Title : 
Recognition online Arabic pattern
         
        
            Author : 
Omer, Marwan Ali H ; Ma, Shilong
         
        
            Author_Institution : 
Sch. of Comput. Sci. & Eng., Beihang Univ., Beijing, China
         
        
        
        
        
            Abstract : 
It very hard to collect a large set especially for Arabic handwriting samples used as training set. We generate a good and clear handwriting Arabic character samples for 28 letters, the algorithm learns the strokes directions and properties of the handwriting Arabic character from those real samples. In this paper, we present decision tree to classify the characters belongs to dots, baseline detection to discover the baseline of the character to detect dot position, and matching algorithm to learn the stroke direction and properties of the Arabic character and recognize the input online Arabic handwriting isolated character. Experiments show the results and achieved a high online recognition rate for isolated Arabic character.
         
        
            Keywords : 
decision trees; handwriting recognition; handwritten character recognition; pattern recognition; Arabic handwriting; Arabic pattern; baseline detection; matching algorithm; online recognition; stroke direction; baseline detection; decision tree; matching algorithm; online Arabic character recognition;
         
        
        
        
            Conference_Titel : 
Advanced Computer Theory and Engineering (ICACTE), 2010 3rd International Conference on
         
        
            Conference_Location : 
Chengdu
         
        
        
            Print_ISBN : 
978-1-4244-6539-2
         
        
        
            DOI : 
10.1109/ICACTE.2010.5579357