Student education projects

Point cloud Track

This is a link to the manual of our scanner. Many of us use the free Cloud Compare software. This is the latest version of Manual

Thesis writing hints

Possible BSc/MSc thesis projects

Sometimes BSc Civil Engineering topics are given at BSc Eindwerken CT3000 . Typically I have more topics then listed there

Completed student thesis projects

In most cases the pdf icon links to the corresponding TU Delft repository page, that has an abstract of the thesis and the full thesis in pdf. BSc theses: incomplete.


PDF icon

Point cloud based improvement of the trajectory of a Railway Mobile Mapping System
Pablo Secco
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2022

PDF icon

Vegetation monitoring in coastal areas using readily available Remote Sensing data
Harm Kathmann
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2022


PDF icon

Stability of rocks on mild slopes
Tom Venrooy
MSc thesis, MSc track Hydraulic Engineering, Delft University of Technology, 2021

PDF icon

Analyzing reconstruction scenarios for dolmen D14 using non-destructive, digital methods
Emy Theuws
MSc thesis, MSc track Geo-Engineering, Delft University of Technology, 2021

PDF icon

Landslide detection and mapping using Synthetic Apertue Radar amplitude imagery
Marja Michielse
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021

PDF icon

The effects of storm Ciara and Denis on the coastline of Noordwijk using weekly and monthly LiDAR data sets
Karla Echeverry Caro
BSc thesis, Applied Earth Sciences, Delft University of Technology, 2021

PDF icon

Photogrammetric Deformation Analysis of a Quay Wall
Stijn Lodder
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021

PDF icon

Water surface heights in the upper Brahmaputra and Nam Co basin with ICESat-2
Simone van Gent
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021

PDF icon

3D Road Boundary Mapping of MLS Point Clouds
Qian Bai
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021

PDF icon

Rail Line Detection Based Photogrammetry: Using image content knowledge to improve 3D reconstruction of railway tracks
Hessel Prins
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021

PDF icon

Automated railway object mapping using imagery and point clouds
Anastasios Vogiatzis
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021

PDF icon

Machine Learning-based Classification of Different 3D Point Cloud Data of Railway Environments using Random Forest and DGCNN
Annie Papalexiou
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021

PDF icon

Detecting Alpine glacier changes from a combination of ICESAT-2 and GEDI data
Timo Bisschop
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2021


PDF icon

Satellite derived bathymetry for times of absent in situ data
Floortje Burgers
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2020

PDF icon

Deep Localization of Static Scans in Mobile Mapping Point Clouds
Fancong Meng
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2020

PDF icon

Land Type Classification of ICESat-2 Global Geolocated Photon Data
Joseph Heywood
MSc thesis, Fact. Aerospace Engineering, Delft University of Technology, 2020

PDF icon

Semantic Segmentation of AHN3 Point Clouds with DGCNN
Qian Bai
Small MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2020

PDF icon

Wave attenuation due to vegetation: Duursche Waarden (case study)
Max de Bruijn
MSc thesis, Civil Engineering (Dept. HE), Delft University of Technology, 2020

PDF icon

Stability of rock on mild slopes
Bas van Wijland
MSc thesis, Civil Engineering (Dept. HE), Delft University of Technology, 2020

PDF icon

Performance evaluation of the CWI BRDF-fitting method under cloud-contaminated conditions: A numerical experiment using PROSAIL
Jigme Klein
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2020

PDF icon

Road Detection from Remote Sensing Imagery
Kaniouras, Pantelis
MSc thesis, Geomatics, Delft University of Technology, 2020

PDF icon

Multi-case Evaluation of Empirical Methods for Satellite-Derived Bathymetry
Sean Zandbergen
MSc thesis, Civil Engineering (HE) Delft University of Technology, 2020

PDF icon

Towards an Antarctic meteorite hotspot map
Veronica Tollenaar
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2020


PDF icon

Combining microwave and optical remote sensing to monitor rivers in monsoon affected regions
Mirja Dooren
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2019

PDF icon

Predictive Modelling of Asphalt Concrete Functional Properties Using Multiple Linear Regression and Gradient Boosting
Giulia Martini
MSc thesis, MSc Civil Engineering, Delft University of Technology, 2019

PDF icon

Random Forest Classification of three different species of trees in Delft, based on AHN point clouds: Additional Thesis
Kirsten van Dongen
Small MSc thesis, MSc Track Geoscience and Remote Sensing, Delft University of Technology, 2019

PDF icon

The stability of stones on mild slopes under wave attack
Lieke Mossinkoff
MSc thesis, MSc Civil Engineering, Delft University of Technology, 2019

PDF icon

Eutrophication prediction in the Dutch coastal waters using remote sensing data and machine learning
Kostas Chatzopoulos Vouzoglanis
MSc thesis, MSc Civil Engineering, Delft University of Technology, 2019

PDF icon

Semantic Segmentation of Large-scale Urban Scenes from Point Clouds
Zhiwei Ai
MSc thesis, 3ME, Delft University of Technology, 2019

PDF icon

The Current State of the St. BavoChurch Tower: A Verification Study and Structural Analysis of the Wooden Tower of the St. Bavo Church in Haarlem,The Netherlands
Pieter Woudenberg
MSc thesis, Civil Engineering, Delft University of Technology, 2019

PDF icon

Mobile laser scan data for road surface damage detection
Bibi van der Horst
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2019

PDF icon

The influence of drone flightpath on photogrammetric model quality
Tom Sassen
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2019

PDF icon

Accurate, Detailed and Automatic Tree Modelling from Point Clouds
Shenglan Du
MSc thesis, Geomatics, Delft University of Technology, 2019


PDF icon

Sensitivity of snow models in the semi-arid Andes of Chile: A comparison between SNOWPACK and SnowModel
Annelies Voordendag
MSc thesis, Track Geoscience and Remote Sensing, Delft University of Technology, 2018

PDF icon

Deep segmentation of the drivable path of a self-driving vehicle using external data: Influence of domain shift factors and depth information
Robbert Bormans
MSc thesis, Track Geoscience and Remote Sensing, Delft University of Technology, 2018

PDF icon

Mapping sub-annual beach growth using terrestrial laser scanning: A study on the application of terrestrial laser scanning on small scale beach variability, to quantify beach resilience on sub-annual time scale
Rik Hobbelen
MSc thesis, Civil Engineering, Delft University of Technology, 2018

