Vous êtes sur la page 1sur 3

Final Year Project Proposal

Government College University, Lahore

Department of Electrical Engineering

Real-Time Traffic Sign Detection and Recognition for Intelligent Vehicles Using Image
Processing

Project Supervisor: Asma Mushtaq

Muhammad Ramiz Amer Ch. 0025-ENGG-2014

Jahanzeb Ahmad Khan 0035-ENGG-2014

Ali Murtaza Zaidi 0038-ENGG-2014

Saad Bin Ayaz 0042-ENGG-2014


ABSTRACT
Automatic traffic sign detection and recognition systems (TSDRs) are the
centre of attention and great interest for engineers and researchers since few
years. Traffic road sign detection and recognition is important to transport
system. The aim of the project is to detect and recognize traffic signs in video
sequences recorded by an onboard camera in a vehicle. Traffic Sign
Detection & Recognition (TSDR) is used to regulate traffic, warn a driver,
and command or prohibit certain actions. A real-time and robust automatic
traffic sign detection and recognition can support and disburden the driver to
a considerable extent and potentially increases driving safety and comfort.
Automatic recognition of traffic signs is also important for automated
intelligent driving vehicle or driver assistance systems technique. The task in
this project is to develop a prototype of an autonomous vehicle which is
capable of detection and recognition of the traffic signs in real-time video
through the camera that is mounted on top of the car. Although the system
comprises of various stages and techniques however, primarily there are two
main stages, detection and recognition. In the detection stage, our main goal
would be to segment the traffic signs through color based segmentation, from
the real-world environment while in recognition our focus would be to
analyze the contents that were obtained from the sign or a signal.

Fig 1. Block Diagram


Methodology
Road sign detection basically consists of two major stages:

 Detection
 Recognition

In the detection phase, the camera that is mounted on the vehicle captures
real-time video of the traffic sign located on the road ahead. The video is then
processed using the image processing techniques which is another important
aspect of this project. The Images that will be captured in RGB (Red, Green,
Blue) color space will be transformed into different color spaces (HSV, YCbCr,
YIQ or CIELab). This transformation is applied because RGB color space is
highly sensitive to light and illumination changes. If the pixel values do not lie
within the threshold ranges, then those pixels are considered as a part of
background image and next image is selected from training data for
segmentation. Furthermore, each captured image is compared with the certain
set of images present in the database to evaluate and validate the traffic sign.
Based on the recognition results, the movement of the vehicle is controlled.

Primary Equipment Required:


 Robotic Vehicle Frame
 Raspberry Pi and OS
 Microcontroller
 Power Pack (5 V)
 Pi Camera Module

Vous aimerez peut-être aussi