अमूर्त

Contourlet transform based efficient shape extraction technique for forensic odontology

Jaffino G, Banumathi A, Ulaganathan G, Vijayakumari B

Owing to the evaluation of information technology and an urge to investigate more cases by forensic experts, it is necessary to automate the victim identification system. For designing an automated dental identification system, dental shape extraction is a prime process. This work explains the contour based shape extraction approach applied for both ante mortem and post mortem dental images using Contourlet Transform (CT). For individual identification, contour information has been proven as a useful measure. In automatic dental identification system, contour extraction is a challenging one and the results obtained proved that contourlet transform based shape extraction techniques give a better shape extraction compared to existing fast connected component approach. In order to extract the more accurate contour, double filter bank is used in Contourlet transform. The Euclidean distance based matching was performed to compare both ante mortem and post mortem dental images. It is observed that 98% of similarity can be obtained by using contourlet transform.