Recognition of Isolated Handwritten Kannada Numerals based on Decision Fusion Approach
Journal Title: International Journal on Computer Science and Engineering - Year 2013, Vol 5, Issue 6
Abstract
combining classifiers appears as a natural step forward when a critical mass of knowledge of single classifier models has been accumulated. Although there are many unanswered questions about matching classifiers to real-life problems, combining classifiers is rapidly growing and enjoying a lot of attention from pattern recognition and machine learning communities. For any pattern classification task, an increase in data size, number of classes, dimension of the feature space and interclass separability affect the performance of any classifier. It is essential to know the effect of the training dataset size on the recognition performance of a feature extraction method and classifier. In this paper, an attempt is made to measure the performance of the classifier by testing the classifier with two different datasets of different sizes. In practical classification applications, if the number of classes and multiple feature sets for pattern samples are given, a desirable recognition performance can be achieved by data fusion. A framework for feature selection and decision fusion has been proposed in this paper to increase the performance of classification. From the experimental results it is seen that there is an increase of 4.55% in the recognition accuracy.
Authors and Affiliations
Mamatha. H. R , Sucharitha Srirangaprasad , Srikantamurthy K
Keywords
Related Articles
- EP ID EP87509
- DOI -
- Views 155
- Downloads 0
Anti-Synchronization of the Hyperchaotic Liu and Hyperchaotic Qi Systems by Active Control
This paper investigates the problem of anti-synchronization of identical hyperchaotic Liu systems (2008), hyperchaotic Qi systems (2008) and non-identical hyperchaotic Liu and hyperchaotic Qi systems using active nonline...
IMPROVED HYBRID MODEL FOR DENOISING POISSON CORRUPTED XRAY IMAGES
Medical practitioners are increasingly using digital images during disease diagnosis. Several state-of-the-art medical equipments are producing images of different organs, which are used during various stages of analysis...
ANALYSIS OF ELECTRONIC VOTING SYSTEM IN VARIOUS COUNTRIES
India is the world’s largest democracy with a population of more than 1 billion; India has an electorate of more than 668 million and covers 543 parliamentary constituencies. Voting is the bridge between the governed and...
Entropy Based Texture Features Useful for Automatic Script Identification
In a multi script environment, a collection of documents printed in different scripts is in practice. For automatic processing of such documents through Optical Character Recognition, it is necessary to identify the scri...
Restoration of Normal Frequency Affected by Small Load Variations Through HVDC Link Using Neuro-Fuzzy Approach
HVDC power transmission is coming up with merits to replace the EHV-AC systems. The controller inverter operation can successfully regulate the power in HVDC link leading to fulfillment of power demand in A.C. networks c...