Traffic Sign Classification for Automated Cars

Traffic Sign Classification for Automated Cars

1 Ms. Pallavi Shejwal, Assistant Professor, Information Technology Engineering Department,

Progressive Education Society’s Modern College of Engineering, Pune - 411 005 (Maharashtra)(India)

2Student VIII Semester B.E(Information Technology),

Aditya Korad1, Kartikeya Bapat2, Harshal Kamble3, Chaitanya Virkar4

Progressive Education Society’s Modern College of Engineering, Pune - 411 005 (Maharashtra)(India)

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Developing an automated driver guidance system is very important in the context of Indian road conditions. Be it a human driver driving a car or a self-driven car, following traffic signs on Indian roads is necessary. A lot of traffic sign violations are seen and heard of, mainly the reasons for which are over speeding, broken traffic signs, distorted signs, night time conditions, etc. In India there are multiple roads where traffic signs do not survive and are mostly broken are covered in trees or any hindrance. Traffic signs are mostly damaged and distorted by the civilians itself which causes more traffic sign violations.

The recent technical development in mobile processors led many automobile manufacturers to deploy computer vision systems into their cars. These systems ensure safety which is a crucial aspect for autonomous vehicles. Traffic sign recognition is one of the most well-known and widely discussed tasks that can be solved by these computer vision systems. It can ensure safety as it allows the vehicle to know what different signs on the road mean and act accordingly. However, such systems come with problems like low accuracy in detection, and some systems are unable to detect traffic signs from different countries.

Key Words: Python Imaging Library (PIL), Deep Learning, Neural Networks.