PDF icon

Lateral persistency of precession-driven floodplain cycles and their relation to fluvial sandbodies
Jasper de Lanoy
MSc thesis, Applied Earth Sciences, Delft University of Technology, 2018

PDF icon

Locating PS-InSAR derived deformation using LiDAR point clouds
Adriaan van Natijne
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2018

PDF icon

Semi-automatic fold detection in geological outcrops using Laser Scanning data: Applied to a limestone-marl sequences in the Vocontian Basin, France
Joppe Roebroeks
MSc thesis, MSc track Geoscience and Remote Sensing, Delft University of Technology, 2018

PDF icon

Methods For Feature Extraction of an Outcrop Using Terrestrial LiDAR: Methods For Feature Extraction of an Outcrop Using Terrestrial LiDAR
Surya Pahal
BSc thesis, Applied Earth Sciences, Delft University of Technology, 2018

PDF icon

Comparing WWII aerial photographs to Sentinel-2 data
Tom Sassen
Small MSc thesis, Track Geoscience and Remote Sensing, Delft University of Technology, 2018

PDF icon

The effectiveness of a snow fence in the dry Andes region of Chile
Coco Antonissen
Small MSc thesis, Track Geoscience and Remote Sensing, Delft University of Technology, 2018

MSc thesis project - Thijs Verweij

Lighting optimization and battery management of solar powered street lights, including weather predictions

MSc thesis project - Nadir

Heat island assessment using WOW meteorological citizen data

MSc small thesis project - Bibi van der Horst

Sentinel logo Second year BSc Applied Earth Science students go for geological fieldwork in the Drome Provencale, France. Here they practice how to characterize sedimentary rock outcrops in terms of geometry. Their efforts could be made reusable by creating and storing a 3D surface model of selected outcrops. In her thesis work, Bibi has investigated if using the open-souce program MicMac it is possible to automatically gerenerate a suitable 3D model from photos that are uploaded by students to a server. A protocol was made for students on how to sample outcrops. This protocol was evaluated on different input using a shell script calling MicMac commands Although the workflow works in most cases, there are several adjustments to do in the future, like to make it more robust, find a method to create the right scales and to display all 3D models in one map

PDF icon

Automatic generation of rock outcrop point clouds from student photo's
Bibi van der Horst
Small MSc thesis, Applied Earth Science, Delft University of Technology, 2018

Small MSc thesis project - Shuangbaihe Zhou

Detecting changes in a Himalaya valley from Sentinel-2 satellite data

Bsc thesis project - Lucas Veldkamp

Validating the usage of satellite data for the OPERANDUM project

Bsc thesis project - Julian van Dingen

Integration of a telecom network administration in a Geographical Information System

Bsc thesis project - Stijn Lodder

ICESat-2 logo Stijn compared new ICESat-2 data to data from the Dutch Atueel Hoogtebestand Nederland (AHN)

PDF icon

Validation of ICESat-2 data in The Netherlands
Stijn Lodder
BSc thesis, Applied Earth Science, Delft University of Technology, 2018

Bsc thesis project - Goof Blokker

ICESat-2 logo Goof assessed the possibilities for ICESat-2 to monitor glaciers in the Alps, with a focus on the Aletschglacier: goal is to see if ICESAT-2 can be used to monitor Alpine glaciers. To be able to answer this a case study on the Aletschglacier is performed, which is the largest glacier present in the Alps. Three subquestions are asked namely, does ICESAT-2 scan the Aletschglacier? How large of an area does ICESAT-2 scan of the Aletschglacier? And how often does ICESAT-2 scan the Aletschglacier? To see if ICESAT-2 scans the Aletschglacier the known orbits of ICESAT-2 are plotted over the Aletschglacier. Resulting in ICESAT-2 scanning the Aletschglacier in a total of 8 different orbits (called repeats by ICESAT-2), each repeat scans with three pairs of beams with a separation between the beams of 3.3km and between the pairs a spacing of 90m. To see how large of an area ICESAT-2 scans of the Aletschglacier a short python script is developed, resulting in 8.1% covered when all 8 orbits are done. How often ICESAT-2 scans the Aletschglacier is as of yet unknown since only the time specific orbits of the first repeat and halve of the second repeat are known. So to see which area’s would have the most data, cross overs of the different repeats are highlighted using QGIS. In short it is concluded that ICESAT-2 when it has been up and running for a couple of years might be a good way to monitor glaciers for change but in the short run it will not have enough data to monitor glaciers for change but it can give relevant height data.

PDF icon

Using ICESAT-2 to monitor Alpine Glaciers
Goof Blokker
BSc thesis, Applied Earth Science, Delft University of Technology, 2018

Bsc thesis project - Marije van Hell

Sentinel logo In a world of climate change, NASA predicts that the amount of category 4 and 5 hurricanes that form over the North Atlantic Ocean will increase not only in frequency, but also in intensity and duration. As a result more damage due to hurricanes is expected. In order for organisations to quickly start with emergency aid at areas that need it most, satellite images can be used. Category 5 hurricane Irma struck Sint Maarten on the 6 th of September 2017, leaving an almost completely destroyed island behind. Sentinel 2 optical images are retrieved from before and after the hurricane. The images are analysed using Google Earth Engine which is a free cloud-based platform that can process satellite data online using JavaScript coding and data on the Earth Engine server. First the Sentinel 2 optical data is retrieved and cloud masked. Training data is retrieved from the before picture and is used to classify both images. Also indices are calculated that highlight vegetation, sand and urban areas in the images. Difference plots are made by subtracting the after image with the before image. A reported widespread browning effect of vegetation appears to correspond to a great decrease in NDVI, indicating a great loss of plants. The NDGI together with the classification results show several areas where sand shifting has taken place, notably at the Prinses Juliana Airport and the North East shoreline of the island. It is discussed that optical imagery is not always a good method to indicate changes in urban area and it is recommended to additionally test Sentinel 1 radar data.

