Using moments features from Gabor directional images for Kannada handwriting character recognition

Abstract

Handwriting character recognition (HCR) for Indian Languages is an important problem where there is relatively little work has been done. In this paper, we investigate the moments features on Kannada handwritten basic character set of 49 letters. Moments features are extracted from the preprocessed original images by most of the researchers. Kannada characters are curved in nature with some symmetry observed in the shape. This information can be best extracted as a feature if we extract moment features from the directional images. So we are finding 4 directional images using Gabor wavelets from the dynamically preprocessed original images. We then extract moments features from them. The comparison of moments features of 4 directional images with original images when tested on Multi Layer Perceptron with Back Propagation Neural Network shows an average improvement of 13% from 72% to 85%. The mean performance of the system with these two features together is 92%.

Projection-based features: A superior domain for handwritten Bangla basic characters recognition

“…Moments features were extracted from Gabor wavelets by Ragha and Sasikumar [17] for Kannada handwritten character recognition.…”

Handwritten Gurmukhi Numeral Recognition using Zone-based Hybrid Feature Extraction Techniques

“…For these applications, the performance (accuracy and speed) of digit recognition is most important factor. While in pattern classification and machine learning communities, the problem of handwritten digit recognition is a good example to test the classification performance [3]. HCR is very valuable in terms of the variety of applications and also as an academically challenging problem.…”