Automatic Detection of Bike-riders without Helmet using Surveillance Videos in Real-time

  S.Saranya,  R. Malveka,  S. Ragavi,  N. Gokul Raj,  P. Narayanan

Now-a-days two wheelers is the most preferred mode of transport. It is highly desirable for bike riders to use helmet. In this paper, we propose an approach for automatic detection of bike-riders without helmet using surveillance videos in real time. The proposed approach first detects bike riders from surveillance video. Then it determines whether bike-rider is using a helmet or not using a visual feature. Later convolutional neural network (CNN) is used to select motorcyclists among the moving objects. Again, we apply CNN on upper one fourth portions for further recognition of motorcyclists driving without a helmet. The performance of the proposed approach is evaluated on two datasets. The experiments on the real-time videos successfully detect 96.66% bike-riders without helmet with a low false alarm rate of 0.5%. Hence the bike riders without helmet are detected.

Keywords- Image Classification, OpenCV, Kera model, Convolutional neural network, region proposal networks (RPN), Histogram of Oriented Gradients (HOG).

Unique Identification Number - IJEDR2001113
Page Number(s) - 605-609
Pubished in - Volume 8 | Issue 1 | March 2020
DOI (Digital Object Identifier) -    http://doi.one/10.1729/Journal.23948
Publisher - IJEDR (ISSN - 2321-9939)

  S.Saranya,  R. Malveka,  S. Ragavi,  N. Gokul Raj,  P. Narayanan,   "Automatic Detection of Bike-riders without Helmet using Surveillance Videos in Real-time", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.8, Issue 1, pp.605-609, March 2020, Available at :http://www.ijedr.org/papers/IJEDR2001113.pdf