PDF icon

Detecting hurricane induced changes on Sint Maarten using Sentinel 2 optical data
Marije van Hell
BSc thesis, Applied Earth Science, Delft University of Technology, 2018


PDF icon

Indentifying glacial features with Sentinel-2 data
Najoua Essaf
BSc thesis, Applied Earth Sciences, Delft University of Technology, 2017

PDF icon

Core sample characterisation using 3D terrestrial laser scanning
Ivanka van Berkom
BSc thesis, Applied Earth Sciences, Delft University of Technology, 2017

MSc thesis project - Eva Stierman

Sentinel logo Chlorophyll (Chl) and Total Suspended Matter (TSM) are important water quality parameters characterizing the amount of oxygen and light penetrating the water. The Dutch Public Works department, Rijkswaterstaat (RWS), has been monitoring the water quality of the Dutch part of the North Sea for years. A research vessel samples parameters such as chlorophyll and TSM every few weeks at fixed locations. Recently, Sentinel-3 satellites started to provide satellite-based information on water quality. It is expected that products from the Ocean Land Colour Instrument (OLCI) sensor, on board of Sentinel-3, can map Chl and TSM at great detail. This study focusses on the validation of these Chl and TSM Sentinel-3A products. OLCI water quality roducts were directly compared to Rijkswaterstaat in-situ measurements from May until September 2017. In addition, the OLCI water-leaving reflectance and aerosol optical thickness data products were compared with observations from the Belgian AERONET-OC station Thornton. The water quality products of OLCI consist of a Chl product determined by the OC4Me algorithm and a Chl & TSM product derived from a neural network. OLCI Chl obtained from the OC4Me algorithm showed an overestimation of a factor 2 compared to the in-situ measurements. The Chl results of the neural network compared well with the in-situ measurements showing a correlation coefficient of 0.77. OLCI TSM showed an unrealistic underestimation of a factor 4 compared to in-situ measurements. The aerosol optical thickness data product showed unrealistic overestimations of OLCI compared to AERONET-OC. The spatial variability of OLCI's aerosol optical thickness is very high with differences of more than 40% per kilometre. In general, all products seem to have unrealistic values around clouds and in coastal areas. Therefore additional validation and algorithm improvement is recommended.

PDF icon

Remote sensing of North Sea water quality
Eva Stierman
MSc thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc thesis project - Leonoor Portengen

Sentinel logo Mangroves are forest ecosystems growing in (sub)tropical saline coastal environments. With their unique root structure they serve as important natural coastal protection. Despite all benefits mangrove forests are disappearing at alarming rates. This research focusses on Ca Mau, the southernmost province of Vietnam with mangroves present along the coast, in the Mui Ca Mau National Park and in mixed mangrove aquaculture farms. Remote sensing has been widely proven to be essential in mapping mangrove ecosystems. Previous research used either expensive optical and radar data sources or free but lower resolution systems. This study is the first that combines the new ESA Sentinel-1 radar and Sentinel-2 multispectral satellite missions that provide free available data with high spatial (10-20 meter) and temporal (10-12 days) resolution. Since optical data is prone to cloud effects and radar data is hard to interpret, both data sets are combined. The data is processed in the new online Google Earth Engine platform providing a powerful tool for big data applications like land cover classification. Optical data is found to separate mangroves by their spectral reflectance mainly in the near-infrared wavelength domain. The C-band radar signal is indicating the density of the canopy. To improve information extraction possibiloties from the radar signal temporal features are added. Seasonal variations are quantified and show an increase according to the spatial succession of mangroves. Pioneer species, such as Avicennia genus, show less seasonal variations than mature species, like Rhizophora. Five classes are defined: urban area, water and three mangrove classes. A classification method is set-up in Google Earth Engine with a Random Forest classifier using the satellite data inputs and ground truth training data of the five classes. A combination of the optical data with the temporal features of the radar data is found to be the best data input for separating those five classes. Classification results for discriminating mangrove types have an overall accuracy of up to 87\%. Comparing my land cover map of 2017 with a mangrove product of 2000 shows a regression along the southern coastline. No significant changes inside the shrimp farms are found between 2016 and 2017 but with the future availability of a long time series of Sentinel-1 and 2 data those can be detected with the method resulting from this study.

PDF icon

Classifying Mangroves in Vietnam using Radar and Optical Satellite Remote Sensing
Leonoor Portengen
MSc thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc thesis project - Stijn Pinson

Sentinel logo Understanding the physics and dynamics of glaciers is important in climate predictions and predictions of fresh water supplies. The Himalayas are the biggest storage of fresh water outside the polar regions with many different local climates and glaciers. One important glacier parameter used in dynamic studies is flow velocity of a glacier. The increased coverage of satellite missions provides the means to large scale velocity monitoring. Robust methods for measuring large scale seasonal and long term velocity dynamics of glaciers, however, remain an elusive goal.The optical Landsat mission is useful for long term flow velocity analysis. Feature tracking algorithms applied on these Landsat images provide means of automatically calculating large scale velocities. Automated approaches for large scale analysis are difficult because of shadows, surface changes and clouds and because the errors arising in the flow velocity calculations are often large compared to the velocities itself. This research proposes, implements and tests a new method for automatically creating large scale velocity time series using the optical Landsat database and feature tracking in the Himalayas. Where normally velocity time series consists of consecutive single velocity fields, the proposed method uses combinations of velocities to estimate these single velocities. This method is tested against results from single velocity fields for the Everest and Karakoram region. The main sources of errors: geo-location and precision of the feature tracking algorithm, are shown to have a large effect on the results. The magnitude of these errors make it difficult to measure seasonal changes in flow velocity for slow moving glaciers. The Landsat 8 velocity results for example, show good similarity to results from other researches and show promise for future research.The new method is not only useful for use on the Landsat database, but could be used to improve any optical or radar flow velocity time series.

PDF icon

Design and implementation of a novel method for creating glacial velocity time series for the Himalayan region using the optical Landsat database
Stijn Pinson
MSc thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc small thesis project - Adriaan van Natijne

Sentinel logo Laser scanners and Interferometric SAR both create point clouds but with different point density and position accuracy. Common web-based visualization of the low density InSar data and high density point clouds is expected to improve the interpretation of notably the InSAR data. Challenging for such visualization are the high data volumes of point clouds, the inhomogeneous coverage of laser measurements and the different coordinate systems involved and limited processing power of web-browsers. Using the PoTree octree structure implementation, a web application is built, suitable for researchers to create 3D visualisations for the greater public. All datasets were brought to the same coordinate system and are optionally enriched with other data such as aerial photographs and maps. Tiling was applied to limit downloads and processing exhaustion of the web-browser. Clustering of InSAR data may be applied to group points with similar behaviour while preserving unique data features. Results of this work are a demo application, a report and a manual on how to make a similar application based on a combination of existing tools. The visualisation at hand will allow for a new approach to InSAR analysis, integrating measurements with their 3D surroundings.

