1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Medical image segmentation using optimized OTSU thresholding method
Authors:	Ma Jie (P51688)
Supervisor:	Md. Jan Nordin, Assoc. Prof. Dr.
Keywords:	Image segmentation Diagnostic imaging -- Data processing Computer vision in medicine Image processing -- Mathematics Genetic algorithms Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia
Issue Date:	16-May-2012
Description:	Medical image is a crucial field and develop rapidly. With the help of medical image processing, it really convenient to used to create images of the human body for clinical purposes. Image segmentation is the technique that divides the image into different regions which have different characteristics or extracts the interested objects from the image. Therefore, application of image segmentation in biomedical is a good method. Medical image segmentation technique is to delineate anatomic structures or other interested regions automatically or semi-automatically, which is helpful to diagnosis and plays a crucial role in many medical imaging applications. Segmentation of medical images is challenging due to complicated or low quality image that result in missing or vague organ/tissue boundaries. The purpose of this research is to use a threshold-based method to get a key value for boundary detection. Genetic algorithm (GA) is a search heuristic that mimics the process of natural evolution. With the help of genetic algorithm, the optimum threshold value can be found. This heuristic is routinely used to generate useful solution to optimization. Finally, the experiments used dataset based on a specific medical format image (DICOM) and were conducted to evaluate and compare the performance of the proposed threshold algorithm using GA with other two kinds of different threshold algorithms; global threshold algorithm and OTSU algorithm. In each comparative experiment the effect of one of the medical image difficulties: target object segmentation accuracy and time complexity. The obtained results verify the superiority of threshold method using genetic algorithm, whereas these method show robustness against all types of tested medical image. Furthermore, the proposed segmentation method was very effective to find an optimum threshold for target object extraction and segmentation. Based on the comparative experiments results, it is observed that the proposed approach segmentation method outperforms the other threshold-based segmentation method.,“Certification of Master’s / Doctoral Thesis” is not available,Master of Information Technology,“Certification of Master’s/Doctoral Thesis” is not available
Pages:	107
Call Number:	TA1638.4.M336 2012 3 tesis
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.