Citation

Kabir, Fardin and Roslee, Mardeni and Ali, Farman and Ullah, Yasir and Santo, Istiyak Amin and Kabir, Fahmid (2025) Toward Safer Roads: A Low-Cost Intelligent Pothole Detection System for Smart Cities. In: 2025 International Conference on Engineering and Emerging Technologies (ICEET), 22-23 October 2025, Kuala Lumpur, Malaysia.

Text 2.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Potholes represent a significant threat to urban road safety because they damage vehicles and endanger drivers. The research develops a small-scale IoT pothole detection system which uses an ultrasonic sensor and GPS module and Sony Spresense microcontroller with an integrated camera for live road surface inspection. The system automatically records geotagged images of potholes that surpass depth limits while it saves time- stamped data to a microSD card for future Google Maps visualization. The proposed system operates independently from the cloud while using minimal power and allowing users to add new components to the system. The YOLO-based image analysis module performs pixel-distance calibration to estimate pothole dimensions during post-processing for intelligent severity assessment. The system performed well during field tests by detecting potholes and mapping their GPS locations with acceptable precision. The system enables large-scale implementation for both road maintenance and smart city infrastructure management through its sensing-vision combination.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Pothole Detection, Ultrasonic Sensor, Camera Imaging, GPS Tracking, Smart City Infrastructure
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800-8360 Electronics > TK7871 Electronics--Materials
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	20 Apr 2026 01:02
Last Modified:	20 Apr 2026 01:02
URII:	http://shdl.mmu.edu.my/id/eprint/15738

Downloads

Edit (login required)