A Novel Segmentation Method of Cervical Cancer affected Surface Tissues based on Superpixels

Abstract

Abstract

Cervical Cancer is one of the pervasive forms of cancer afflicting the female population worldwide. A Digital Colposcope is a self-illuminated powerful microscope which acquires the image of the affected cervix. The raw cervix image acquired by the colposcope is known as a cervigram. The raw cervigram is preprocessed by removing the specular reflections and then the region of interest is sought. Different novel segmentation algorithms which were proposed in our ongoing research work are applied to the cervigram to detect the tissues, sometimes with erroneous segmentation results. In this paper, we try to overcome those weaknesses by adding excess information into the segmentation process. This excess information relates to the local continuity of pixel features within the image plane. It shifts from a probabilistic pixel-based clustering scheme to clustering of small coherent regions in the image plane termed superpixels. Our proposed algorithm takes a desired number of approximately equally-sized superpixels as input, and by Mahalonobis distance metric, cluster analysis and classification.

Keywords: Cervical cancer, colposcopy, segmentation, superpixels, mahalonobis distance

Cite this Article

Abhishek Das. A Novel Segmentation Method of Cervical Cancer affected Surface Tissues based on Super-pixels. Journal of Image Processing & Pattern Recognition Progress. 2017; 4(3): 1–7p.