Forensic Retrieval of Striations on Fired Bullets by using 3D Geometric Data

Currently, optical devices, such as microscopes and CCD cameras, are mainly utilized for identification of bullets and tool marks in the field of forensic science. While these optical methods are easily manageable and effective, they are under great influence of illumination condition. Besides these appearance-based approaches, we can utilize 3D geometric data of tool marks that are free from lighting condition. Nevertheless, a perfect correspondence of two striation patterns is rarely encountered, even if the two bullets have been fired from the same firearm. Therefore, more robust methods are required. In this study, we propose a two-stage comparison method focused on 3D geometric. At first, we have aligned global shapes and evaluated a global shape similarity. Then small elevations are compared by neural networks; that is local matching. In this stage, rendered images under a unified lighting condition are utilized. By using 2-stage comparison, we have developed a robust method that searches for similar striation patterns.