A comparative study of noise elimination algorithms for a 3D terrain model through object clustering and the differential method

Abstract

The technology to automatically detect the surrounding working environment for modeling of the result is the key essential technique for the automation of the construction equipment developments. When noise takes place during 3D modeling of the front work area of the automated construction equipments, the point of noise occurrence during the surface modeling can result in ground level phenomenon in the form of triangular pyramid to lower the quality of the modeling. This can greatly affect the detection of the objects around the automated equipment. This study proposed revised object clustering and differential algorithms for noise elimination of 3D terrain model and compared the noise elimination performance of existing algorithm and proposed algorithm on the images of actual earthwork working environment. It is expected that the noise elimination algorithm proposed in this study will be very useful as a widely used essential technology required for the development of automated technology not only in earthwork field but also in other general construction and civil engineering fields by noise elimination of the 3D working environment model.