Efficient Streetlamp Detection with YOLO Models on Edge Device

Beibut Amirgaliyev, Sunggat Aiymbay, Аinur Zhumadillayeva

2024

Closed access Metadata (CSL-JSON)
DOI 10.1109/mosicom63082.2024.10881770
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This study presents a real-time streetlamp detection system using a Raspberry Pi and dashcam, leveraging YOLOv5 and YOLOv8 models. Tested in various conditions, YOLOv8n outperformed YOLOv5n with a mean Average Precision (mAP) of $73.7 \%$ versus $\mathbf{6 9. 4 \%}$, along with higher Precision and Recall. YOLOv8n demonstrated superior accuracy and reliability, making it better suited for IoT-based smart lighting applications in smart cities. This research underscores the potential of affordable edge-computing for adaptive urban lighting solutions.