The Design and Implementation of Digital Image Segmentation in HSV Color Space

Segmentation subdivides an image into its constituent parts or
objects. The purpose of segmentation is to simplify or change
from a digital image to the more meaningful and easier to analyze.
Most of the segmentation process is done in RGB representation.
In this paper, the segmentation process is done in HSV
representation. The process consist several steps: changing the
representation from RGB to HSV, search the centroid points using
maximin algorithm, clustering by K-Means, and post processing
of the results of clustering to eliminate noise and unnecessary
detail. This application was created with C #. net with Microsoft
Visual Studio 2005 as its IDE. The result indicates that the
segmentation results are influenced by the centroid points. With
determining the exact number of centroids, it gives clear
segmentation result. Factors that can affect the centroid are
intensity light and color range on the digital image.