Home
Call For Papers
Submission
Author
Registration
Publications
About
Contact Us

  Handwritten Character Recognition Using Residual Network  
  Authors : Anju Mohandas; Kavitha V K; Dr.Radhakrishnan B
  Cite as:

 

Handwritten character recognition has been one of the active research area in deep learning. This Recognition processing includes many applications such as reading bank cheques, converting written documents to structural text form. Handwritten recognition is a challenging task for computer system. Deep learning techniques are used for understanding the handwritten data through training. Recently used network is Convolution neural network for recognition process. In this paper, we used residual network for recognition. Before applying CNN we had performed image processing operations like pre-processing, conversion to greyscale, thresholding, image segmentation etc. With the use of residual networks, we can achieve fast training process and can attain more accuracy than other networks. Residual network is differing from Convolution neural network due to which residually adding a parallel connection to the layers of convolution neural network in order to providing better performance.

 

Published In : IJCSN Journal Volume 8, Issue 3

Date of Publication : June 2019

Pages : 243-245

Figures :01

Tables : 01

 

Anju Mohandas : received her B Tech (CSE) degree from University of Kerala in 2017. She is currently pursuing her Masters in Computer Science & Engineering from KTU. She has published one paper on data mining.

Kavitha V K : K is working as Assistant Professor in Computer Science and Engineering Department. She has more than 10 years' experience in teaching. Her research interests focus data mining and machine learning. She has published several papers on data mining.

Dr. Radhakrishnan B : is working as the Head of CSE department. He has more than 14 years experience in teaching and has published papers on data mining and image processing. His research interests include image processing, data mining, image mining.

 

Convolution Neural Network, Residual Network, Machine Learning, Python

There are many developments possible in this system in the future. Currently this system can only recognize text in English languages. Furthermore, languages can be recognized in future. Presently this system supports letter, digits and some special symbols. There are many applications of this system. Some of the applications are Processing of cheques in Banks, helping hand in Desktop publishing, recognition of text from business cards, helping the blind in recognizing handwritten text on letters.

 

[1] Handwritten Digits Recognition with Neural Networks and Fuzzy Logic; Wei Lu, Zhijian Li and Bingxue Shi. [2] Online Handwriting Recognition Using Support Vector Machine; Abdul Rahim Ahmad Christian Viard-Gaudin, Marzuki Khalid, Emilie Poisson. [3] Support Vector Machine (SVM) For English Handwritten Character Recognition; Dewi Nasien, Habibollah Haron, Siti Sophiayati Yuhaniz. 2010 Second International Conference on Computer Engineering and Applications. [4] Simple Convolutional Neural Network on Image Classification; Tianmei Guo, Jiwen Dong,Henjian Li'Yunxing Gao. 2017 IEEE 2nd International Conference of Big Data Analysis. [5] Handwritten Character Recognition Using DeepLearning; Rohan Vaidya, Darshan Trivedi, Sagar Satra, Prof. Mrunalini Pimpale. Proceedings of the 2nd International Conference on Inventive Communication and Computational Technologies.