When browsing for smartphone apps that provide gasoline prices there were issues that were commonly present. One of them being inaccurate or out of date information.
What it does
GasoLean makes use of artificial intelligence to recognize prices from a picture of a sign that is found at every gas station. The user is then prompted to confirm whether the information recognized is accurate, but never given the option to manually input information. By taking this approach we are able to minimize human error while maintaining an up to date database. Then, by extracting metadata from the picture taken, the date the picture was taken and the geolocation information are used to both update the information on the corresponding gas station and maintain up to date information of the pricing respectively.
How we built it
Using the Microsoft computer vision api we were able to provide text recognition to the app. In order to compare match price information to the correct gas station we used the google maps places api. The main functionality is provided by the java programming language.
Challenges we ran into
Figuring out how to use the different APIs was the main challenge of this project, since we had no prior knowledge on how to use them beforehand, so we had to learn as we worked on the project.
Accomplishments that we're proud of
We're proud of being able to succesfully implement the mentioned API's to provide the necessary functionality for the program.
What we learned
We learned a lot about the possible uses for the different APIs provided by both Google and Microsoft and discovered more about java programming.
What's next for GasoLean
The original idea for this project was making an android app in which the camera was used directly within the app to provide the necessary info. Since we had no prior knowledge of android studio, we were unable to make it into an android app, but mainly that is the next step. Additionally, we would like to make use of a database to store the values so that everyone can have access to them and use location services further to suggest low prices around the user's location