AutoAlert

Inspiration

Our idea came from a heartbreaking personal experience. On Christmas eve of Last year one of our families car was stolen from the driveway, this car was sentimental to one of us and also one of our group members nice sports car his dad had wanted his whole life. When it was stolen he was extremely sad and his dad was as well. During the start of this weekend he was told to brainstorm some ideas for our project, and after some thinking he realized that a major issue, so bad that the police said to leave your key fobs outside the house to protect your house, was car stealing. This combines a major issue which is plaguing society with his own personal motivation to solve this issue, and within my group everyone agreed this was a great idea.

What it does

With AutoAlert, you are keeping your car safe. Through real time tracking it is able to detect if someone is being suspicious near your car or trying to actively steal it. After someone is detected trying to steal your car or being suspicious it will alert you and the police, the alert is through SMS and is done because every second matters, and so you can keep your car safe.

How we built it

Our project utilizes Python, CV2, and trained a model using Roboflow to detect and track suspicious activity around vehicles in real-time by parsing video frames. Upon detecting potential threats, our system promptly sends SMS alerts to emergency services and users via the Twilio API. We developed a mobile application using React Native, enabling users to customize alert timing intervals and car details, offering a personalized and responsive security solution for vehicle owners

Challenges we ran into

Initially, we had no experience with React Native or iOS app development. A big issue was setting up the React Native environment, which involved learning to use command line tools and managing dependencies with npm. As we progressed, we faced challenges such as parsing lengthy videos, which slowed down the system due to the high number of frames. Additionally, we encountered difficulties with GitHub, including conflicts and branching errors. We also experienced issues with integrating Twilio for SMS notifications, which required troubleshooting and problem-solving to resolve.

Accomplishments that we're proud of

-We undertook the task of training our own AI model using collected data, which required significant learning and development. Concurrently, we learned React Native to build the front-end of our application. Our efforts resulted in the creation of a unique service/product that had not previously been developed. Throughout the process, we made substantial progress in teamwork, collaborating effectively to overcome challenges and achieve our goals.

What's next for AutoAlert

In the future, we plan to develop our own hardware camera integrated with our software, providing a seamless experience for users. Additionally, we aim to offer a subscription service for customers using third-party camera systems. We also intend to partner with government agencies, including law enforcement, and other security companies to collaboratively reduce criminal activity related to automobiles.