Adaptive Denoising Algorithm for Scanning Beam Points based on Angle Thresholds

摘要

A local adaptive neighborhood model is proposed in this paper in order to deal with the mistake judgment in the existing scanning beam point cloud denoising algorithms. Such a model regards larger curvatures as the potential noises, can select angle thresholds of noise points and the median values of filtering windows adaptively, so as address the issues of mistake judgment and missing judgment of the point clouds denoising algorithms with different curvatures. The adaption theory in the angle threshold denoising algorithm classifies the noise points and data points. Therefore, it can ensure the smoothness in low frequency, and as well keep the high frequency characteristics. The new method improves the accuracy of median filtering, prevents the diffusion of noise, remove noises effectively while preserving sharp features, and avoid fuzzy data margin.