Oral Tumor Segmentation and Detection using Clustering and Morphological Process

Abstract

Oral tumor is one of the most widely recognized tumors growing globally, continuously promoting a high mortality rate. Because early detection and treatment remain the most effective interventions in improving oral cancer outcomes, developing complementary vision-based technologies that can reveal potential evil high-quality oral diseases (OPMDs), which carry the risk of developing cancer, represent significant opportunities for the oral screening process. This paper proposes a morphological algorithm to preserve edge details and prominent features in dental radiographs. This technique, in the early stage identifies the oral tumor detection using clustering and morphological processing. This algorithm would allow for the identification of tumors in these images. Applying pre-processing in images leads to over-segmentation even though it is pre-processed.