Observation Selection, Total Variation, and L-Curve Methods for LiDAR Data Denoising

• Published in: Advances in meteorology Vol. 2022; pp. 1 - 17
• Main Authors: Zhang, Jie, Chan, Pak Wai, Ng, Michael K.
• Format: Journal Article
• Language: English
• Published: New York Hindawi 22-01-2022; John Wiley & Sons, Inc; Hindawi Limited
• Subjects: Airports; Algorithms; Detection; Lidar; Lidar observations; Methods; Noise reduction; Observatories; Optical radar; Quality control; Regularization; Remote sensing; Signal to noise ratio; Smoothing; Wavelet transforms; Wind profiles; Wind speed; Wind velocities; China; Hong Kong
• ISSN: 1687-9309; 1687-9317
• DOI: 10.1155/2022/3792225

In this article, we propose a light detection and ranging (LiDAR) data denoising scheme for wind profile observation as a part of quality control procedure for wind velocity monitoring and windshear detection. The proposed denoising scheme consists of several components. (i) It selects LiDAR observations according to their SNR values so that serious noisy data can be removed. (ii) A polar-based total variation smoothing term is employed to regularize LiDAR observations. (iii) The regularization parameters are automatically determined to balance the data-fitting term and the total variation smoothing term. Numerical results for LiDAR data collected at the Hong Kong International Airport are reported to demonstrate that the denoising performance of the proposed method is better than that of the testing LiDAR data denoising schemes in the literature.