A Simple and Novel CBIR Technique for Features Extraction Using AM Dental Radiographs

Dental image processing is most immerging field for human identification. Dental features remain more or less invariant over time compared to other identification clues like fingerprint, iris, etc. which are not available in some case of major accidents. The purpose of dental image processing is to match the post-mortem (PM) radiograph with the ante mortem (AM) radiograph based on some characteristic or feature of the radiograph for human identification. Image enhancement is necessary because of poor quality and low contrast of dental image at primary stage. Thereafter segmentation algorithms are applied to the enhanced dental x-ray image which helps to find two major regions namely gap valley and tooth isolation. The main crucial part is tooth and feature extraction of dental image. In this paper, we propose a simple and novel CBIR technique to extract individual tooth and thereafter we extract geometrical features of dental x-ray radiographs for human identification purpose. We compared feature vectors of database with query image and calculated the distance vector for matching purpose.