MS-RBC: A Range Bias Correction Method for Saturated Returns in Multiwavelength Underwater LiDAR

Abstract

Multiwavelength underwater LiDAR can simultaneously provide 3-D geometric and spectral information for precise seafloor topography mapping and benthic habitat characterization. However, achieving both high-ranging accuracy and cross-wavelength ranging consistency remains challenging. Systems employing a supercontinuum laser source are subject to wavelength-dependent pulse broadening, which results in wavelength-dependent ranging bias when combined with receiver saturation under high-gain conditions. Existing range bias correction (RBC) methods for saturated returns are mainly designed for single-wavelength systems with fixed transmitted pulse widths and are therefore not effective in handling this wavelength-dependent bias. To address this issue, we propose a maximum slope RBC (MS-RBC) method. The proposed method estimates the interface arrival time from the MS time of the waveform leading edge, thereby reducing the influence of saturation and broadening of the pulse trailing edge. Experimental results under varying distances, wavelengths, pulse energies, target reflectance, and water quality conditions demonstrate that MS-RBC significantly improves both ranging accuracy and cross-wavelength consistency. Specifically, it achieves a mean absolute ranging error of 0.017 m and a cross-wavelength ranging deviation below 0.02 m. These results indicate that the proposed method enables reliable co-registration of geometric and spectral measurements and provides a robust foundation for underwater topographic mapping and spectral analysis.