Retro Reflectivity Measurement of Pavement Marking by Using Image Processing

Pavement marking is an essential element of traffic operation and road safety. Pavement markings are beneficial because they provide information to drivers about road alignment and conditions ahead. The quality of pavement markings at nighttime is measured by their retro-reflectivity value. This study aims to investigate a technique to predict the retro-reflectivity value of pavement markings using image processing. A relationship between brightness value and retro-reflectivity value will be utilized to make predictions. In this study, computer coding was developed to analyze pictures of pavement markings at night to identify the brightness of markings 15 meters away from a testing vehicle. This involved selecting different ranges of thermoplastic pavement marking retro-reflectivity values by choosing testing sites at King Mongkut's University of Technology Thonburi. The analysis employed an exponential function to create a retro-reflectivity prediction model based on the relationship between brightness value and retro-reflectivity value.