Traffic light counter detection comparison using you only look oncev3 and you only look oncev5 for version 3 and 5

Abstract

<p>This project aims to develop a vision system that can detect traffic light<br />counter and to recognise the numbers shown on it. The system used you only<br />look once version 3 (YOLOv3) algorithm because of its robust performance<br />and reliability and able to be implemented in Nvidia Jetson nano kit. A total<br />of 2204 images consisting of numbers from 0-9 green and 0-9 red. Another<br />80% (1764) from the images are used for training and 20% (440) are used for<br />testing. The results obtained from the training demonstrated Total<br />precision=89%, Recall=99.2%, F1 score=70%, intersection over union<br />(IoU)=70.49%, mean average precision (mAp)=87.89%, Accuracy=99.2%<br />and the estimate total confidence rate for red and green are 98.4% and 99.3%<br />respectively. The results were compared with the previous YOLOv5<br />algorithm, and the results are substantially close to each other as the YOLOv5<br />accuracy and recall at 97.5% and 97.5% respectively.</p>