Feature extraction in dental radiographs in human extracted and permanent dentition

Abstract

Feature extraction in dental images in the form of radiographs involves the identification of major defect areas. While analyzing complex radiograph images, one of the major problems stems from the types of defects present. Analysis with a large number of defects present generally requires a large amount of memory and computational power. Feature extraction applied over the radiographs, once the edge detection process is accomplished, derives combinations of the defects to get around the problems while still describing the problem areas with sufficient accuracy. The process has been implemented over a set of 20 of extracted human dentition for the identification of similar features to actualize the presence of defects in the dentition.