KNN and ANN-based Recognition of Handwritten Pashto Letters using Zoning Features
This work addresses the lack of standard datasets and recognition systems for handwritten Pashto letters, but it is incremental as it applies existing methods to a new domain.
The paper tackled handwritten Pashto letter recognition by creating a dataset of 4488 images and using zoning features with KNN and neural network classifiers, achieving accuracies of 70.05% and 72% respectively.
This paper presents a recognition system for handwritten Pashto letters. However, handwritten character recognition is a challenging task. These letters not only differ in shape and style but also vary among individuals. The recognition becomes further daunting due to the lack of standard datasets for inscribed Pashto letters. In this work, we have designed a database of moderate size, which encompasses a total of 4488 images, stemming from 102 distinguishing samples for each of the 44 letters in Pashto. The recognition framework uses zoning feature extractor followed by K-Nearest Neighbour (KNN) and Neural Network (NN) classifiers for classifying individual letter. Based on the evaluation of the proposed system, an overall classification accuracy of approximately 70.05% is achieved by using KNN while 72% is achieved by using NN.