Sadr, H., Nazari Solimandarabi, M., Mirhosseini Moghadam, M. (2017). Categorization of Persian Detached Handwritten Letters Using Intelligent Combinations of Classifiers. Journal of Advances in Computer Research, 8(4), 13-21.
Hossein Sadr; Mojdeh Nazari Solimandarabi; Mahsa Mirhosseini Moghadam. "Categorization of Persian Detached Handwritten Letters Using Intelligent Combinations of Classifiers". Journal of Advances in Computer Research, 8, 4, 2017, 13-21.
Sadr, H., Nazari Solimandarabi, M., Mirhosseini Moghadam, M. (2017). 'Categorization of Persian Detached Handwritten Letters Using Intelligent Combinations of Classifiers', Journal of Advances in Computer Research, 8(4), pp. 13-21.
Sadr, H., Nazari Solimandarabi, M., Mirhosseini Moghadam, M. Categorization of Persian Detached Handwritten Letters Using Intelligent Combinations of Classifiers. Journal of Advances in Computer Research, 2017; 8(4): 13-21.
Categorization of Persian Detached Handwritten Letters Using Intelligent Combinations of Classifiers
1Department of Computer Engineering, Parand Branch, Islamic Azad University, Parand, Iran
2Young Researchers and Elite Club, Lahijan Branch, Islamic Azad University, Lahijan, Iran
3Department of Computer Engineering, Rasht Branch, Islamic Azad University, Rasht, Iran
Abstract
Abstract—detecting optical characters is the main responsibility to convert printed documents and manuscripts to digital format. In this article, detecting Persian handwritten letters by using the combination of classifiers and features were assessed, hence geometric and statistical sections' features were used. In order to detect each letter, we divide it into two parts; the major and the minor parts. Then, we present some features for them. Preprocess algorithm prepare the possibility to unify dimension features for multiple words and deliver to classifier for detecting . We can get the hierarchy classification by separating the letters. After that, the optimal answer will be reached by using GA method of different SVM, ML and KNN classifications. Extraction algorithm of needed features was proved by using the evaluation of the basis of PCA. Empirical results represent classification of 94.3 and 92 accuracy in simple and multiple parts in 20 times repetition, respectively. eywords— Classifiers' Combination, Optical Character recognition, Persian handwritten, Reducing feature.
Top