Car Body Damage Detection System Using YOLOv7
Muhammad Remzy Syah Ramazhan, Alhadi Bustamam, Rinaldi Anwar · 2023
Car damage inspection is an important step when submitting car insurance claims. Currently, insurance companies manually gather car damage assessment reports. However, this approach takes a lot of time and is susceptible to fraud. Advances in data science and artificial intelligence offer solutions for automated systems for car damage detection. This system can detect and classify different types of car damage. Implementation of this system can reduce operational costs, save time, and minimize leakage. In this study, we proposed an object detection algorithm that is YOLOv7 to automate car damage detection. Three variations from the YOLOv7 algorithm will be used in this paper. There are YOLOv7-tiny, YOLOv7, and YOLOv7x. Our experiment shows that YOLOv7x achieved the greatest performance with a validation F1-score of 0.949 for car damage detection.