Detection of License Plate Numbers and Identification of Non-Helmet Riders using Yolo v2 and OCR Method

At the present circumstances, there are a variety of traffic regulation issues in India, which can be addressed using various approaches. Riding a motorbike or scooter without a helmet is a traffic offence that has increased the number of accidents and deaths. The current system largely monitors traffic offences through CCTV records, in which traffic cops must zoom into the frame where the traffic violation is occurring and look at the license plate on the off chance that the rider isn't wearing a helmet. However, because traffic offences are common and the number of persons riding motorcycles is growing day by day, this will take a lot of labour and time. Hence, this paper proposes a device that uses CNN to identify bike drivers who are not wearing helmets. Bicycle distinguishing proof, helmet versus no head defender, and bicycle label affirmation are all featuring parts of the system. The bikes are filtered using the HOG component vector. When CNN recognises a cruiser, it checks to see if the rider is wearing a protective headgear. Tesseract OCR is used to recognise the bike's tag if the motorcyclist isn't wearing a helmet. Keywords: Protective cap Detection, Convolutional Neural Network (CNN), Tesseract Optical Persona Cognizance (OCR), License Plate Extraction, Histograms of Oriented Gradients (HOG)