Recognition of Meetei Mayek characters using hybrid feature generated from distance profile and background directional distribution with Support Vector machine classifier

Abstract

In this paper we have discussed the recognition of Meetei Mayek script with a Support Vector machine classifier. Distance profile feature and background directional distribution features are used as the feature vectors for training the SVM classifier. A comparative study is made on the performance between profile feature and background directional feature efficiency using SVM. Then a hybrid feature is generated by combining these two features and comparison of accuracy is done with the existing feature. Isolated handwritten documents are collected in some forms and experiment is performed over this dataset. For training the system the collection of documents is done from people from varying age group with different work background, so that the system can work well if we take the testing dataset from real world documents.