Online Handwriting Recognition of Amharic Words Using Hmm
| dc.contributor.advisor | ASSABIE, YAREGAL (PhD) | |
| dc.contributor.author | Kornsa, Getnet | |
| dc.date.accessioned | 2018-06-20T07:01:13Z | |
| dc.date.accessioned | 2023-11-04T12:22:23Z | |
| dc.date.available | 2018-06-20T07:01:13Z | |
| dc.date.available | 2023-11-04T12:22:23Z | |
| dc.date.issued | 2011-11 | |
| dc.description.abstract | Computers are greatly influencing the lives of human beings and their usage is increasing at a tremendous rate. The ease with which we can exchange information between user and computer is of immense importance today because input devices such as keyboard and mouse have limitations in comparison with input through natural handwriting. We can use the online handwriting recognition process for a quick and natural way of communication between computer and human beings. Over the years, handwriting recognition is in research and has attracted many researchers across the world. The main goal of this thesis is to develop an online handwritten Amharic word recognition system. In this work, we present writer-independent HMM-based Amharic word recognition for online handwritten words. In our approach, the central idea is to build the HMM model for each word. The underlying units of the recognition system are a set of primitive strokes whose combinations form handwritten Amharic words. For each word, possibly occurring sequences of primitive strokes and their spatial relationships, collectively termed as primitive structural features, are stored as feature list. In the training phase the extracted features of each word are used as feature vectors which will be given as input parameters to each HMM model. In the case of recognition, a model for each separated word is built up using the same approach. This model is used by the system to perform the recognition using the Viterbi decoding algorithm. We also present a dataset collected for training and testing online recognition systems for Amharic words. The dataset is prepared in accordance with the international standard UNIPEN format. The recognition system is tested with the collected dataset and we achieved word recognition rates of 90.9% for numeral words and 73.94 % for other words. The overall recognition rate of the system is 79.54% for all words in the dataset. Keywords: Handwriting recognition, Amharic word recognition, Online Recognition. | en_US |
| dc.identifier.uri | http://etd.aau.edu.et/handle/123456789/2017 | |
| dc.language.iso | en | en_US |
| dc.publisher | Addis Ababa University | en_US |
| dc.subject | Handwriting Recognition; Amharic Word Recognition; Online Recognition | en_US |
| dc.title | Online Handwriting Recognition of Amharic Words Using Hmm | en_US |
| dc.type | Thesis | en_US |