PDF icon

Web based visualization of 3D radar and LiDAR data
Adriaan van Natijne
MSc small Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

BSc thesis project - Esther Roosenbrand

Solar logo Solar panels are an effective way for sustainable energy generation. Solar panels are literally becoming more flexible, and can now be bended around a light pole, as the young Delft company Flex Solutions demonstrated. There is, however, one important issue with solar energy: it requires a minimum amount of sunlight to work properly! In this thesis the solar potential of a specific section of the TU campus, the Mekelpark, will be investigated. This will be calculated in three phases using both spatial and meteorological information. First, the amount of hours of sun at a certain location is calculated, according to the geometry of the surroundings. Then, using KNMI data, this calculation will be adjusted for average weather conditions. This will generate a model of the expected kW per day for an area of the Mekelpark. The amount of kW per day varies throughout the seasons, not only because the sun intensity varies, but also because the sun doesn’t reach evenly high throughout the year. In her thesis, Esther shows for example that the amount of sunlight at two closeby locations ranges from 42% to 88% of the maximum amount of sunlight.

PDF icon

Investigating the Solar Potential in Mekelpark
Esther Roosenbrand
BSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

BSc thesis project - Timo Bisschop

Sentinel logo With laser scanning or photogrammetry 3D models of rock outcrops can be made. For analysation and visualisation of such models the software Lime was created by the Virtual Outcrop Geology group in Bergen, Norwat. The aim of this thesis is to examine what functions Lime includes and how they can be used to visualise, interpret and measure the geologic features of an outcrop. To do this, a laser scan model was used of an outcrop containing limestone-marlstone alternations deposited during the Hauterivian age. This outcrop is located near the village of La Charce, in the Drome-Provencale, France. A method was developed to process raw laser scan data into a textured mesh that can be imported into Lime. This includes registration in the Leica Cyclone software, georeferencing and meshing in the free point cloud software CloudCompare and texturing in the free mesh software Meshlab. Functionalities that were examined in Lime include the use of texture functions, visualising the stratigraphic layers, measuring dip and dip direction of the formation, measuring layer thickness and importing a digital elevation model to show the location of the outcrop.

PDF icon

Virtual Outcrop Analysis using Lime
Timo Bisschop
BSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc thesis project - Philip van der Lugt

Sentinel logo Glacier fluctuations are regarded as a significant indicator of climate change. Changes in glacier can be observed with optical Earth observation satellites. New satellites, such as ESA's Sentinel-2A/B, provide high spatial resolution images and short revisit times. Clouds, however, present a challenge. Prior to glacier analysis, clouds must be separated from cloud-free pixels. Several techniques exist to automatically detect clouds but show sloppy results over snowy, icy and high relief landscapes. This study compares different cloud detection techniques. Six sentinel-2A scenes from the Indian Bara Shigri glacier acquired between July and December 2016 were processed using 5 different cloud detection techniques: Default L1C, Sen2Cor, Fmask, Temporal Averaging and Maximum Likelihood Classification. A manual mask was used as reference. All automatic cloud detection methods had there issues over mountainous terrain. Maximum Likelihood Classification was the most accurate automatic technique, followed by Sen2Cor. Cloud detection methods often misclassified mountain peaks and glaciers as clouds while at lower altitudes, cloud detection often misclassified clouds as cloud-free instead. The majority of snow and ice pixels are relatively easy to distinguish from clouds using Sentinel-2's band B3 and B11 and the NDSI feature. However, a minority of snow and ice pixels, particularly near cloud edges, are almost indistinguishable from clouds. Results strongly suggest that mountaintopography negatively affects cloud detection. This study came to the conclusion that cloud detection methods have the potential to improve, but currently are not yet strong enough to support glacier research.

PDF icon

An Assessment of Cloud Detection Methods Concerning High Altitude Snow and Glacial Environments With Sentinel-2
Philip van der Lugt
MSc Thesis, MSc Track Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc thesis project - Charis chatzikyriakou

Sentinel logo Drone systems are upcoming tools for flexible monitoring of dynamic processes at local scale. Typically a camera is used and often 3D information of the observed terrain is reconstructed from resulting photos using photogrammetric principles. This has some disadvantages though, as this procedure relies on ground control points, while photogrammetry has some problems with penetrating vegetation. Therefore the company of Shore Monitoring acquired a laser systems that could potentially solve these issues. In her MSc thesis project, Chara evaluated the Alpha AL3-32 UAV laser scanning system that was operated from a DJI Spreading Wings S1000 Octocopter. She considered the a priori and a posteriori error budget of the system and evaluated the quality of resulting digital terrain model.

PDF icon

UAV Laser Scanning for DTM Generation in Coastal Areas
Charis chatzikyriakou
MSc Thesis, MSc Track Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc additional thesis project - Auke Blokland

Sentinel logo After a train accident the site is measured for damage inspection and accident analysis. Traditionally this is done by on the ground surveys which is labour intensive and often dangerous in case chemicals are involved for example. An alternative is to use a drone with camera. In his work Auke investigates if quality requirements on measuring in the scene can be met by a 3D model obtained from such drone images using photogrammetric principles. In a practical test, the drone 3D model is compared to independent laser scanner data from such drone images using photogrammetric principles. In a practical test, the drone 3D model is compared to independent laser scanner data.

PDF icon

Usage of drones in the event of incidents on the rail-Measuring efficient and accurate
Auke Blokland
Additional MSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc additional thesis project - Dirk van der Valk

Sentinel logo In the Second World War Dutch New Guinea was a strategic battle front for Japanese and Allied forces in the Pacific War. A lot of airstrips were constructed and bombed, of which at least three (Mongosah, Otowari and Sagan) staid free from any activity after the war. This provided a great opportunity to find potential war heritage and airstrip equipment. An expedition is foreseen for in-situ heritage inventory. To do so, they need a classification map giving information on the type and location of vegetation. This map helps to know where to land with a helicopter, to setup base camp, to find travel ways, etc. Thus, the main objective of this thesis is to check whether it is possible to create a proper classification image with freely available data. Dirk used optical data from the Sentinel 2 mission, L-Band Radar data from the ALOS PALSAR mission and elevation data from the SRTM Mission. Dirk pre-processed the data and used Maximum Likelihood Classification. He masked clouds via three different cloud masking methods, the MLC Method, a Threshold Method and Sen2cor. Dirk compared the three different methods and found no significant differences. The classifications have been cross-validated with a reference validation dataset and the classification accuracy is on average 90 %.

