Reading Fuel Information Boards at Gas Stations with Artificial Intelligence

There have been many studies in the field of image processing from past to present. Studies in the field of image processing are quite useful in cases where a system is automated and the human eye cannot follow. These studies use image processing techniques in order to ensure control during the production phase or in order to ensure correct data flow during the inspection phase. Our work falls within the scope of the inspection phase. Within the scope of our work, it is aimed to transfer the information on the fuel information panels at gas stations to a central system with image processing techniques. In this way, the prices reflected on the panels at gas stations will be controlled. Within the scope of this project, nearby gas station panel images were collected. Labeling was performed for each feature in the collected data set. The Yolov8 model is used for object detection.