) then |
/* Start learning */ |
For each result label of Top- ( to 10) |
Search the same label as Top-1 among Top 2–10; |
End For |
If (Numbers of the same labels as that of Top-1 > ) then |
Convert the label of Top-1 to the learning index ; |
If ( setting == TRUE) then |
For each |
/* is the number of the DTW-distance with */ |
Search the worst (the largest) DTW-distance; |
End For |
Return the template item with the worst DTW-distance; |
/* Removal Process */ |
Remove the template item with the worst DTW-score in the database; |
/* Process of learning data */ |
Feature extraction; |
If (Removal process finished == TRUE) then |
Add into the referenced pattern index; |
For each frame ( to total frames of the learning data) |
Add featurest into reference templates; |
End For |
End If |
End If |
Else |
/* Improper data and no learning */ |
End If |
Else |
/* End of learning process (No learning) */ |
End If |