Review of Automatic Handwritten Kannada Character Recognition Technique Using Neural Network
- Mukesh Kumar
- Dr.Jeeetendra Sheethlani
Back Propagation Neural Network, Form Processing, Histogram of Gradients, Kannada Script, Principal Component Analysis.
Data processing and management is common now a days. In this paper, automatic processing of forms written in Kannada language is considered. A suitable pre-processing technique is presented for extracting handwritten characters. Principal Component Analysis (PCA) and Histogram of oriented Gradients (HoG) are used for feature extraction. These features are fed to multilayer feed forward back propagation neural network for classification. Only 57 characters are used for recognition. Performances of two features are compared for different number of classes. HoG is found to have better recognition accuracy than PCA as number of classes increased. This is implemented in Visual Studio 2010 using Open CV library.
Mukesh Kumar, Dr.Jeeetendra Sheethlani. "Review of Automatic Handwritten Kannada Character Recognition Technique Using Neural Network".INTERNATIONAL JOURNAL OF ENGINEERING DEVELOPMENT AND RESEARCH ISSN:2321-9939, Vol.5, Issue 4, pp.726-730, URL :https://rjwave.org/ijedr/papers/IJEDR1704118.pdf
Volume 5 Issue 4
Pages. 726-730
