Forensic identification to victims of criminal acts or mass deaths due to disasters to determine the identity of victims using fingerprints, DNA, and teeth is very important for various reasons. In some cases, such as a fire disaster, DNA and fingerprints cannot be used. Identification using teeth is one of the alternatives that can be used because it is the most preserved part and the position, structure and shape of teeth are unique. Basic principle of dental identification is comparison of antemortem (AM) and postmortem (PM) dental images. For the current work only AM data is available, so it is necessary to manipulate dental x-ray images on AM data to produce PM image data using three methods, those are gamma correction, image rotation and image distortion using integral projection. From both datasets (AM and PM), important image features are extracted through representation learning by PCA, transfer learning using ResNet50's architecture methods. The results of representation learning namely the AM and PM datasets are compared using cosine similarity, which will give output the values of similarity each images. The best results of dental image identification have the best accuracy value using the transfer learning -ResNet50 method on all types of PM datasets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.