A Robust Pedestrian Detection Approach for Autonomous Vehicles

Nowadays, utilizing Advanced Driver-Assistance Systems (ADAS) has absorbed a huge interest as a potential solution for reducing road traffic issues. Despite recent technological advances in such systems, there are still many inquiries that need to be overcome. For instance, ADAS requires accurate and real-time detection of pedestrians in various driving scenarios. To solve the mentioned problem, this paper aims to finetune the YOLOv5s model for handling pedestrian detection challenges on the real-world instances of Caltech pedestrian dataset. We also introduce a developed toolbox for preparing training and test data and annotations of Caltech pedestrian dataset into the format recognizable by YOLOv5. Experimental results on the Caltech pedestrian dataset have verified that the mean Average Precision (mAP) of our fine-tuned model for pedestrian detection task is equal to 91 percent when performing at the highest rate of 70 FPS. Moreover, our proposed approach can outperform other existing methodologies in terms of accuracy and speed.