I will be talking about creating a computer vision classroom color poll sheet recognition project using Python, OpenCV and raspberry Pi. Specifically, looking at detecting color poll-sheets in a classroom of about 50 - 150 students. Much like how one can poll an audience of people using a clicker technology and show the results of the audience with regards to a question asked.
Audience is anyone interested in computer vision, what computer vision can do and achieve, as well as the hard work required to make it all happen.
After the talk the audience can have a glimpse into what it takes to create a computer vision project, as well as the typical process and algorithm one would follow in a simple task of creating a computer vision project using python, opencv and raspberry pi.
A few aspects I will cover include: - Computer vision goals - Simplified overview of the algorithm used - Going into a bit of detail where the challenges arise - Example of the the project i worked on - Results achieved - A few global interesting projects around computer vision
I will end with a look into the python code involved and talk through the results of the polling program, as well as any hard to detect areas and how the system detects these. :)