Automatic Detection of License Number Plate of Motorcyclists Without Helmet

Authors

  • Nikita Saklani

Keywords:

Helmet detection, Daubechies 8 wavelet transform, (SVM) Support Vector Machine, Template Matching.

Abstract

Nowadays Vehicles are at a reasonable price and it can economically give by every person this is the reason there is rapid growth in the rate of accidents since most of the motorcyclist does not wear a helmet which has made it an ever-present danger situation to travel by motorcycle. From this year in 2019, the Government has made it a punishable offence to ride a motorcycle without a helmet. The main cause of death is due to the injury caused to the head region of the motorcyclist. According to section 129 of motorcycle vehicle act, Government has made it mandatory for two-wheeler driver to wear a helmet while driving but many of the traffic rule violators do not obey them. So, it is very important to take prompt and strict action against these violators. This project presents a smart surveillance system for automatic detection of two-wheeler driver without a helmet and traces the license number plate of the motorcycle. For detecting motorcyclist with and without helmet we have used Daubechies 8 wavelets transform to extract features of the upper part of the segmented image i.e. motorcycle and these features were fed as an input to the SVM (Support Vector Machine) to train the classifier based upon the features derived. For vehicle number plate detection, we have used image- based template matching. To set up our purpose we have created a database of l images of a motorcyclist with and without helmet taken from different angles. The experimental results show that the system efficiency for classification of helmet and non-helmet is 95%.

 

Downloads

Published

2019-09-06

How to Cite

Nikita Saklani. (2019). Automatic Detection of License Number Plate of Motorcyclists Without Helmet. Journal of Science & Technology (JST), 4(5), 24–28. Retrieved from https://jst.org.in/index.php/pub/article/view/78