Please use this identifier to cite or link to this item: https://www.um.edu.mt/library/oar/handle/123456789/94123
Title: Implementation of an on-road vehicle detection system
Authors: Sammut, Josef (2013)
Keywords: Algorithms
MATLAB
Vehicle detectors
Traffic flow -- Measurement -- Equipment and supplies
Issue Date: 2013
Citation: Sammut, J. (2013). Implementation of an on-road vehicle detection system (Bachelor’s dissertation).
Abstract: On-road vehicle detection has become one of the most important researches in vision-based computer applications for various reasons, mainly to reduce traffic congestion and accidents. Vehicle detection technologies are constantly being developed in order to provide better traffic surveillance with the aim of improving traffic flow and increase on-road safety. The development of intelligent video surveillance is replacing the traditional surveillance system which required a large amount of manpower and labour in order to monitor any particular street. This dissertation will put into effect a vehicle detection system which is able to detect vehicles from a medium density traffic scene. The proposed vehicle detection system is based on the Block Matching Algorithm. This algorithm is used to find motion estimation from video frames. Its objective is to divide video frames into fixed sized block, called macroblocks and find the displacement of each macro block in the current frame when compared to a reference frame. In order to determine the best matching macroblock a cost function is used to compare the macroblock with all the other macroblocks in the search window. The minimum cost generated by this function indicates that a matching macro block in frame i+ 1 was found for frame i. The displacement, if any, of a macro block from frame i to frame i+ I is represented with a motion vector. This motion vector represents the change in an image from one frame to the next. The on-road vehicle detection system was first prototyped using the MATLAB environment, which is specialized software suitable for image and video processing applications. The prototyped software was then implemented on a DSP board where the aim was to be able to process vehicle detection in real time.
Description: B.Sc. IT (Hons)(Melit.)
URI: https://www.um.edu.mt/library/oar/handle/123456789/94123
Appears in Collections:Dissertations - FacICT - 2013
Dissertations - FacICTCCE - 1999-2013

Files in This Item:
File Description SizeFormat 
B.SC.(HONS)ICT_Sammut_Josef_2013.PDF
  Restricted Access
7.7 MBAdobe PDFView/Open Request a copy


Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.