PDF icon

Remote Sensing of Japanese WWII airstrips in the Papua Province Republic of Indonesia
Dirk van der Valk
Additional MSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

See also: Remote Sensing of Japanese WWII airstrips in West Papua on the Leiden, Delft, Erasmus Centre for Global Heritage and Development Website

BSc thesis project - Kathelijne Beenen, tree roughness

BSc thesis project - Jeras Dieleman

Sentinel logo Lava flows strongly affect the direct surroundings of volcanoes. Roads may be blocked and in the worst case complete villages are destroyed. At the same time, historical flows contain information on the processes causing them. Using modern mapping technologies such as digital photogrammetry and LIDAR, elevation data of ongoing and past lava flow can be easily obtained from different flying platforms. For his BSc thesis project, Jeras studied airborne LIDAR data of a past lava flow in Oregon. He shows that by estimating and subtracting the large scale topography it is possible to map the variations in roughness of the lava flow, which can consecutively be linked to physical processes.

PDF icon

Estimating lava flow roughness from elevation data
Jeras Dieleman
BSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

BSc thesis project - Jacintha Tjia

Sentinel logo Some plants move their leafs in a near daily cycle. The plants close at night and open up during the day. Using a terrestrial laser scanner it is possible to more precisely characterize the movement of such plants without interfering strongly in the experiment by exposing the plant to a lot of artificial light. In her BSc project, Jacintha notably studied the movement of the Calathea Roseo-Picta by scanning it repeatedly for two days, one day with the open curtains and one day in a closed room. Her results demonstrate the strong effect of sunlight on the leafs of this plant.

PDF icon

Circadian rhytms of plants by TLS
Jacintha Tjia
BSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

MSc thesis project - Nils Vollenhoven

nn logo

HTML icon Boreal forest change detection with the spaceborne Geoscience Laser Altimeter System
Vollenhoven, S.N.
MSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2017

Assessing drone LIDAR data of an intertidal area

MSc thesis project - Shelley van der Graaf

nn logo Often values of for example elevation are required at spatial locations where has not been measured. To estimate elevation a number of deterministic and stochastic interpolation techniques are available. Deterministic techniques are often fast, but do not incorporate spatial correlation into account. Ordinary Kriging is a well-known stochastic technique, but is potentially slow. In her thesis project, Shelley combines Voronoi diagrams and Ordinary Kriging. The Voronoi diagrams are used to determine the so-called natural neighbours of an interpolation location. Kriging is used to interpolate only these natural neighbours. Results indicate that the resulting natural neighborhood Kriging method often outperforms Ordinary Kriging, notably from a computational view.

PDF icon Natural neighbour kriging and its potential for quality mapping and grid design
Van der Graaf, S.C.
MSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2016

PDF icon

Spatiotemporal aeolian sediment transport variabilities at Oysterville (WA) beach
A.H. Verheijen
MSc thesis, Civil Engineering, Delft University of Technology, 2018

MSc thesis project - Elizabeth Prentice

Sentinel logo During fieldwork, geologists characterize rock outcrops to reconstruct geological history. To do so also geometric information is collected on notably the orientation of the outcrop, the orientation of recognizable layers in the outcrops, and the thickness, alterations and anomalies (faults and folds) in these layers. Traditionally, these information is extracted using ruler and compass, which is often a bit subjective and restricts the geologist to the accessible part of the outcrops. As an alternative, Elizabeth considers terrestrial laser scanning to extract these geometric information. She took a scanner to the Drome Provencale in France, scanned several outcrops and developed a workflow to automatically extract bedding planes from the collected point cloud data.

PDF icon

html icon

4TU icon
Characterizing Sedimentary Outcrops with Laser Scanning:
Applied to Cretaceous deep marine limestone-marl sequences in the Vocontian Basin, France
Elizabeth Prentice
MSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2016

Online trip report on Elizabeth's thesis fieldwork: Laser Scanning Rocks in the Foothills of the French Alps

3D Laser Scan Data: La Charce Sedimentary Outcrop, France in the 4 TU Data Repository

MSc thesis project - Isabelle de Lange

Sentinel logo There is some need to monitor objects in undeep water, like harbour infrastructure, underwater vegetation or sorting of riverbed deposits. But monitoring in shallow water is difficult. There is often physically not enough space for designated underwater sensors like echo sounding equipment. An alternative is to use optical sensors that look in the water. One possibility is to use a green light terrestrial laser scanner. Green laser is known to penetrate the water column to some extent. The Department of Geoscience and Remote Sensing has a `green` Leica C10 terrestrial laser scanner. In her MSc thesis project, Isabelles studies under what circumstances useful bathymetric measurements can be obrained with this scanner.

PDF icon

Observing through the water surface with a green-wavelength terrestrial laser scanner
Isabelle de Lange
MSc Thesis, Geoscience and Remote Sensing, Delft University of Technology, 2016

PDF icon

Next Generation Precise Satellite Positioning: Analysis of advanced positioning techniques with multiple navigation satellite systems using Curtin's PPP-RTK User Platform
Giulia Martini
MSc thesis, Track Geoscience and Remote Sensing, Delft University of Technology, 2016

BSc thesis project - Rianne Broeksma

Sentinel logo The Belgium Ardennes contains a sandstone formation that daylights at several former quarries. Terrestrial laser scanning is one technique that can be used to characterize the resulting outcrops. Rianne scanned the Durnal quarry with the Leica C10 scanner. In her thesis she describes a workflow to extract fracture planes and their orientation from the scan data.

PDF icon

Method for Detecting and Measuring Fracture Plane Orientation in the Condroz Sandstone with 3D Terrestrial LIDAR
Rianne Broeksma
BSc Thesis, Applied Earth Sciences, Delft University of Technology, 2016

BSc thesis project - Merve Sinem Günes

Sentinel logo Sentinel 1 and 2 data from ESA's Copernicus program provides world wide high repeat coverage at relative small pixel size. Sentinel 1 is a SAR mission, while Sentinel 2 is a spectral mission. Data products from both missions have the potential to give insight in the state of mountain glaciers. In her BSc thesis work Merve compares recent Sentinel 1 and 2 data over Mt. Ararat, a prominent vulcanoe in the East of Turkey, covered by somes glaciers.

PDF icon

Mapping the outline of the glaciers on the Mt. Ararat using Sentinel-1 and Sentinel-2 data
Merve Sinem Günes
BSc Thesis, Applied Earth Sciences, Delft University of Technology, 2016

BSc thesis project - Ruben Egbers

Sentinel logo Typically, in the warm season, the higher part of glaciers are still covered by snow, while at lower parts the bare ice is visible. These two parts of a glacier are separated by a snowline. The highest snowline in a season indicates the Equilibrium Line Altitude and its evolution throughout the years is used to assess the health of a glacier. In his BSc project, Ruben investigates the possibilities of using spectral Sentinal 2A data for identying snowlines and other glacial features over some glaciers in the Karakorum in High Mountain Asia.

