No abstract

Offline Handwritten Character Recognition of Malayalam scripts have gained remarkable attention in the past few years. The complicated writing style of Malayalam characters with loops and curves make the recognition process highly challenging. This paper presents a comparative study of Malayalam character recognition using 4 different feature sets-Zonal features, Projection histograms, Chain code histograms and Histogram of Oriented Gradients. The performance of these features for isolated Malayalam vowels and 5 consonants are evaluated in this study using feedforward neural networks as classifier. The final recognition results were computed using a 5 fold cross validation scheme. The best recognition accuracy of 94.23 % was obtained in this study using Histogram of Oriented Gradients features.

ABSTRACT

This paper presents a multiple classifier system for the recognition of offline handwritten Malayalam characters. The features used are the gradient and density based features. These feature sets are fed as input to two feedforward neural networks. The results of both these neural networks are combined using four different combination schemes: Max rule, Sum rule, Product rule and Borda count method. The best combination ensemble with an accuracy of 81.82% is obtained by using the Product rule combination scheme.

A hypergraph consisting of hyperedges and nodes can be made intuitionistic fuzzy hypergraph (IFHG) by assigning membership and nonmembership degrees for both nodes and edges. Just as a hypergraph, an IFHG is also having hyperedges consisting of many nodes. Many flavours of a given IFHG can be created by applying morphological operators like dilation, erosion, adjunction, etc. The focus of this paper is to define morphological adjunction, opening, closing, and Alternative Sequential Filter (ASF) on IFHG. The system modeled in this way finds application in text processing, image processing, network analysis, and many other areas.