Minimally segmenting performance Bangla optical character recognition using Kohonen network
Includes bibliographical references (page 5).
| Главные авторы: | , |
|---|---|
| Другие авторы: | |
| Формат: | Статья |
| Язык: | English |
| Опубликовано: |
BRAC University
2010
|
| Предметы: | |
| Online-ссылка: | http://hdl.handle.net/10361/630 |
| id |
10361-630 |
|---|---|
| record_format |
dspace |
| spelling |
10361-6302019-09-29T05:27:35Z Minimally segmenting performance Bangla optical character recognition using Kohonen network Shatil, Adnan Mohammad Shoeb Khan, Mumit Center for Research on Bangla Language Processing (CRBLP), BRAC University Bangla optical character recognition Includes bibliographical references (page 5). This paper presents a method to use Kohonen neural network based classifier in Bangla Optical Character Recognition (OCR) system, providing much higher performance than the traditional neural network based ones. It describes how Bangla characters are processed, trained and then recognized with the use of a Kohonen network. While there have been significant efforts in using the various types of Artificial ,eural ,etworks (A,,) in optical character recognition, this is the first published account of using a segmentation-free optical character recognition system for Bangla using a Kohonen network. The methodology presented here assumes that the OCR pre-processor has minimally segmented the input words into easily segmentable chunks, and presenting each of these as images to the classification engine described here. The size and the font face used to render the characters are also significant in both training and classification. The images are first converted into grayscale and then to binary images; these images are then scaled to a fit a pre-determined area with a fixed but significant number of pixels. The feature vectors are then extracted from the rectangular pixel map, which in this case is simply a series of 0s and 1s of fixed length. Finally, a Kohonen neural network is chosen for the training and classification process. Although the steps are simple, and the simplest network is chosen for the training and recognition process, the resulting classifier is accurate to better than 98%, depending on the quality of the input images. Adnan Mohammad Shoeb Shatil Mumit Khan 2010-10-24T04:57:57Z 2010-10-24T04:57:57Z 2006 2006 Article http://hdl.handle.net/10361/630 en 5 pages application/pdf BRAC University |
| institution |
Brac University |
| collection |
Institutional Repository |
| language |
English |
| topic |
Bangla optical character recognition |
| spellingShingle |
Bangla optical character recognition Shatil, Adnan Mohammad Shoeb Khan, Mumit Minimally segmenting performance Bangla optical character recognition using Kohonen network |
| description |
Includes bibliographical references (page 5). |
| author2 |
Center for Research on Bangla Language Processing (CRBLP), BRAC University |
| author_facet |
Center for Research on Bangla Language Processing (CRBLP), BRAC University Shatil, Adnan Mohammad Shoeb Khan, Mumit |
| format |
Article |
| author |
Shatil, Adnan Mohammad Shoeb Khan, Mumit |
| author_sort |
Shatil, Adnan Mohammad Shoeb |
| title |
Minimally segmenting performance Bangla optical character recognition using Kohonen network |
| title_short |
Minimally segmenting performance Bangla optical character recognition using Kohonen network |
| title_full |
Minimally segmenting performance Bangla optical character recognition using Kohonen network |
| title_fullStr |
Minimally segmenting performance Bangla optical character recognition using Kohonen network |
| title_full_unstemmed |
Minimally segmenting performance Bangla optical character recognition using Kohonen network |
| title_sort |
minimally segmenting performance bangla optical character recognition using kohonen network |
| publisher |
BRAC University |
| publishDate |
2010 |
| url |
http://hdl.handle.net/10361/630 |
| work_keys_str_mv |
AT shatiladnanmohammadshoeb minimallysegmentingperformancebanglaopticalcharacterrecognitionusingkohonennetwork AT khanmumit minimallysegmentingperformancebanglaopticalcharacterrecognitionusingkohonennetwork |
| _version_ |
1814309838295924736 |