Waveform Information Accurate Extraction for Massive and Complex Waveform Data of Hyperspectral Lidar

Hyperspectral lidar (HSL) has high-precision geometric information and high-resolution spectral information, and its advantageous detection capability has been recognized by scientists at home and abroad. However, how to extract massive and complex HSL data effectively and accurately is an important issue in the current development of HSL. A methodological system that caters to HSL data features is required to achieve high-precision spatial–spectral integrated data interpretation. This requirement represents a significant scientific challenge, for which research on appropriate HSL waveform data processing methods remains scarce. This study aims to address the challenges posed by the massive data and complex waveform situations associated with HSL. Based on an experimental verification, the single-channel algorithm suggested in this article proves to be advantageous over Gaussian decomposition, specifically for asymmetric and overlapping echoes. This algorithm produces an average R² increase of 0.023 and reduces the standard deviation by 63%. It also accurately extracts hidden overlapping echoes. Furthermore, this study proposes a multi-channel-assisted optimization algorithm that can precisely extract faint and overlapping echoes that a single-channel algorithm cannot extract. Its accuracy is remarkably high, with the ranging accuracy boosted by 98% compared with that of the single-channel algorithm.