Research on the Classification of Road Surface Types and Conditions Using Artificial Intelligence 


Vol. 14,  No. 7, pp. 524-532, Jul.  2025
https://doi.org/10.3745/TKIPS.2025.14.7.524


PDF
  Abstract

Accurately identifying the type and condition of road surfaces in various road environments is crucial for safe traffic operations and road maintenance. However, electromagnetic wave-based sensors commonly used, such as cameras, radar, and lidar, have problems with recognition errors or detection range limitations in situations where road surface characteristics rapidly change, such as ice, snow, and slush. As an alternative solution, non-contact ultrasonic sensor-based road condition detection techniques are emerging. This study proposes a method that preprocesses acoustic reflection signals acquired through continuous ultrasonic transmission (continuous wave) and uses a LightGBM-based machine learning model to classify road material (asphalt, cement, and soil) and surface conditions (dry, damp, wet, snow, ice, and slush) in real-time. The proposed method is lighter and faster than conventional impulse ultrasonic transmission or artificial neural network (e.g., PatternNet) based approaches, and achieves high accuracy across various road/weather conditions. Experimental results on 18 road scenarios (3 types of materials and 6 types of conditions) implemented in stationary conditions showed classification accuracy of over 99% even in extreme environments. This confirms the potential for AI-based road surface recognition technology to be utilized in improving understanding of road infrastructure, maintenance, and safety assessment.

  Statistics
  Cite this article

[IEEE Style]

P. S. Hyun, K. H. Jun, H. S. Jun, J. J. Eun, K. Min-Hyun, "Research on the Classification of Road Surface Types and Conditions Using Artificial Intelligence," The Transactions of the Korea Information Processing Society, vol. 14, no. 7, pp. 524-532, 2025. DOI: https://doi.org/10.3745/TKIPS.2025.14.7.524.

[ACM Style]

Park Sang Hyun, Kwon Hyun Jun, Hong Seong Jun, Jeong Jung Eun, and Kim Min-Hyun. 2025. Research on the Classification of Road Surface Types and Conditions Using Artificial Intelligence. The Transactions of the Korea Information Processing Society, 14, 7, (2025), 524-532. DOI: https://doi.org/10.3745/TKIPS.2025.14.7.524.

Forms

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)

Old Journals

Indexing

All publications of TKIPS is indexed in DOI, EBSCO, Google Scholar, Crossref, and CrossCheck.

TKIPS is also selected as the Journal for Accreditation by NRF (National Research Foundation of Korea).

Related Journals