KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Author: search.filters.author.Thanapong, Chaichana

Search Results

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Automatic segmentation of blood vessels in retinal image based on fuzzy K-median clustering
    (2007-12-01)
    Supot, Sookpotharom
    ;
    Thanapong, Chaichana
    ;
    Chuchart, Pintavirooj
    ;
    Manas, Sangworasil
    This paper presents an efficient method for automatic segmentation of blood vessels in retinal images. Specifically, we also delineate vascular intersections/crossovers. The proposed algorithm is composed of three steps: matched filter, fuzzy k-median (FKMED), and length filter. The segmentation results are compared with clinically generated vessel segmentation and are evaluated in terms of sensitivity and specificity. The results are encouraging and will be used for further application such as personal identification. © 2007 IEEE.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Segmentation of magnetic resonance images using discrete curve evolution and fuzzy clustering
    (2007-12-01)
    Supot, Sookpotharom
    ;
    Thanapong, Chaichana
    ;
    Chuchart, Pintavirooj
    ;
    Manas, Sangworasil
    The region clustering of a Magnetic Resonance Imaging (MRI) image is more complicate than a Computed Topography (CT) image because a MRI image composes of three components such as T1-weighted, T2-weighted, and Proton Density (PD) in each layer. However, the MRI images provide more detail than the CT images. Therefore, we propose a technique of the region clustering of MRI image by using Fuzzy c-means (FCM). The fuzzy c-means algorithm is an iterative operation, that is very time-consuming and makes the algorithm impractical for using in image segmentation. To cope with this problem, the discrete curve evolution (DCE) technique is applied to find the actual cluster center to refine the initial value of the fuzzy c-means algorithm, which reduces the convergence time. In experimental results, the proposed technique provides the same segmentation accuracy as the fuzzy c-means technique. Moreover, this technique takes lower computational time comparing to the previous method. © 2007 IEEE.