Automatic Vehicle Recognition Using Morphological And Template Matching Technique

Samsudin, Khairul Ikhwan (2017) Automatic Vehicle Recognition Using Morphological And Template Matching Technique. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text Automatic Vehicle Recognition Using Morphological And Template Matching Technique.pdf Download (426kB)

Abstract

This document is about an automatic vehicle recognition using morphological process and template matching technique.This project will design a system from a MATLAB software on automatic number plate recognition.This project have 3 main objectives which is to develop a vehicle plate number recognition by using morphological and template matching technique,to apply the programming knowledge by using the MATLAB software and Simulink in order to compute the project and lastly is to analyze whether the performance and the accuracy of the technique used can efficiently recognize the vehicle plate number. In order to enhance the image quality for future process, the morphological process was used to make the input image clearer and sharper and the process that are involved are dilation and erosion and the process are based on the shapes.The process are done by applying the structuring elements into an input image and the process will compare each pixel value with its neighbour in order to produce output image.Dilation will adds the pixels to the boundaries of objects in an image while erosion will removes the pixels on object boundaries.The input image must be in good condition in order to being processed in the future process which is segmentation and recognition of the character process.To recognize the plate number, this document will use the template matching technique where it will compare the segmented characters with the character in the database and character that have the most similarity level will be choose as an output.As a result from this research,from a 30 samples that have been tested,18 plate number responded successfully and give the percentage of 60% of accuracy for the system.To be conclude,the system was being successful developed in this project despite some problems that lead to inefficiency of the system to recongnize some kinds of plate number.

Actions (login required)

View Item