>
Fa   |   Ar   |   En
   recognition of handwritten persian two-digit numerals using a novel hybrid svm/hmm algorithm  
   
نویسنده moradi vahid ,razzazi farbod ,behrad alireza
منبع majlesi journal of electrical engineering - 2016 - دوره : 10 - شماره : 3 - صفحه:13 -19
چکیده    there is a vast range of potential applications for recognition of handwritten persian / arabic digits (e.g. bankingtransactions, business registration forms and postal code recognition to name a few). in this paper, a new method ispresented for automatic recognition of joint two-digit persian numerals. the proposed method is composed of acombinational structure of support vector machines (svm) and a hidden markov models (hmm). in this approach,we used svm and hmm for classification and segmentation goals respectively. due to the higher performance ofsvm in classification with respect to hmm, the main core of recognition is a svm classifier. in contrast, we usedhmm to detect the location of the boundary for two-digit numerals. to evaluate the method, we employed a selectionof hadaf persian isolated characters corpus. we employed a 4 scale gabor filter bank (24, 12, 6 and 3 scales) in 6directions (0, 30, 60, 90. 120. 150 degrees) for feature extraction. the results showed the digit recognition rate ofabout 98.75 percent for the proposed algorithm on persian two-digit numerals, while the recognition rates were 98.58and 95.93 for separate svm and hmm engines on isolated characters respectively.
کلیدواژه persian handwritten numeral recognition ,svm/hmm combining classifier
آدرس islamic azad university, science and research branch, electrical and computer engineering department, iran, islamic azad university, science and research branch, electrical and computer engineering department, iran, shahed university, electrical and electronic engineering department, iran
پست الکترونیکی behrad@shahed.ac.ir
 
 

Copyright 2015
Islamic World Science Citation Center
All Rights Reserved