PDF icon

Sentinel-2 data processing and identifying glacial features in Sentinel-2 imagery
Ruben Egbers
BSc Thesis, Applied Earth Sciences, Delft University of Technology, 2016

BSc thesis project - Bibi van der Horst

AHN logo Recently it has been shown that land cover can be estimated to some extend from LIDAR data. The Netherlands is currently sampled for the third time for the Dutch national archive `Actueel Hoogtebestand Nederland`. In her BSc thesis project, Bibi took AHN3 data sampling the island of Ameland as input and classified it into suitable classes. She considered both the 5m and 0.5m raster product and compared results to existing landcover products. Notably Bibi showed that classifications of remarkable detail can be obtained from the 0.5m data.

PDF icon

Landcover classification based on LIDAR elevation - On Ameland
Bibi van der Horst
BSc Thesis, Applied Earth Sciences, Delft University of Technology, 2016

BSc research minor - Tom Sassen

AHN logo Ameland and the adjacent Waddenzee is subject to change. Except for ordinary changes in e.g. vegetation, houses, roads and morphology, some subsidence is expected at the eastern part of the island due to gas extraction. In his study Tom Sassen compared different editions from the Dutch national laser archive `Actueel Hoogtebestand Nederland` and some dedicated laser altimetry datasets over the mudflats. In his report, Tom discusses methodology for and results of change detection, but also different issues with the available data.

PDF icon

Change of heights on Ameland - as detected from repeated airborne laser scan data
Tom Sassen
Final report BSc Research Minor, Applied Earth Sciences, Delft University of Technology, 2016

BSc thesis project - Jurjen Kamphuis

IGF logo Climate change is expected to induce local changes in vegetation. Changes in micro topography are monitored by the Institute of Mountain Research in Innsbruck, Austria. As part of his BSc thesis project, Jurjen acquired and processed multispectral data from a handheld camera at plot level at some mountain summits in the Alps. The camera collected images consisting of a NIR, RED and GRN band. Using Agisoft Photoscan orthophotos and texture classes were obtained from these images using a Structure from Motion approach. The NDVI of individual pixels was determined to enable a classification into vegetation or rock. The resulting 3D models seem to have local errors in the order of centimetres.

PDF icon

Vegetation detection in Structure-from-Motion derived 3D-models
Jurjen Kamphuis
BSc thesis, Applied Earth Sciences, Delft University of Technology, 2015

MSc thesis project - Dieuwertje Wesselink

Norut logo Avalanches form a threat to people and infrastructure in mountainous regions. Forecasting services are interested in knowledge on avalanche activity to verify their warning system. By applying Synthetic Aperture Radar (SAR), large areas can be monitored at once with both high spatial resolution and high acquisition frequency. Avalanche debris has an increased surface roughness compared to the surrounding unperturbed snow causing a higher backscatter signal. The main goal of this research project proposed by and performed at NORUT is to optimise avalanche detection in SAR images by exploring automatic detection of debris fields. Hence, we present a method to automatically detect avalanche debris fields in SAR images. It is designed and tested on both RADARSAT-2 Ultra Fine (UF) mode and Sentinel-1A Extra Wide swath (EW) mode images. For the RADARSAT-2 UF mode images the best result is obtained by using a median filter and a threshold value of 1.9dB, while for the Sentinel-1A EW mode images a so-called RSO filter in combination with a threshold value of 3.4dB resulted in optimum detection. None of the designed automatic detection methods resulted in 100% detection and zero false alarms, but they do confirm that automatic detection of avalanches in these SAR images is possible.

PDF icon

Optimisation of Method for Snow Avalanche Detection in SAR images: Supporting the development of snow avalanche mapping and monitoring of Svalbard (75MB)
Dieuwertje Wesselink
MSc thesis, Track Geoscience and Remote Sensing, Delft University of Technology, 2015

Bachelor thesis project - Adriaan van Natijne

ALS logo Shoals in The Oosterschelde in the South West of The Netherlands are loosing sand and are artificially strengthened by sand deposits. To monitor their dynamics, drone photogrammetry could be a flexible method that doesn't require people to actually access the mudflats. In his BSc thesis project Adriaan analyzes how a digital terrain model can be created from a large number of drone photo's. Notably the relative poor texture on the shoals is creating difficulties for the photogrammetric reconstruction of the terrain geometry.

PDF icon

Creation of precise shoal elevation models from UAV-photogrammetry
Adriaan van Natijne
BSc thesis, Technische Aardwetenschappen, Delft University of Technology, 2014

On monitoring shoals with drone photogrammetry

Bacheloor thesis project - Leonoor Portengen

ALS logo Glaciers on the Tibetan Plateau are suspect to change, but identifying and quantifying these changes is challenging. One possibility is to check for each year where on glaciers the bare ice is visible. A snowline, the line separating snow from ice, that is going up through the years, is an indication that a glacier looses mass. Leonoor developed methodology for identifying snowlines from Landsat images and analysed several glaciers in the Nyainqêntanglha Mountains in central Tibet.

PDF icon

Landsat derived snowline variations in Nyainqêntanglha Mountains on the Tibetan plateau
Leonoor Portengen
BSc thesis, Technische Aardwetenschappen, Delft University of Technology, 2014

Master thesis project - Peter Pietrzyk

ALS logo Forest trees are constantly changing because of e.g. growth, storms and human intervention. Novel high density airborne laser data has the potential to monitor such changes, but it is quite challenging to extract tree structure from a point cloud sampling a forest patch. Instead, Peter first extract points representing substantial change between scans from different years and next analyses if complete trees can be recognized from the extracted points.

PDF icon

Change Detection in Forests Using Multi-Temporal High Density Airborne Laser Scanning Data
Peter Pietrzyk
MSc thesis, Geomatics, Delft University of Technology, 2015

BSc thesis project - Michiel Cruijsberg

ICESat logo Recently, lake level changes and changes in glacier thickness were estimated from ICESat laser altimetry data over the Tibetan Plateau. But, ICESat measurements were sparse, so it is not obvious that the observed glaciers and lakes are in each others vicinity. Michiel analysed exactly which observed glaciers have their meltwater flowing in which observed lakes, and found several such cases. For some of these cases, Michiel performed a case study discussing possible correlations between glacial and lake level changes.

