An Intelligent System For Writer Identification Of Devanagari Handwritten Documents Using ANN
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2015, Vol 3, Issue 12
Abstract
Many studies about writer identification have been done on different handwritten documents but less no of researches has been reported on Hindi or Devanagari documents. Here, in this technique, we accept handwriting as texture image and feature extraction done from preprocessed documents image. Significantly, the feature of presented method uses feature extraction which is suitable for structure of Hindi handwritten texts. And a new feature extraction technique is also presented which is based on classifier design. Firstly, we study different methods for feature extraction from output of preprocessing. This paper represents performance of slant estimation techniques on Devanagari handwritten documents. Slant is a unique feature; it acts as a vital role in writer identification, OCR and graphology systems. For estimation of slant, Radon Transform and Hough Transforms are applied on the word images. Further to improve the performance of these techniques, the extracted images are reconstructed using Zernike moments. For conducting this experiment, samples available on CPAR-2012 for Hindi datasets are used. The experimental result shows that slant estimated from images reconstructed using Zernike moment gives better results.
Authors and Affiliations
Praveen Sagar, Akhilesh Pandey
A Survey on Digital Watermarking with Wavelet Transform Domain
Digital video watermarking is a method for embedding some data into digital video sequences e.g. text, audio, image, video. Digital watermarking was developed to provide the copy right protection and owners’ authenticat...
Advanced Surveillance Robot
In the current context of increased surveillance and security, more sophisticatedly advanced and technologically innovative surveillance systems are needed. The detection or tracking of human in highly restricted areas...
Energy Efficient Street Light Supervising and Control System
In this proposal, we propose an energy efficient streetlight supervising and control system that can monitor and control streetlight more efficiently and can provide a safe night time environment for all road users incl...
A Robust But Flexible Privacy Preserving Scheme Supporting Cloud Environments
Fog is the Ruby cloud services library. The fog binary allows quick and easy access to configured cloud services. Whether you need compute, storage, or a multitude of other services, fog provides an accessible entry poi...
Analysis of Multicloud Priority Based Job Scheduling For Loosely Coupled Tasks
Cloud computing technologies can offer important benefits for IT organizations and data centers running MTC applications: elasticity and rapid provisioning, reduced capital costs, access to potentially unlimited resourc...