Scribd
Upload
21 views

Handwritten Character Recognition

This document summarizes handwriting character recognition using a convolutional neural network (CNN). It discusses the main goals of handwriting recognition in applications like processing bank checks and mobile addresses. It introduces CNNs for their ability to better extract features compared to artificial neural networks, while using fewer parameters. The document describes training a CNN model on a dataset containing over 37,000 handwritten alphabet images in 28x28 pixels for the tasks of training and testing the recognition abilities of the network. It concludes that CNNs achieve fairly good performance for handwriting recognition.

Uploaded by

Keerthi Prem
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
21 views

Handwritten Character Recognition

This document summarizes handwriting character recognition using a convolutional neural network (CNN). It discusses the main goals of handwriting recognition in applications like processing bank checks and mobile addresses. It introduces CNNs for their ability to better extract features compared to artificial neural networks, while using fewer parameters. The document describes training a CNN model on a dataset containing over 37,000 handwritten alphabet images in 28x28 pixels for the tasks of training and testing the recognition abilities of the network. It concludes that CNNs achieve fairly good performance for handwriting recognition.

Uploaded by

Keerthi Prem
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 15

HANDWRITTING CHARECTOR

RECOGNITION
CONTENTS

1. Main Goals and Application

2. Introduction

3. Neural Network -CNN

4. Conclusion
01 Main Goals and
Application
Main Goals and Applications

• Handwritten Digit Recognition is used to recognize the


charectors which are written by hands.
• A handwritten digit recognition system is used to visualize
artificial neural networks.
2017
• Its already widely used in the automatic processing of
bank cheques , postal addresses in mobile phones etc.
02 Introduction
Introduction

• Scientist believe that the most intelligent device is the


human brain.
• There is no computer which can beat the level of
efficiency of human brain .These inefficiencies of the
computer has lead to evolution of “Artificial neural
Network”.
• They differ from conventional systems in the sense that
rather than being programmed these system learn to
recognize patterns.
CONVOLUTION
03 NEURAL
NETWORK
Why CNN?
• Convolution is a simple mathematical operation between
two matrices in which one is multiplied to the other
elements wise and sum of all these multiplied is calculated.
• Convolutions are performed for various reasons-
 Convolutions provide better feature extraction.
 They save a lot of computation compared to ANNs.
 Less number of parameters are created than those in pure
fully connected layers.
 Due to less no. of required parameters, lesser fully
connected layer are needed.
CNN

Training Testing Preprocessing Segmentation


Dataset
dataset
CONVOLUTION NEURAL NETWORK
04 Training Dataset
Training Dataset

• A-Z Handwritten Alphabets is the name of the dataset,


which is kept in a.csv file.
• There are two sets of 372450 letter images in it: a
training set and a testing set.
2017
• All the images in the dataset are of 28* 28 pixels,where
each pixel’s binary intensity value.
• The intensity of each pixel is positioned in the centre of
the greyscale images.
TEMPLATE

CONCLUSION

TEMPLATE
CONCLUSION

• The handwritten recognition using convolution neural network has


proved to be a fairly good efficiency.

• It works better than any other algorithms , inclusing artificial neural


networks.
THANK YOU

You might also like