A Novel
Approach for Segmentation of Human Metaphase Chromosome Images Using Region
Based Active Contours
Tanvi
Arora
Department of Computer Science and
Engineering, Dr. B.R Ambedkar National Institute of Technology, India
Abstract:
The
chromosomes are the genetic information carries. A healthy human being has 46
chromosomes. Any alteration in either the number of chromosomes or the
structure of chromosomes in a human being is diagnosed as a genetic defect. To
uncover the genetic defects the metaphase chromosomes are imaged and analyzed. The
metaphase chromosome images often contain intensity inhomogeneity that makes
the image segmentation task difficult. The difficulties caused by intensity inhomogeneity
can be resolved by using region based active contours techniques. These techniques
uses the local intensity values of the nearby regions of the objects and find
the approximate intensity values along both sides of the contour. In the
proposed work a segmentation technique has been proposed to segment the objects
present in the human metaphase chromosome images using region based active
contours. The proposed technique has been quite efficient from prospective of
number of objects segmented. The method has been tested on Advanced Digital
Imaging Research (ADIR) dataset. The experimental results have shown quite good
performance.
Keywords:
Chromosomes,
segmentation, active Contours, intensity inhomogeneity.