Week 4: Finished Product
Introduction
The aim of this week was to carry out further tests on the handwriting recognition software program to make certain that it works as expected and any limitations are known and considered. This main objective was to have a deliverable product ready by the end of the week for assessment (during week 5's bench inspection).
Project Progression
Some of the results form further tests carried out during this week:
Test Result 1 (unsuccessful at fully recognising all the written characters) |
Test Result 2 (unsuccessful at fully recognising all the written characters) |
As can be seen from the pictures above, further testing on the program revealed a limitation that hadn't been observed prior: some letters being completely ignored by the program and therefore not being recognised at all. After several attempts of testing different handwritings, it was found that writing characters straighter and each with the same size had a greater chance of being recognised by the program. Additionally, when characters were written too closely to each other, some of them weren't recognised i.e. it was best to leave little gaps between each letter (even in a word) for the program to notice all written characters in the word. By adapting to the program, the written sentences were recognised more accurately and consistently as can be seen in the picture below.
Successful Recognition of Test Result 2 |
As a result of time constraints, the team members had to adapt to program rather than changing the program to adapt to the users (team members) which could possibly have been done by using bespoke training data rather than pre-processed datasets. Such limitations caused by time constraints have been considered and explored in the report writing as well as the presentation for the bench inspection.
Lastly, the poster and presentation for the bench inspection was completed this week with plans to finish report writing after week 5.
This concludes our project on Handwriting Recognition Software!