UCSY
[ Back ]

Each day, billion of business and financial documents have to be processed by computer. The great bulk of them are still processed manually by human operators, the most common and labor-consuming operation being document amount reading and typing. A common way to automate the process is to replace the human operator with an off-line recognition system that is able to do the operator’s job. Pattern recognition includes such applications as speech and character recognition for various languages, visual image recognition and classification. Writing has been the most natural mode of collecting, storing and transmitting information through the countries, now serves not for communication among human but also serves for communication of human and machines.

Handwritings can be used for forensic tasks and for person authentication. According to the way handwriting data is generated, two different approaches can be distinguished: on-line and off-line. The data are captured during the writing process by a special pen on an electronic surface in online system. In off-line recognition system, the data are acquired by a scanner after the writing process is over. Recognition of off-line character is more-complex than the on-line case due to the presence of noise in the image acquisition process and the loss of temporal information such as the writing sequence and the velocity. This information is very helpful in a recognition process.



Images of Myanmar printed and hand written characters

Method to recognize handwritten words are well known and widely used for many different languages. Hidden Markov Models (HMM) have been very successfully implemented for recognizing cursive written words. The process of recognition system involves preprocessing, segmentation, feature extraction, and classification. Several type of decision method including statistical methods, neural networks, structural matching (on trees, chains, etc) and stochastic processing (Markov chains, etc) will use along with different types of features.


The both of Myanmar handwriting and English handwriting are performed in the Character recognition research work. An automatic data entry system for Myanmar bank cheque and Myanmar passport are developed.

Publications

  1. San San Mon and Myint Myint Sein: Recognition of Myanmar Handwriting Text Based on Hidden Markov Model, The proceeding of Fifth International Conference on Computer Application(ICCA), 2007.[pdf]
  2. Nang Aye Aye Htway, Thi Thi Soe and Myint Myint Sein: Automatic Extraction of Payee's Name and Legal Amount from Myanmar Cheque by Using Hidden Markov Model, The proceeding of Fifth International Conference on Computer Application(ICCA), 2007.[pdf]
  3. Nang Aye Aye Htway, San San Mon and Myint Myint Sein: Recognition on User-entered Data from Myanmar Bank Cheque, The proceeding of Sixth International Conference on Computer Application(ICCA), 2008.[pdf]
  4. Mya Mya Thinn and Myint Myint Sein: Implementation of Automatic Data Entry System Using Myanmar Passport, The proceeding of the third Malaysian Software Engineering Conference, MySEC'07.[pdf]
  5. Mya Mya Thinn and Myint Myint Sein: Hand Written Recognition System of Passport for Automatic Data Entry, The proceeding of 28th ASEAN Conference on Remote Sensing (ACRS 2007), Kuala Lumpur, Malaysia, November (2007).[pdf]
  6. Mya Mya Thinn and Myint Myint Sein: Automatic Data Entry of Passport for Security System, The proceeding of Fifth International Conference on Computer Application(ICCA), 2007.[pdf]
  7. Mya Mya Thinn and Myint Myint Sein: Robust Segmentation for Automatic Data Extraction from Passport, The proceeding of Sixth International Conference on Computer Application(ICCA), 2008.[pdf]
News
Computer Universities
Contact us
Contact:
No.(4) Main Road ,
Shwe Pyi Thar Township,
Yangon, Myanmar.
Fax: +95 01 610633
Phone: +95 01 610655
Email: ucsy.admin@ucsy.edu.mm
Facebook page: https://www.facebook.com/ucsygn/