An estimated 4.25 million adults aged 40 years in Nigeria are visually impaired or blind. Reference and this number is expected to rise by average of 400,000 every year.
The naira does not have any marking to aid the visually inpaired to know which note they are holding.
This people interact with the community every day. Imagine taking a ride on a bus and on getting there he is required to pay, the man will depend on the honesty of the driver to know the amount he/she is holding.
The android application is to help them know to the amount in thier hands before giving the driver.
We would solve this problem with Machine learning by training a neural network using keras with tensorflow backend.
With images of the naira note from #5 denomination to #1000. At different position to train the model for better accuracy.
This will improve their lifes becuase they wiil no longer depend on the honesty of the drivers to know the naira note in their hands
The building process will be in two phase
- Prove of concept
During this time, we would train with two naira note to understand.
Do some tweaking, hyperparameter tuning of the model. Just to know what works best. - Full model build
This phase will involve training the model on the eight (8) naira notes, beta testing and tuning the model for better accuracy, precision,and sensitivity.
Building the android version
- Keras with Tensorflow backend
- Android studio 3.0+
- Android 5.0+ upwards
- Counting of the notes
- Audio voice for Other languages in Nigeria
- Image Recognition for other Country note