抽象的な

THREE DIMENSIONAL DATA REDUCTION ALGORITHM BASED ON SURFACE FITTING

Hussein Abo-surrah, Saidghoniemy, E.A.Zanaty, Ashraf Afifi

This paper proposes an approach for 3D data reduction based on estimating the surface normal vectors for handling the amount of data acquired by laser scanning. The data points are partitioned into cells based on their x, y, and z-axis positions.For normal vector computation, wefit the point data of each cell to implicit general quadric. Then, the shortest distance is directly estimated by intersecting the implicit surface with a line passing through the given point according to the estimated orthogonal orientation, which is necessary for normal vectors computation. The points in each cell are assigned their corresponding estimated normal vectors. For each cell, the points are reduced according to the normal vectorsdirection. The median point is chosen as representative point of the cell if the assigned normals point to the same direction, otherwise the average point is selected. The performance of the proposed method is illustrated using a range of point clouds scanned from typical engineering surfaces. Keyword: Normal vector, surface fitting, 3D data reduction