Automated Helmet Detection

Author(s):

Sandra Rose Antony , St.Josephs College of Engineering and Technology, Palai; Sonia Maria Siby, St.Josephs College of Engineering and Technology, Palai; Sajol Sunny Vettom, St.Josephs College of Engineering and Technology, Palai; Sreelakshmi Raju, St.Josephs College of Engineering and Technology, Palai; Dr.Giby Jose, St.Josephs College of Engineering and Technology, Palai

Keywords:

Image Processing, Faster R-CNN, YOLOv3, Pattern Recognition, Helmet Detection

Abstract:

In the last couple of years, most deaths in motorcycle accidents were due to damage to the head of the motorcycle riders. Helmets are being utilized to minimize such damages to the head. Lately, the Government has enforced strict penalties on motorcycle riders who don’t wear helmets while driving. But many violations are still being pursued. To overcome these violations, our team proposes a project to detect motorcycle riders without helmets. The proposed approach first detects bike riders from surveillance video using background subtraction and object segmentation. Then it determines whether the bike-rider is using a helmet or not, attains the number plate of the motorcycle and sends a message stating the fine to the defaulter using a mail. Image processing is used here and processes live video streams from traffic inspection recordings. We propose to use Python programming to implement this project.

Other Details:

Manuscript Id	:	IJSTEV7I1021
Published in	:	Volume : 7, Issue : 1
Publication Date	:	01/08/2020
Page(s)	:	44-48

Download Article