Environment Awareness with Hyperspectral LiDAR Technologies

Abstract

LiDAR (Light Detection And Ranging, also known as LADAR) is an active optical remote sensing technique that can measure the distance by counting the time of flight (ToF) of the transmitted laser and acquire other physical properties of a target by illuminating the target with a light source, often using a pulsed laser due to the high spatial and temporal coherence nature. Traditionally a monochromatic laser beam can map physical features with a very high spatial resolution in a non-contact manner. Limited by the employed monochromatic laser source, traditional LiDAR sensors usually operate at one or several discrete wavelength bands. These systems can provide high-accuracy spatial information, but restricted spectral information: compared with a passive spectrometer, the spectral information and spectral resolution are inadequate for some particular remote sensing applications. The motivation of this research is straightforward: investigating SCL-based hyperspectral LiDAR (HSL) technology, more specifically: investigating HSL technique with different system configuration, confirming their accuracy both in spatial and spectral measurements, and conducting various feasibility studies towards environmental awareness applications, more specific, in in forestry, plant science and mining. The emphasis was on system development and methodology development. The specific aims of this dissertation based on six prototyped hyperspectral LiDAR were as follows: • to investigate different hardware techniques including optics, electronics, and post-processing techniques for HSL system development covering the spectrum from visible to near-infrared (VNIR) to shortwave infrared (SWIR) (for better eye-safety operation) with discrete or continuous spectral channels for remote sensing applications; • to develop methods to evaluate the performance of the HSL systems concerning the spectral measurement accuracy, the range stability and resolution over the covered spectral range ; • to assess the feasibility of the developed methods in forestry, mining and plant sciences related environment awareness applications with the point clouds or range measurement containing the spectral information.