Hand Writing Recognition

The recognition of handwritten text is one of the most difficult problems in the field of image detection. One of the reason is that there are virtually infinite ways a human can write the same message. Hand writing recognition with the help of deep learning approach has shown some comparatively acc

Project Title

Hand Writing Recognition

Project Area of Specialization

Artificial Intelligence

Project Summary

The recognition of handwritten text is one of the most difficult problems in the field of image detection. One of the reason is that there are virtually infinite ways a human can write the same message. Hand writing recognition with the help of deep learning approach has shown some comparatively accurate results.. Generally, the handwritten text recognition process consists of four steps: data preprocessing, segmentation, the feature extraction and selection and the application of supervised learning algorithms. In this project we aim to investigate all the past existing deep learning based hand written text recognition algorithms and then propose a system based upon the most accurate algorithm. 

Project Objectives

We aim to achive the following obejectives

1. A new Hand Written Text Recognition system trained on a new local data set having higher accuracy than the existing systems.
2. To make thekind of system that will be able to help in medical fields such as doctor prescription writing in online doctor services and decrease the error probability in medicine identification in  pharmacies.
3. All the research findings and  conclusions  to be compiled in the form of a research paper.

Project Implementation Method

 Our system will depend upon the the scanned images of written text. For this purpose the user will use his camera to feed a image to our system and then our system will convert the image to digital text. One more method is that the user will scan the text paper and then feed the image to our system to convert the text.

Benefits of the Project

This project will benefit anyone who wants to convert any form of handwritten text into digital text. Our project will target the medical fields such as online medical services and medicine identification in pharmacies. Our project will be able to decrease the error probability in medicine identification.
This project will benefit in storing the meeting notes, invoices, taxes and questionaires in a digital manner. This will help in better preservation and a more organised information extraction.

Technical Details of Final Deliverable

Our final deliverables will be

1- Convert any form of soft or hard copy hand written text into digital text.
2- Explore all the past HWTR neural network algorithms and propose a new better algorithm using the past or combinations of past algorithms.

3- Detect the medical prescriptions with high accuracy.
4-  Prepare a local hand written database.

Final Deliverable of the Project

Software System

Type of Industry

IT , Medical

Technologies

Artificial Intelligence(AI), Big Data

Sustainable Development Goals

Good Health and Well-Being for People, Decent Work and Economic Growth, Sustainable Cities and Communities

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com