At the 35th Japan Road Conference ( November 1-2, 2023, hosted by the Japan Road Association), a paper presented by Dr. Bo-Woo Cho, President of MHI, received the Award for Excellence and was commended.
Title of paper: "Development of Road Surface Rut Evaluation Technology Using Cameras Installed in Vehicles
■ Awards Overview:
The award is given to papers whose contents and presentations are deemed particularly outstanding from the perspective of contributing to the improvement of road technology and the promotion of road projects, taking into consideration originality, usefulness, reliability, clarity, and presentation attitude and question-and-answer session.
Summary of Research Results:
We proposed a technique for evaluating the cross-sectional profile and the amount of rut excavation based on 2D images taken by a camera installed inside a vehicle. Field tests showed that the proposed method outperforms the conventional SfM method in estimation accuracy, and the majority of the rut excavation estimation errors were less than 3 mm. Furthermore, the proposed method was also applied to the estimation of pothole depth, and it was confirmed that the relative error to the manually measured value was about 10%. Since the proposed method can be applied not only to cross-sectional profiles but also to longitudinal profiles, it is expected to be utilized in constructing 3D profiles of road surfaces.
*The above research results were developed through a matching project by the Kanto Regional Development Bureau in FY2022. (Research theme: Development of a simple evaluation method for rut moat and improvement of productivity of inspection work by automatic calculation of MCI)
We proposed a technique for evaluating the cross-sectional profile and the amount of rut excavation based on 2D images taken by a camera installed inside a vehicle. Field tests showed that the proposed method outperforms the conventional SfM method in estimation accuracy, and the majority of the rut excavation estimation errors were less than 3 mm. Furthermore, the proposed method was also applied to the estimation of pothole depth, and it was confirmed that the relative error to the manually measured value was about 10%. Since the proposed method can be applied not only to cross-sectional profiles but also to longitudinal profiles, it is expected to be utilized in constructing 3D profiles of road surfaces.
*The above research results were developed through a matching project by the Kanto Regional Development Bureau in FY2022. (Research theme: Development of a simple evaluation method for rut moat and improvement of productivity of inspection work by automatic calculation of MCI)
