Codes: https://drive.google.com/drive/folders/1-L2Mzz5-u91SKeV4pMdN4XyrjlrUQ2xq?usp=sharing

Aim: To build an object detection model which can detect whether person on the construction site is wearing helment or not.

Model Architecture: I have used YOLOv5 for the task, model will be used on the construction site for the live video frame so need an architecture which can run on higher fps.

Result: I have trained the model for 100 epoch which give the following result:

Model is evaluated on the video : https://www.youtube.com/watch?v=6PoPwZ0WO9w

result video named "output.mp4" can be found in the drive link. https://drive.google.com/file/d/1-8EFQ0I_bI_ixhDv4fYKWJFvTx7_aqe6/view?usp=share_link

We have also evaluated on the 250 test images and the result is "test-img-and-result"