Patterns (May 2022)
A fine-grained network for human identification using panoramic dental images
Abstract
Summary: When accidents occur, panoramic dental images play a significant role in identifying unknown bodies. In recent years, deep neural networks have been applied to address this task. However, while tooth contours are significant in classical methods, few studies using deep learning methods devise an architecture specifically to introduce tooth contours into their models. Since fine-grained image identification aims to distinguish subordinate categories by specific parts, we devise a fine-grained human identification model that leverages the distribution of tooth masks to distinguish different individuals with local and subtle differences in their teeth. First, a bilateral branched architecture is designed, of which one branch was designed as the image feature extractor, while the other was the mask feature extractor. In this step, the mask feature interacts with the extracted image feature to perform elementwise reweighting. Additionally, an improved attention mechanism was used to make our model concentrate more on informative positions. Furthermore, we improved the ArcFace loss by adding a learnable parameter to increase the loss of those hard samples, thereby exploiting the potential of our loss function. Our model was tested on a large dataset consisting of 23,715 panoramic X-ray dental images with tooth masks from 10,113 patients, achieving an average rank-1 accuracy of 88.62% and rank-10 accuracy of 96.16%. The bigger picture: DNA, fingerprints, faces, etc. have been used in human identification, but they are susceptible to decay when people die. Teeth do not decay, so experts use teeth as an effective feature in individual identification. In earlier times, experts did the comparation manually. Our model contains a branch devised specially to extract tooth contour features, which have proved to be meaningful in previous methods. With other improvements added, our model is able to identify the target person in 1,000 X-ray dental images with an accuracy of 88.62. There also exist limitations. The proposed model rests on masks, so in subsequent studies, we will perform unsupervised methods on teeth or other structures.Compared with DNA, panoramic dental X-ray images are easier to access, so our model provides a feasible approach for identifying unknown bodies if they took panoramic dental X-ray images when alive, even if these bodies are ossified.
