Visual driver support system for traffic signs

It has been well documented that traffic accidents that can be avoided occur when the motorists miss or ignore traffic signs. With the attention of drivers getting diverted due to distractions like cell phone conversations, missing traffic signs has become more prevalent. Also, poor weather and other unfriendly driving conditions sometimes makes the motorists not to be alert all the time and see every traffic sign on the road. Besides, most cars do not have any form of traffic assistance. Because of heavy traffic and proliferation of traffic signs on the roads, there is a need for a system that assists the driver not to miss a traffic sign to reduce the probability of an accident. Since visual information is critical for driving, processed video signals from cameras have been chosen to assist drivers. These inexpensive cameras can be easily mounted on the automobile. The objective of the present investigation and the traffic system development is to recognize the traffic signs electronically and alert drivers. For the case study and the system development, five important and critical traffic signs have been selected. They are: STOP, NO ENTER, NO RIGHT TURN, NO LEFT TURN, and YIELD. The system was evaluated processing still pictures taken from the public roads, and the recognition results were presented in an analysis table to indicate the correct identifications and the false ones. The system reached the acceptable recognition rate of 80% for all five traffic signs. The processing rate was about three seconds. The capabilities of MATLAB, VLSI design platforms and coding have been used to generate a visual warning to complement the visual driver support system with a Field Programmable Gate Array (FPGA) on a XUP Virtex-II Pro Development System.