Week 1: A Start To The Project
Introduction
After completing initial research into handwriting recognition systems and understanding how they work, plans were made to begin developing the final code on Python. Furthermore, the first academic advisor meeting regarding the project was held where ideas for how the project would be developed was discussed.
Project Progression
After researching into different types of handwriting recognition software, a word document summarising the useful information that was learnt was created. This document contains information such as definitions for key terminology used in a handwriting recognition software (e.g. neural networks, activation function, epoch etc..) as well as notes on training, validation and test sets that will be implemented into the program during its development. Below are a few pictures highlighting some of the information found in this document.
In short, this document will be used by the team to quickly refer back to concepts they learnt during the research week (week 0) which they might forget in the upcoming weeks.
The team also managed to create an initial draft of the requirements specification identifying some of key benchmarks they want to achieve by the end of the project.
With guidance form the academic advisor, a structure on how the report would be written out was drafted.
The algorithm that was created in the previous week was finalised after a discussion with the academic advisor. Additionally, the team members came together during the week to decide the dataset they would use to train the handwriting recognition software: EMNIST was chosen. This dataset can be found from the following link: https://www.kaggle.com/crawford/emnist/code?datasetId=7160&sortBy=voteCount. The team members decided on this dataset as it contains a balanced number of samples of letters and numbers as well as being less heavy with only 100k samples unlike the typical 800k. Even with fewer samples, the dataset had been found to give a high accuracy result.
GUI Implementation Research
After an initial discussion with the academic advisor, it was concluded that a GUI would be implemented into the handwriting recognition software so as to make the operation of reading images with handwritten words and displaying a text version of it on screen easy and efficient. Research on how to write code to create a simple GUI was carried out. This was done by following video guides on YouTube for which some of the links are below:
Plans for week 2
- Begin and complete designing functions for the handwriting recognition system on python using the requirements specification and practice test code as a basis
- Start writing up aspects of the report not relevant to the code and its testing (e.g. introduction to project, project specification, what programming used for the project and why)
- If possible, begin testing the functions that was created and plan for future improvements
Comments
Post a Comment