Jurassic Parking

• 

  Real tests were successful!

• 

  Training Data Testing

• 

  Confusion Matrix

• 

  Training Graph

Inspiration

The inspiration behind Jurassic Parking comes from the universal frustration of wasting time circling parking lots searching for an open spot. It's driven by the desire to leverage technology for solving everyday problems, making urban living more efficient and reducing the environmental impact of vehicles idling unnecessarily. The advancements in AI and machine learning, specifically object detection with YOLO v8 and TensorFlow, presented a unique opportunity to tackle this issue head-on. The idea was born from a vision of seamless urban mobility, where technology and data analytics work together to ease congestion, lower stress levels, and improve the quality of life in crowded cities.

What it does

Jurassic Parking utilizes advanced object detection algorithms to scan and analyze parking lots in real-time, accurately identifying the number of available parking spots. By processing live images using YOLO v8 and TensorFlow, it provides instant updates on parking availability, which drivers can access via an app or digital signage. This system not only guides drivers directly to open spaces, saving time and reducing traffic within lots, but also offers valuable data analytics for parking lot operators to optimize space usage and improve overall efficiency.

How we built it

We kicked off building Jurassic Parking with some serious tech and a bunch of parking lot images—over 8,000, to be precise. First up, we got our hands dirty with YOLO v8 and TensorFlow, two powerhouse tools in the AI world, to teach our system how to spot empty parking spaces like a pro. It was a bit like training a super-smart dog to fetch, but instead of sticks, we were after those elusive open spots.

Challenges we ran into

We ran into a lot of technical difficulties throughout the project when it came down to integrating the front-end into the backend. The challenges of staying productive past the 2am mark was very difficult, and pushing through that was something we never thought we could do.

Accomplishments that we're proud of

We are immensely proud of the accomplishments we achieved with Jurassic Parking, particularly in the realm of technological innovation and real-world impact. The successful integration of YOLO v8 and TensorFlow for precise object detection stands as a testament to our commitment to leveraging cutting-edge technology for practical applications. Our system's ability to discern available parking spaces with remarkable accuracy was a pivotal achievement, underscoring our expertise in machine learning and computer vision.

The processing and analysis of over 8,000 images to train our algorithm not only demonstrated our technical prowess but also highlighted our dedication to creating a robust and reliable solution. This extensive dataset ensured that JP could operate effectively in diverse parking environments, showcasing our system's adaptability and scalability.

What we learned

We learned a lot this past weekend, one being not to procrastinate too much otherwise you might go the weekend without sleep. However we also learned how to work as a team and how to coordinate together. We also learned the important role of a PM.

What's next for Jurassic Parking

We will be starting our Series A Funding run soon along with joining Y Combinator. Jokes aside, the real next step is scaling up, refining our algorithms, and maybe, just maybe, keeping an eye out for any parking needs in the Mesozoic era. Because at Jurassic Parking, we're all about solving tomorrow's parking problems today... and possibly yesterday's, if you count the dinosaurs.

Built With

• ai
• javascript
• ml
• neural-networks
• python
• react