A Comparative Analysis for Traffic Officer Detection in Autonomous Vehicles using YOLOv3, v5, and v8
Juan P. Ortiz, Juan Diego Valladolid, Denys A. Dutan-Sanchez · 2024
This article focuses on generating an alternative to identify traffic officers during driving. This research employed the You Only Look Once (YOLO) model, using a sixphase methodology: data collection, data preparation involving resizing and labeling, implementation of various filters to avoid overfitting, model training, prediction evaluation, and result interpretation. The YOLO model was applied across three iterations using a dataset of 1862 images. The graphics processing unit (GPU) acceleration was utilized to enhance training efficiency and detection speed, further enhancing the experimental process. The results of this study revealed that the YOLOv8x variant produced the most promising results. This proposed model attained a remarkable F1 score of 0.95, bolstered by a confidence score of 0.631, with the potential to increase to 0.80 in confidence without significantly compromising the F1-score. These findings are poised to contribute substantially to the broader research landscape, particularly in advancing the effectiveness of detection models for traffic officers.