cargoProtecTPro
Justice is one text away.
Inspiration
After trying out many, many ideas, we decided to tackle the challenge of cargo theft using facial recognition. We wanted to develop something that solved a real world problem while learning to work with new technologies.
What it does
A facial recognition system in our app keeps track of the faces of registered drivers for the cargo company, and is able to detect whether someone in the driving seat is unregistered. If the person is not recognized, our app immediately sends an alert in SMS form to the company that a cargo theft is occurring. An alert is also sent if unrecognized people are detected in the vehicle cargo storage.
How I built it
We used OpenCV and machine learning for the facial recognition and human body detection in Python.
For the SMS alert system, we used Android Studio and Java.
In order to connect the two systems, we explored using REST APIs, web servers, and PHP.
Challenges I ran into
We did not know what project to make, and the idea generation alone took us hours. Once we did decide to use facial recognition for cargo theft, we wanted to make an immediate call to the company rather than send a text, but it was not working and a text alert would also be more useful in this situation. We also had challenges with connecting our Android Studio project to the OpenCV project.
Accomplishments that I'm proud of
Most of us learned how to use Android Studio for the first time. Coming up with the idea and working on the process was also an accomplishment.
What I learned
Using Android Studio, OpenCV, etc.
What's next for cargoProtecTPro
We hope that we can include location services in our SMS alert, so companies could immediately find out where the theft is occurring. We want to improve the accuracy of our facial recognition system.
Log in or sign up for Devpost to join the conversation.