No abstract

No abstract

In developing countries like Timor-Leste, regular road condition monitoring is a significant subject not only for maintaining road quality but also for a national plan of road network construction. The sophisticated equipment for road surface inspection is so expensive that it is difficult to introduce them in developing countries, and the monitoring is usually achieved by manual operation. On the other hand, the utilization of ICT devices such as smartphones has gained much attention in recent years, especially in developing countries because the penetration rate of the smartphone is remarkably increasing even in developing countries. The smartphones equip various high precision sensors, i.e., accelerometers, gyroscopes, GPS, and so on, in the small body in low price. In this project, we are developing an integrated road condition monitoring system that consists of smartphones, dashcams, and a server. There are similar trials in advanced countries but not so many in developing countries. This system assumes to be used in developing countries. The system is very low cost and does not require trained specialists in the field side. The items that are automatically inspected in this system were carefully selected with the local ministry of public works and include paved and unpaved classification, road roughness, road width, detection and size estimation of potholes, bumps, etc., at present. All the inspected items are visualized in Google Maps, Open Street Map, or QGIS with GPS information. The survey results are collected on a server and updated to more accurate values by the repeated surveys. On the analysis, we use several state-of-the-art machine learning and deep learning techniques. In this paper, we summarize related works and introduce this project’s target and framework, which especially focused on the developing countries, and achievements of each of our tasks.

Road condition monitoring usually requires extremely expensive special vehicles, equipment, or many human resources. On the other hand, with the development of ICT and data science technologies in recent years, there are several research trials in which the heavy technical tasks of road asset condition monitoring are replaced by automatic inspection systems consisting of common devices such as smartphones and dashcam videos. As the system consists of low-price devices, it also suitable for developing countries. However, there are many differences in the situation and the inspection items on road condition monitoring between advanced countries and developing countries. There are few trials to develop such a road condition monitoring system in developing countries. Our project is developing an integrated road condition monitoring system focusing on developing countries like Timor-Leste. In developing countries, many parts of the road are still unpaved, and the “road width” is an important item to be inspected. In this paper, we discuss the road width and pothole size estimation as a part of the integrated system we are developing. We survey the road width of both paved and unpaved roads. We use a common dashcam to take video along the road. The estimated values are integrated into a database with GPS information and visualized in Google Map, QGIS, or the original visualization system which we developed. To estimate the real width of the road and pothole size, we need to transform the captured forward view image of dashcam video into bird’s-eye-view. For the transformation, we need to estimate the vanishing point in a captured image. However, unlike the advanced countries, it is difficult to detect the vanishing point in developing countries because there are usually no straight lines in the images in the unpaved road of the province. In this study, we propose to use the optical flow method to detect the vanishing point in the rural road. To identify the area of road and the existence of potholes in images, we apply state-of-the-art semantic segmentation using deep learning.