Abstract:
In busy city traffic, it's super important to quickly spot emergency vehicles like
ambulances. That’s what our project is all about - making traffic lights smarter by
detecting ambulances in real-time. We want to create a clever traffic signal system that
can see when an ambulance needs to get through. When that happens, the signal
changes from red to green so the ambulance can keep going without any delays on its
way to the hospital. To make this happen, we went out on the streets of Karachi and
collected data about all kinds of ambulances. This big set of information is the core of
our project, helping us train and check how well our system can spot ambulances. We
made an important choice to switch from YOLOv5 to YOLOv8, a move we carefully
thought about. YOLOv8 has some cool improvements that match what we need,
making our system really good at spotting ambulances accurately. Our project is not
just about technology; it's like a journey into the world of computers and learning
machines. All our hard work paid off, and we now have results showing our system is
more than 90% accurate in spotting ambulances during practice and testing. But our
goal is not just numbers - we want to make things easy for people to understand. So,
the next step for us is to create a simple interface that shows how our system works in
real life. This way, people can see for themselves how our system spots ambulances in
videos, proving that our traffic signal idea is practical and useful. In the end, our project
is a mix of technology, new ideas, and making city traffic better. We're using the latest
computer tricks to help traffic lights work smarter, especially in emergencies.
Switching to YOLOv8, gathering data in Karachi, and focusing on making things
simple for everyone are all part of our commitment to improve how traffic is managed.
