Abstract
Rivers have been bringing us benefits since ancient times, while occasionally raging. Living spaces and rivers are closely related with each other, and the social responsibility of river management is extremely important in ensuring secure, safe, and comfortable life. Among river management duties, periodic cross-sectional survey in river, which actually measures the open space along river to create cross sections, is performed over class-A rivers nationwide since it can be used for embankment improvement plans or grasping discharge capacity etc. Actual comprehensive measurement of all the class-A rivers nationwide incurs a huge cost. Therefore, efforts have been made on generation of cross sections using point cloud data obtained by the airborne laser. However, there still remain measurement errors or problems such as the noise of vegetation having an influence on the result. Thus this paper proposes a method of solving the remaining problems and creating a cross section with high precision from point cloud data.
