This study aims to validate the feasibility of using LiDAR reflectivity data to quantitatively estimate the retroreflectivity of road lane markings. The goal is to establish the optimal scanning conditions considering the channel position, angle of incidence, and vehicle speed for an accurate and consistent retroreflectivity assessment in mobile environments. Fifteen standard lane marking samples with known retroreflectivity values were scanned using an OS1-128 LiDAR sensor under controlled field conditions. A two-phase experiment was conducted: (1) a speed-based test to assess the influence of vehicle velocity (20-80 km/h) on LiDAR reflectivity measurements, and (2) a channel–angle–distance test using a static testbed to analyze the relationship between retroreflectivity, LiDAR channel position (that is, the angle of incidence), and measurement distance. Ground truth retroreflectivity values were obtained using a high-precision handheld retroreflectometer. Reflectivity measurements showed a strong correlation with standard retroreflectivity values, particularly at scanning angle between 100-115° and distances of 4.9-5.6 m. The coefficient of determination (R2) exceeded 0.97 across optimal conditions. Speedrelated tests confirmed that the LiDAR-based reflectivity remained stable with a minimal RMSE (< 5), even under high-speed driving scenarios. LiDAR sensors provided reliable and contactless estimates of pavement marking retroreflectivity when the channel angle and scanning distance were appropriately selected. The findings demonstrated that channel-specific calibration and incidence angle correction significantly improved the measurement accuracy. This suggests a practical path forward for automated large-scale retroreflectivity monitoring in road asset management systems.

This project aimed to understand the near-infrared (NIR), intensity, and reflectivity characteristics of LiDAR for measuring retroreflectivity and to understand the correlation between the characteristics of LiDAR and retroreflectivity. A 600 m-testbed was investigated using a survey vehicle equipped with LiDAR, and the testbed retroreflectivity and LiDAR data measurement values were compared. The reflectivity and intensity at night were not affected by sunlight compared with daytime, enabling stable data collection. However, NIR reacted very sensitively to sunlight, and the difference between daytime and nighttime NIR values was very large. In addition, by comparing the absolute error between the retroreflectivity and LiDAR data, we observed that the reflectivity was consistent with the data difference between day and night, and it was not significantly affected by sunlight. However, the intensity showed that the daytime measurement data were more scattered than the nighttime measurement data, resulting in low-precision collection stability caused by sunlight. An analysis of the correlation between retroreflectivity and LiDAR data using 40 data points revealed that the reflectivity and intensity data at night were highly correlated with retroreflectivity, with a P-value of less than 0.05. Reflectivity and intensity values at night correlate with retroreflectivity. The NIR light is sensitive to sunlight. Thus, it can be used as a solar correction index for future retroreflectivity analyses using intensity.