A New Feature Extraction Method from Dental X-ray Images for Human Identification

Using dental radiography is an alternativeapproach to identify a deceased person, especially in cases thatother biometric traits cannot be handled. This paper proposes anew method for feature extraction from dental radiographyimages to identify people. First, dental works are segmented inthe X-ray images using image processing techniques. Then,radius vector function and support function are extracted foreach segmented region. These functions are independent of imagetranslation. The presented algorithm modifies both functions tobe invariant under image rotation as well. Also, by normalizingthe functions, the problems due to image scale variations can besolved. Image translation, rotation and scale variations are basicchallenges when dental features are compared in spatial domain.Experiments prove suitable recognition accuracy in the proposedapproach which does not require teeth alignment at the matchinglevel

کلیدواژه ها:

Human identification ، biometrics ، dental X-ray images ، radius vector function ، support function