An Efficient Model for Isolated Vietnamese Handwritten Recognition

Isolated vietnamese handwriting recognition embedded system applied combined feature extraction method

“…That is the reson for why algorithms applied in image recognition are effective to approach the character recognition systems. Correctly, many algorithms hidden Markov model [1], neural network [2], or support vector machine (SVM) [6] confirm the high performance; however, most researches have referred to SVM algorithm as the best method illustrated in [4][6] [8]. Normally, both training and recognition processes are implemented on high power PC to tackle the complicated training SVM algorithm and long converge time.…”

“…In terms of Vietnamese characters, a combination between alphabet characters and sign has challenged the handwriting recognition system. That is also the main reason that just some researches have mentioned the isolated Vietnamese handwriting as presented in [6] [7]. As regards evaluation of handwriting recognition system, almost authors have made decision to choose the software approaches to experience the system performance on other high power personal computers [6] [7].…”

“…That is also the main reason that just some researches have mentioned the isolated Vietnamese handwriting as presented in [6] [7]. As regards evaluation of handwriting recognition system, almost authors have made decision to choose the software approaches to experience the system performance on other high power personal computers [6] [7]. This becomes critical issue to be able to promote the handset applications in limited resource conditions or meet the real time requirements strictly.…”

“…6 The steps from 1 to 5 are executed again to detect the next word lines. Based on the values of histogram image, the word separation is identified by the zero value continuously.…”