Lane detection algorithm based on open source computer vision

Date of Publication

2009

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Electronics and Communications Engineering

College

Gokongwei College of Engineering

Department/Unit

Electronics and Communications Engineering

Thesis Adviser

Edwin Sybingco

Defense Panel Chair

Enrique M. Manzano

Defense Panel Member

Aaron Don M. Africa
Elmer Jose P. Dadios

Abstract/Summary

Every year, there is a significant increase in the number of road accidents on both cars and trucks alike. Such accidents may have been triggered by the human error of drivers, poor driving conditions, or the mechanical failure of the vehicle's components. One way of decreasing the occurrence of such accidents is by lessening, if not eliminating, the cause of human errors while driving. In this paper, a machine vision system is introduced, which enables the computer to detect lanes using an algorithm based on the Open Source Computer Vision libraries. The system produced, compared to those out in the market, is developed using low cost materials, gadgets, and software that are freely available in the market. It is capable of detecting and tracing the different kinds of road lanes such as straight, broken, and continuous. Also, it is capable of producing a robust and accurate detection of the said lanes in various road conditions such as daytime, night time, rainy, and overcast weather. The result of the said detection is displayed to the user using a graphical user interface developed using the same low cost programs and software. The accuracy of the system is then gauged by the discrepancy of the projected lane of the algorithm, with respect to the original lane to be detected. A higher accuracy is then given to the system if the disparity of the projected and the original lane is within the bounds of an acceptable margin.

With the development of such a system, the authors wish for the eventual automation of automobiles with the use of the developed low cost lane detection system.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU14868

Shelf Location

Archives, The Learning Commons, 12F, Henry Sy Sr. Hall

Physical Description

201 leaves : ill. (some col.) ; 28 cm.

Keywords

Algorithms; Image processing; Computer vision; Image analysis; Automobiles--Automatic control

This document is currently not available here.

Share

COinS