Automatic Extraction and 3D Reconstruction of Bundle Conductor Based on Improved DPC Algorithm

This paper presents a method for automatic extraction and 3D reconstruction of bundle conductor based on improved DPC algorithm. Firstly, the least square method is used to fit the point clouds of bundle conductors with a parabola equation in three-dimensional space, and then the spatial coordinates of each bundle sub-conductor relative to the fitting curve are calculated. Next, an improved DPC algorithm is adopted to cluster the point clouds of the bundle sub-conductors in order to discriminate the number of the bundle sub-conductors and extract the point clouds of each bundle sub-conductor. The method is proved to be more robust than the K-MEANS algorithm. Finally, the least square method is developed to reconstruct the three-dimensional model of the bundle sub-conductors.