As a team, we are continuing to work on our individual subsystems and we are making sufficient progress. Currently, the most significant challenge we are facing is the lack of documentation on the eYs3D camera we are using, which could make it more difficult to integrate it with the Raspberry Pi and the ML models. Furthermore, the addition of OCR means that more time needs to be allocated for the ML component of the device.
Opalina’s Status Report 3/8
Over the last two weeks, I began training YOLOv8 on one of the online airport datasets. I also realized the need for Optical Character Recognition (to interpret words and numbers in addition to arrows) and delved into ways to implement and integrate it into the software subsystem. By next week, I hope to have a functional YOLO model for our purposes and robust implementation plans for OpenCV and OCR.