PDF icon

Connecting Tibetan lake level and glacial thickness changes
Michiel Cruijsberg
BSc thesis, Technische Aardwetenschappen, Delft University of Technology, 2014

scanner BSc thesis project - Adriaan Visser

The Dutch Oosterschelde is an estuary in the South-West of The Netherlands, whose natural dynamics got hampered after the building of the Oosterschelde Dam. Oyster reefs were therefore constructed to prevent the drowning of shoals. The effect of the oyster reefs is monitored by NIOZ using terrestrial laser scanning. In his BSc thesis, Adriaan shows how morphological changes can be detected from repeated laser scan data, and how, in addition, parameters of ripples can be estimated. The results indicate that the reefs have the anticipated effect.

PDF icon

Detecting morphology changes due to oyster reefs in a tidal basin using Terrestrial Laser Scanning
Adriaan Visser
BSc thesis, Technische Aardwetenschappen, Delft University of Technology, 2014

BSc thesis project - Marcella Schoenmaker

Strynevatnet is an isolated lake in Norway that used to be part of Nordfjord. About 8150y BP a huge tsunami, caused by the Storegga landslide, stirred up sediments in the shallow part of the lake and deposited in the deeper part. In her BSc project Marcella reconstructed the influx of sediment during the Holocene based on low frequency echosounder data acquired by TU Delft.

PDF icon

Mapping and interpretation of Holocene lake infill deposits
Marcella Schoenmakers
BSc thesis Technische Aardwetenschappen, Delft University of Technology, 2014

scanner Master thesis project - Hanne Vanhaekendover

For the construction of breakwaters large amounts of armourstone are needed which are mined in quarries. The better the estimation of the in-situ block size distribution (IBSD) of such quarry, the more accurately the prediction of the quarry production. In this thesis, 3D terrestrial laser scanning is proposed as an alternative to gather the discontinuity data of the rocks which is the major input to the ISBD estimation. The research is based on laser data gathered in the Dan-quarry in Benin. From the rock scans discontinuity planes are estimated that are expected to form the block faces after the blast. For estimating the sizes of the blocks a new so-called voxel method is developed, that basically counts how many voxels of fixed size are contained in each block. The results of the voxel method are compared to two more traditional methods that evaluate the position an orientation of the discontinuities along a so-called scanline.

PDF icon

In-situ block size estimation using 3D terrestrial data
Hanne Vanhaekendover
Master thesis Civil Engineering and Geosciences, Delft University of Technology, 2013

Master thesis project - Marcel Kleinherenbrink

Cryosat is a radar altimetry mission operating in three different modes. Over high relief terrain it operates in the so-called SarIn mode. In his thesis project Marcel studied the potential of Cryosat SarIn data for estimating lake levels over the Tibetan Plateau. Signals obtained over the lakes but near their shores are often poluted by topography of the surrounding mountains. To avoid their influence a special majority voting procedure was impleneted that estimates the most likely lake level from a series of consecutive signals.

PDF icon

Implementation and application of a numerical Cryosat retracker to retrieve lake level estimates on the Tibetan plateau
Marcel Kleinherenbrink
Master thesis, Delft University of Technology, 2013
Best MSc thesis award, Geoscience Depts. Research Meeting 2014

Master thesis - Jamie Morris argus

Our understanding of nearshore coastal zone processes can be improved significantly with the availability of high spatiotemporal resolution bathymetry data. Two algorithms that use remote sensing information to infer the corresponding bathymetric evolution are cBathy and Beach Wizard. cBathy models the temporal evolution of bathymetry by assimilating up-to-date bathymetry observations with a prior running average bathymetry. These bathymetry observations are determined using video-derived estimates of wave celerity via the linear dispersion relationship. Beach Wizard is a data-model assimilation scheme that works by comparing multiple sources of observed information with numerical model results. The bathymetry is progressively updated in time in order to make the observed and computed properties converge. Its main source of information is video-derived roller energy dissipation. The aim of the present study was to evaluate whether a combination of the two algorithms can improve our ability to estimate the nearshore bathymetry. Results demonstrated that a possible method of combining the sources is to use cBathy bathymetry estimates as Beach Wizard sources during calm conditions and roller dissipation maps as Beach Wizard sources during stormier conditions.

PDF icon

Estimation of the Nearshore Bathymetry using Remote Sensing Techniques: Combining Beach Wizard and cBathy
Jamie Morris
Master thesis Hydraulic Engineering, Delft University of Technology, 2013

Master thesis - Ilhame Oujamaa

The concept of a rapid flood model is a model which makes prediction of a flood event in 1) a short period with the use of 2) publicly available data 3) for any flood prone area in the world. This thesis evaluates the usability of a rapid flood model for river flooding and crisis management. This study consists of three parts. The first two, a 1D and 2D schematization, determine the most influential parameters and conditions of a flood for different types of river areas. Publicly available data is used for these most influential parameters and conditions to simulate the third study case, the 2011 Thailand flood event. For a relative flat area the type of land use is an influential condition, while in a mountainous area the river geometrics are the main influential parameters. The analysis gives an insight into which data the available resources, both time and money, should be spend on, if they are not publicly available. Over all, this evaluation shows that a rapid flood model can provide a reasonable estimate of a flood event in a short amount of time by using only publicly available data.

PDF icon

Evaluation of the usability of a rapid flood model
Ilhame Oujamaa
Master thesis Civil Hydraulic Engineering, Delft University of Technology, 2013

Diplomarbeit - Maximilian Streicher

Deltares constructs a new Delta Flume, a large wave flume facility (length = 230 m, width = 5 m, depth = 9 m) Hydraulic experiments near to full scale conditions can be conducted and waves with significant heights of Hs = 2 are generated. Together with the new facility, monitoring equipment will be installed. In this study the possibilities of Terrestrial Laser Scanning for determining water wave parameters were studied. Data sets obtained with a Faro, SICK and Riegl TLS device are analyzed with the purpose of wave field reconstruction and quality assessment of the results. In addition experiments with a SICK TLS device were conducted in the small wave flume facility of the TU Delft. Processing of the data gave insight in the possibilities and expected problems of the new measurement method.

