Citation

Al-abayechi, Alaa Ahmed Abbas (2015) Enhancement of Dermoscopic Images and Feature Extraction for Classification of Skin Lesions. PhD thesis, Multimedia University.

Full text not available from this repository.

Abstract

Early detection of skin lesions, such as melanomas, permits early treatment, hindering their spread throughout the body. It will lead to a decrease in the number of the new melanoma cases presented every year. Since most dermatology centres lack expensive equipment and specialists, there is a need to develop a highly trustworthy computer-aided system that is able to reliably conduct preliminary early diagnosis. The aim of this study is to use the available common equipment for imaging techniques such as a digital camera or mobile phone camera, instead of the expensive tools used nowadays, to increase the earliest skin lesion diagnosis accuracy, especially in early stage detection that led to reduce the number of deaths. Thus, this work is established to help dermatologists detect melanoma with minimum cost. The proposed system consists of 5 main stages: pre-processing, segmentation, post-processing, feature extraction, and classification. Each main stage is further subdivided. For example, pre-processing involves resizing images (the images used in this work are of different sizes), which is followed by noise removal within the lesion region. Lastly is the image enhancement technique to reduce artefacts and obtain a smooth image using different proposed algorithms, which are essential for improving accuracy in the next step. The various approaches of next stage (segmentation) aim to detect the location of the skin lesion region and refining the region that is segmented in post-processing step using different morphological operations to create binary image.

Item Type:	Thesis (PhD)
Additional Information:	Call No.: RC78.7.D53 A43 2015
Uncontrolled Keywords:	Diagnostic imaging
Subjects:	R Medicine > RC Internal medicine > RC71-78.7 Examination. Diagnosis
Divisions:	Faculty of Engineering (FOE)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	23 Sep 2016 07:30
Last Modified:	23 Sep 2016 07:30
URII:	http://shdl.mmu.edu.my/id/eprint/6368

Downloads

Edit (login required)