Normal view MARC view ISBD view

Object detection system using haar-classifier / Wan Najwa Binti Wan Ismail

By:

Wan Najwa Wan Ismail

Material type: Text

TextPublication details: Kuantan, Pahang : UMP, 2009Description: 59 p. : ill. (some col.) ; 30 cm. + 1 computer discISBN:

THE0006944(Local)

Other title:

Object detection system using haar-classifier [electronic resource]

Subject(s):

Detectors

Dissertation note: Project paper (Bachelor of Electrical Engineering (Electronics)) -- Universiti Malaysia Pahang - 2009 Abstract: The invention of new algorithms had encouraged to the reinforcement of image processing’s application. An algorithm for the design object detection systems is presented. Haar-classifier is utilized as the algorithms for this object detection system. The exertion of Haar-Classifier had boosted to the upgrade system which is faster and more accurate. In this system, Haar-Classifier is conjunct with the Adaboost machine learning algorithms wherefore the performance of the system is upgraded. Development of this project is categorized into two phase which are training phase and execution phase. Training phase use OpenCV utilities such as haartraining.exe to train the object by calculating the object’s weak constraints. This is for the purpose of finding the different features of the object of interest. The list of these weak constraints is converted to the xml file to be included in the coding which had been developed using Visual Studio 2005. The execution process will result on the detection process of object of interest. System will detect rounded image in any image which had been included in the system itself. Object detection system using Haar-classifier algorithm can perform best performance of high detection rate and high level of accuracy rate.

Tags from this library: No tags from this library for this title. Log in to add tags.

Average rating: 0.0 (0 votes)

Holdings ( 2 )
Title notes ( 2 )

Holdings
Item type	Current library	Call number	Copy number	Status	Date due	Barcode
Final Year Report	UMPLIB PEKAN	TK7872.D48 N35 2009 rs Bc. (Browse shelf(Opens below))	1	Not for loan		0000037856
Final Year Report	UMPLIB PEKAN	CD 3303 \| TK7872.D48 N35 2009 rs Bc. (Browse shelf(Opens below))	1	Not for loan		0000037857

Project paper (Bachelor of Electrical Engineering (Electronics)) -- Universiti Malaysia Pahang - 2009

The invention of new algorithms had encouraged to the reinforcement of image processing’s application. An algorithm for the design object detection systems is presented. Haar-classifier is utilized as the algorithms for this object detection system. The exertion of Haar-Classifier had boosted to the upgrade system which is faster and more accurate. In this system, Haar-Classifier is conjunct with the Adaboost machine learning algorithms wherefore the performance of the system is upgraded. Development of this project is categorized into two phase which are training phase and execution phase. Training phase use OpenCV utilities such as haartraining.exe to train the object by calculating the object’s weak constraints. This is for the purpose of finding the different features of the object of interest. The list of these weak constraints is converted to the xml file to be included in the coding which had been developed using Visual Studio 2005. The execution process will result on the detection process of object of interest. System will detect rounded image in any image which had been included in the system itself. Object detection system using Haar-classifier algorithm can perform best performance of high detection rate and high level of accuracy rate.