Improve Tire Pattern Imprint Image Recognition with YOLO

Many car accidents around the world cause damage to both life and property. The incidents occur where the perpetrators cannot be tracked. Although there are efforts to find evidence from the scene of the accident, especially tire pattern testing to find the perpetrator's car, this is considered important evidence to help find the perpetrator. In addition, the development of current automobiles that need to respond to driving efficiency has resulted in the development of tire production with a large number of different tire patterns. From the researcher's perspective, tire pattern analysis at the scene of the accident is still a challenging problem. In order to provide useful clues in the case of road accidents and improve the accuracy of tire pattern recognition by using Discrete Wavelet Transform and Local Binary Pattern, we propose that YOLO train tire pattern recognition for recognition by comparing it with SVM, YOLOv5, YOLOv8, and YOLOv9.