Handwritten English Character Recognition Using Logistic Regression and Neural Network

Tapan Kumar Hazra, Rajdeep Sarkar, Ankit Kumar

Abstract: Hand written character recognition is a challenging task often resulting in ambiguous labels. Using the concepts of Machine Learning we have tried to develop an Optical Character Recognition (OCR) system where an algorithm is trained on a data set of known letters and then can learn to accurately classify new data. Our optical character recognition (OCR) system for handwritten English characters comprises of two steps-Generating training set data using an OCR tool and then Applying different machine learning algorithm on the training set and start the learning process. A variety of algorithms have shown good accuracy for the handwritten letters, two of which are looked here.

Keywords: neural network, classification, optical character recognition, regularization, logistic regression

How to Cite?: Tapan Kumar Hazra, Rajdeep Sarkar, Ankit Kumar, "Handwritten English Character Recognition Using Logistic Regression and Neural Network", Volume 5 Issue 6, June 2016, International Journal of Science and Research (IJSR), Pages: 750-754, https://www.ijsr.net/getabstract.php?paperid=NOV164228, DOI: https://dx.doi.org/10.21275/NOV164228

Download Citation: APA | MLA | BibTeX | EndNote | RefMan