A systematic review of CNN based model approaches for human identification using bitemark evidence

A total of 22 recent three-year papers were examined in this paper to examine the application of CNN models to digital images. In forensic dentistry, the use of bitemark evidence for human identity is nothing new. Convolutional Neural Networks (CNNs) have been used for the examination of bitemark evidence thanks to recent developments in machine learning. An abstract of a project whose objective is to create a CNN model for automatically identifying human bitemarks based on dental traits is presented in this publication. The suggested model optimises on a dataset of dental pictures using a pre-trained CNN architecture. A variety of bitemarks with various patterns, shapes, and sizes make up the dataset. The model is trained to recognise particular dental characteristics like tooth size, orientation, and spacing to precisely match bitemarks to the offender. Accuracy, precision, recall, and F1-score measures are used to assess the performance of the suggested model. The findings demonstrate that the suggested CNN model can distinguish human bitemarks with good precision and recall rates. The suggested model may enhance forensic dentistry's precision and effectiveness when identifying people based on their bitemarks.