So, I’ve been messing around with this idea for a while now – tracking foot traffic in public transport stations. You know, like counting how many people are coming in and out, and when it’s the busiest.

First, I gathered some cheap cameras, nothing fancy, just ones that could capture a decent video feed. I set them up at a couple of exits at a local bus station. Not exactly high-tech, more like a DIY project, really. I had to climb some ladders, it was fun.

Next, I needed something to process the video. I found this open-source computer vision library – it’s free, which is great! I started playing around with it, trying to get it to recognize people in the video frames.

  • I spent hours, maybe days, just tweaking the settings.
  • It was a lot of trial and error.
  • Sometimes it would count a trash can as a person.
  • Sometimes it would miss people altogether.

But I kept at it. I read some blogs, watched some tutorials, and slowly, it started to get better. The system started to accurately count the number of people entering and exiting the station.

Then, I wrote a simple script to store the counts over time. I just wanted to see the patterns. Like, when are the rush hours? How many people use the station on a typical day? The script saved these numbers into a file.

The Outcome

Finally, I built a basic dashboard to visualize the data. I used some free charting tools to create graphs. And guess what? It worked! I could actually see the foot traffic patterns in the station. It was pretty cool to see the numbers go up and down throughout the day. It clearly showed the rush hours, the slow periods, everything.

This whole thing was more of a fun side project than anything else, but it was really satisfying to see it come together. I mean, I started with some cheap cameras and ended up with a working foot traffic tracking system. It’s not perfect, of course, but it’s a start. Who knows, maybe I’ll keep improving it. It’s amazing what you can do with a little bit of tinkering and a lot of patience.