University of Tasmania
Gautam_whole_thesis.pdf (12.73 MB)

Direct georeferencing and footprint characterisation of a non-imaging spectroradiometer mounted on an unmanned aircraft system

Download (12.73 MB)
posted on 2023-05-27, 09:43 authored by Gautam, D
Point-measuring spectroradiometers have gained interest in recent years for airborne spectroscopy from unmanned aircraft systems (UAS) facilitated by miniaturisation and improvements in signal quality. Spectroradiometers sample radiance and reflectance measurement at a very high spectral resolution, and are commonly used for laboratory measurements and outdoor proximal remote sensing. When mounted on a UAS, they offer a unique potential in quantifying vegetation properties, such as solar-induced fluorescence (SIF), in unprecedented spatial detail. To understand and to correctly interpret the spectral signal measured from the UAS, the extent and geolocation of the measurement spot, referred to as the spectral footprint becomes crucial. This thesis investigates how the footprint of a point-measuring spectroradiometer on a UAS can be determined. The focus of the thesis is on the pre-requisites for the footprint determination, specifically a) selection of position and orientation (pose) measurement sensors, b) their detailed geolocation error budget and c) geometric calibration, followed by d) the footprint determination and validation. This study is motivated by the upcoming Fluorescence Explorer (FLEX) satellite mission of the European Space Agency, equipped with a fluorescence imaging spectrometer. This study is an enabler for mapping the spatial distribution of the SIF signal in detail, which subsequently has the potential to contribute to validation of FLEX observations and an improved understanding of the fluorescence signal. A UAS-based spectroradiometer system, equipped with inertial measurement units (IMUs), global navigation satellite system (GNSS) receivers, a monochrome camera, and an on-board computer, was designed for this study. The primary IMU/GNSS is used to measure the spectroradiometer pose, while the secondary IMU/GNSS measures the orientation of the UAS airframe. To determine the most suitable primary IMU/GNSS, two microelectromechanical systems (MEMS)-based units were tested in a series of groundbased experiments. The IMU/GNSS units were assessed in their performance accuracy, size and weight, and synchronisation capability with the spectroradiometer. Accordingly, the Spatial Dual IMU with its dual-frequency dual antenna GNSS was determined as the most suitable sensor to measure the spectroradiometer pose on a UAS. The selected IMU/GNSS unit was used to develop an error budget model aiming to estimate the uncertainty associated with the footprint geolocation. The model propagates, through an aerial data georeferencing formula, the input uncertainties originating from a) the on-board IMU/GNSS sensors, b) the sensor calibration, and c) the digital surface model (DSM). The model was used to investigate the effect of various spectroradiometer field of view (FOV), integration time, UAS flying speed and flying height values, as well as the grade of the IMU device on the footprint geolocation error. The orientation accuracy of the spectroradiometer, resulting from IMU and boresight angle errors, was found to be the dominant source of the footprint geolocation uncertainty. Subsequently, the lever-arm offset and boresight angle must be properly calibrated and the spectroradiometer FOV correctly determined in order to achieve the accurate computation of the footprint. The lever-arm offsets were measured using a scaled threedimensional point cloud representation of the UAS-spectroradiometer system. The point cloud was created from the photos of the airframe processed with the structure-frommotion (SfM) algorithm. The boresight angles were estimated from stationary experiments that computed the difference between the orientations of IMU, spectroradiometer, and camera. The footprint size of the spectroradiometer for a known distance to the target was measured experimentally, thereby enabling the FOV angle to be determined. The lever-arm and boresight correction was applied to the in-flight dataset and the correction was validated through data acquired by the co-mounted camera processed using the SfM algorithm. The footprint geolocation was derived via a ray-casting algorithm that uses the calibrated spectrometer pose measurement together with a DSM of the observed terrain. The combined effect of spectroradiometer integration time, UAS flying speed, spectroradiometer pointing angle, and terrain slope was incorporated in the footprint calculation. Using the ray-casting algorithm, the constituent boundary points of the footprint were computed for each IMU/GNSS epoch, and these points were connected to form an instantaneous footprint. The series of instantaneous footprints collected during the integration time of the spectroradiometer were combined to form the final footprint shape. To validate the spectroradiometer footprint location, the ray-casting technique was applied to isolated pixels of the co-mounted camera. The resulting geolocation of the isolated pixels were compared with the surveyed ground control points. The footprint spectral validation was performed by comparing UAS-based and ground-measured reflectance signatures of natural targets. The overarching goal of this thesis, to determine geolocation and extent of the spectroradiometer footprint, was achieved with a spatial accuracy of 15 cm ¬¨¬±1˜ìvâ for a flying height of 10 m. The achieved level of spatial detail and accuracy can be considered as sufficient to retrieve a SIF signal over heterogeneous vegetated surfaces. Hence, this study made significant steps towards robust UAS-based acquisition of spatially explicit SIF measurements, which will facilitate the validation of future FLEX satellite observations. The work in this thesis has contributed to improved acquisition and understanding of the SIF signal, which will in turn improve our ability to monitor photosynthetic activity of vegetation at the global scale.


Publication status

  • Unpublished

Rights statement

Copyright 2019 the author Chapter 2 appears to be the equivalent of a post-print version of an article published as: Gautam, D., Lucieer, A., Malenovsk√Ω, Z., Watson, C., 2017. Comparison of MEMS-based and FOG-based IMUs to determine sensor pose on an unmanned aircraft system, Journal of surveying engineering 143(4), 1-11 Chapter 3 appears to be the equivalent of a pre-print version of an article published as: Gautam, D., Watson, C., Lucieer, A., Malenovsk√Ω, Z., 2018. Error budget for geolocation of spectroradiometer point observations from an unmanned aircraft system, Sensors, 18(10), 1-16. c 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ( Chapter 4 appears to be the equivalent of a pre-print version of an article published as: Gautam, D., Lucieer, A., Watson, C., McCoull, C., 2019. Lever-arm and boresight correction, and field of view determination of a spectroradiometer mounted on an unmanned aircraft system, ISPRS journal of photogrammetry and remote sensing, 155, 25-36 Chapter 5 appears to be the equivalent of a pre-print version of an article published as: Gautam, D., Lucieer, A., Bendig, J., Malenovsk√Ω, Z., 2019. Footprint determination of a spectroradiometer mounted on an unmanned aircraft system, IEEE transactions on geoscience and remote sensing, Early view, 6 November 2019. Copyright 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Repository Status

  • Open

Usage metrics

    Thesis collection


    No categories selected


    Ref. manager