Inspiration
We saw that bilnd and partially impaires perple have a difficulty to calculate their bills and determine their value on their own. They usually use the help of others to determine the value of bills which puts them in the danger of being scammed and being dependent to others. Our device serves as an answer to this problem.
What it does
It uses computor vision and AI to determine the value of the bill then it informs the user of the value by audio through a wired headset. It also Saves the images used for the AI training to use it later and improuve the AI model.
Challenges we ran into
The time-limit was our biggest challenge. We had a design challenge to create the proof-of-concept in the time limit. The availability of a reliable internet connection. The ideation and managing to stick to the specfication book. Training the model.
Accomplishments that we're proud of
The 3d design and the 3d print. The AI model and learning more obout vision.
What we learned
Using mqtt protocole. Time management. Team work. Using audio with raspberry.
What's next for Dinary Vision
Our next step would be to create a more advanced design. Add more features such as: navigatio detection and pedestrian crossing assistant
Built With
- 3dprinting
- airclone
- cloud
- cnn
- machine-learning-knl
- mlops
- opencv
- python
- tensorflow
Log in or sign up for Devpost to join the conversation.