A Sequential K-Means Clustering For Mammogram Segmentation
K-means is the simple, efficient clustering technique from past 50 years. Application of k-means is an important
characteristic in many applications including medical image segmentation. One of the drawbacks in k-means algorithm is it
does not use the spatial information of image space in the clustering process. This paper proposes a simple algorithm that
combine intensity and texture based clustering in sequence using k-means to determine regions of interest in mammograms.
Experiments are conducted on each image of MIAS database. The results demonstrated the accuracy and efficiency of the
algorithm in identifying the masses of mammograms.
Index Terms- -means, intensity, Region of interest, segmentation, texture.