Parking Scanner

Main and Front Page

Inspiration

We were inspired to build this project to help solve a problem people and even ourselves struggled with at one point. Parking signs, when spending almost an hour looking for parking, the last thing you want to see when finally getting into a spot is a pole with 20 different messages and not knowing whether you can park or not, do you pay or not, is it this specific spot?

What it does

Our application, parking scanner allows you to upload photos of any source and have it returned back to you, explaining what to do for that location. Also we incorporated multiple language support, to further strengthen society

How we built it

We built this through javascript, html, css and react as the front end, handling the file upload and overall display of the website. We had routes associate the linked connection between not only the upload portion but summary portion. Then we used flask, python and a couple of major python libraries to establish a system of uploading but also incorporating google ai for our summarization.

Challenges we ran into

The challenges we ran into was at first, merging different flask files, but also the connection between uploading on the main html, which then gets sent to the backend through javascript. The connection wasn't there at first.

Accomplishments that we're proud of

We were able to finish off a good majority of it in 1 sitting on the first day, and show strong progress of submission, however minor debugs still existed. Overall, we were proud of the front end and how react and css was able to make the website feel interactive. Alongside the connective backline files into one.

What we learned

Our project successfully transformed confusing city parking signs into instant, universal clarity by building a powerful full-stack application. We developed a highly polished, visually dynamic frontend featuring sophisticated animations and an intuitive drag-and-drop zone, ensuring the complex process felt simple and elegant to the user. On the backend, we created a robust pipeline using OCR (Optical Character Recognition) to extract text from images and fed that data directly into the Gemini AI.

What's next for Parking Scanner

The next evolution is making this a mobile app so it works exactly where the problem happens—at the curb. It will use your camera, GPS, and clock to instantly answer: "Can I park here right now?" This makes it a real-time parking assistant that eliminates stress and prevents tickets.