(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Image segmentation using firefly algorithm

Image segmentation using firefly algorithm / Ong Chin Loong
Kanser payu dara adalah salah satu punca kematian kanser di kalangan wanita. Teknik pengesanan awal adalah penting untuk mengurangkan kadar kematian pesakit. MRI adalah salah satu teknik saringan untuk mengesan tumor. Segmentasi kawasan tumor payu dara yang disyaki dalam imej MRI telah menjadi satu cabaran sejak kewujudan teknologi ini disebabkan penampilan fizikal tumor di dalam imej. Dalam kajian ini, segmentasi tumor payu dara MRI telah diterokai. Algoritma kelip-kelip (FA) adalah salah satu algoritma metaheuristik yang boleh digunakan dalam segmentasi imej. Matlamatnya adalah untuk melaksanakan algoritma kelip-kelip untuk mengesan tumor. Lima keping imej MRI diberikan untuk segmentasi tumor. Pra-pemprosesan seperti penapis median dan penyingkiran pinggiran payu dara digunakan untuk meningkatkan kualiti imej. Operasi morfologi digunakan untuk membuangkan titik kecil putih di dalam imej MRI. Keputusan segmentasi tumour dibanding dengan keputusan dari pakar radiologi adalah memuaskan. Ini menunjukkan algoritma kelip-kelip boleh dilaksanakan dalam imej MRI untuk segmentasi tumor. _______________________________________________________________________________________________________ Breast cancer is one of the causes of cancer death in women. An early detection technique is important to reduce the death rate of patients. MRI is one of the commonly used screening techniques for detecting tumour region. Segmentation of suspected breast tumour region in MRI image has been a challenge since the existence of this technology due to the physical appearance of tumour in the image. In this study, MRI breast tumour segmentation has been explored using firefly algorithm (FA), one of the metaheuristic algorithms. The goal is to implement the firefly algorithm to segment out the tumour region from suspected regions. Five MRI images are given for the purpose of tumour segmentation. Pre-processing such as median filter and removal of breast borderline are implemented to enhance the quality of images. Morphological operation is implemented to remove the small white dots in MRI images. The comparisons of final results of MRI images after tumour segmentation with benchmark result of radiologist are satisfied. This shows that firefly algorithm can be implemented in the breast images for tumour segmentation.
Contributor(s):
Ong Chin Loong - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875004751
Language:
English
Subject Keywords:
Breast; cancer; (FA)
First presented to the public:
6/1/2012
Original Publication Date:
7/2/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 52
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-07-02 16:15:44.431
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Image segmentation using firefly algorithm1 2020-07-02 16:15:44.431