PDF icon

Applying Laser Ranging for the Monitoring of Water Waves in the new Deltares Delta Flume Facility
Maximilian Streicher
Master thesis Hydromechanik und Kuesteningenieurwesen, Technische Universitaet Braunschweig, 2013

Bachelor Thesis - Leo Visser

Het digitaal simuleren van overstromingen is in de laatste jaren in grote mate geoptimaliseerd. Er is vanuit deze ontwikkeling de behoefte ontstaan om op een snelle manier een stroomsimulatie uit te voeren met hoogtegegevens, die vrij beschikbaar, werelddekkend en compact van formaat zijn. Het grid van de Shuttle Radar Topography Mission (SRTM) is er zo een. De vraag is, of dit grid nauwkeurig genoeg is om te gebruiken voor Rapid Flood Analysis. In dit onderzoek zijn gegevens van SRTM vergeleken met het fijnere grid van het Actueel Hoogtebestand Nederland (AHN). Er is vergelijking gedaan in de verticale verschillen in hoogte, en in de ‘smoothness’ van deze grids. De conclusie is, dat SRTM een grotere ruis dan AHN bevat, en in vlakke gebieden een grotere precisie heeft; van een fout van 10 meter in heuvelachtig terrein reduceert de fout naar 4 meter op vlak terrein.

PDF icon

Het gebruik van DEMs voor snelle stroomsimulaties
Leo de Visser
BSc thesis Civiele Techniek, TU Delft, 2013

Masterproef - Pieter Vandoren en Wouter Vandermast

Master thesis - Penelope Rammos argus

This thesis focuses on the ecological monitoring of an intertidal mudflat located in the Oosterschelde known as the Galgeplaat. It possesses an ARGOS monitoring platform standing 15 meters tall producing imagery that was previously used for monitoring the morphology of the mudflat. The goal is to examine the potential of this imagery for automated ecological monitoring of the mudflat. Two separate case studies were considered: 1. automatic detection of microphytobenthos, and 2. automatic detection of bird numbers. The main inhibiting factor for microphytobenthos detection was the presence of macroalgae in the images, which possess similar spectral properties to that of microphytobenthos. Two methods were used to detect microphytobenthos: I. maximum likelihood classification combined with the masking of the macroalgae (the undesired target) and II. Kohonen’s self organising maps (SOM). The results of this case study indicated that distinguishment between microphytobenthos and macroalgae was best achieved with the Self organizing map (SOM) approach. For the detection of bird numbers consecutive snapshot images of the camera were used such that the motion of birds could be taken advantage of. Background subtraction using a weighted mean background image and a standard deviation image was the most promising of the methods used to count the birds in the 20 frame video sequences.

PDF icon

Automatic detection of benthos & birds: Microphytobenthos cover and bird number detection on the Galgeplaat mudflat using terrestrial imagery
Penelope Rammos
Master thesis Civil Engineering and Geosciences, Delft University of Technology, 2012

Master thesis - Lennert van den Berg ICESat logo

Flow velocity is an important parameter for assesing the state of a glacier. Using repeated SAR radar images it is possible to estimate the local velocity variations in two essentially different ways. InSar processing decomposes the phase differences between repeated acquisitions into several components, one of which is deformation. On the other hand feature tracking determines correlation between features in repeated SAR images and consecutively maps moved features into displacements. Both techniques have been implemented and compared for the Petermann glacier in Greenland and the Siachen, the Gyachung and Rongbuk glaciers in the Himalaya-Karakorum.

PDF icon

Estimating glacial flow from repeated SAR images using feature tracking and InSAR
Lennert van den Berg
Master thesis Aerospace Engineering, Delft University of Technology, 2012

Masterproef - Jorn van Espen scanner

LIDAR (Light Detection And Ranging) is een techniek waarmee, gebruik makend van laserpulsen, de afstand tot een object gemeten wordt. Met behulp van LIDAR wordt ondermeer vanuit de lucht een terrein ingemeten. Deze techniek is gebruikt om zowel Nederland als Vlaanderen in te meten, resulterend in het Actueel Hoogtebestand Nederland (AHN) en het Digitaal Hoogtemodel Vlaanderen (DHM). In deze masterproef wordt onderzocht of het mogelijk is om wegen uit de AHN en DHM data te filteren. De hellingen die hiervoor gebruikt worden zijn zes bekende hellingen uit de wielersport, drie uit de Ronde Van Vlaanderen en drie uit de Amstel Gold Race. Er zullen enkele methodes gehanteerd worden om de hellingen in de puntenwolk terug te vinden. Vervolgends worden de gemiddelde en maximale hellingspercentages ervan berekend. Deze percentages zijn ook op verschillend wielersites terug te vinden. Zo kunnen de bekomen resultaten vergeleken worden. Uiteraard is er ook meer betrouwbaardere informatie nodig om de resultaten mee te vergelijken. Deze informatie wordt gevonden door zelf GPS-metingen uit te voeren. Ook wordt er vergeleken met metingen uitgevoerd door een professioneel landmeter. Er wordt ook geanalyseerd wat de verschillen en gelijkenissen tussen de AHN en DHM data zijn.

PDF icon

Bereken van hellingspercentages uit Amstel Gold Race en Ronde Van Vlaanderen met behulp van LIDAR data
Jorn Van Espen
Masterproef in de industriele wetenschappen: Bouwkunde, afstudeerrichting landmeten, Lessius Mechelen, Campus de Nayer, 2012

Master thesis Geomatics - Marjolein Koudijs KNMI logo

This study evaluates the potential of the Ice, Cloud and Elevation Satellite (ICESat) with the onboard Geoscience Laser Altimeter System (GLAS) instrument for the monitoring of water levels of rivers. The research question in this study considered the development of a method to derive water level estimations and eventually water level trends in the Mekong River from available GLAS elevations. For this purpose a workflow is developed and validated that identifies those GLAS elevations that are reflected by the water in the Mekong river. Furthermore, the strengths and weaknesses of GLAS altimetry data for hydrological monitoring of water levels in the Mekong River are evaluated

PDF icon

Using ICESat/GLAS laser altimetry for water level estimations in the Mekong River
Marjolein Koudijs
Master thesis Geomatics, Delft University of Technology, 2012
Winnares Marina van Damme Grant, 2014

Master thesis Aerospace Engineering - Edwin Jongkind KNMI logo

Water vapour in the atmosphere is a key variable in the prediction of heavy rain and thunderstorms. Water vapour in the atmosphere is still badly modeled although more and more ground based, airborne, as well as spaceborne measurements are retrieved. The main input for current water vapour maps are Integrated Water Vapour (IWV) measurements obtained every 15 minutes from GPS ground stations. My graduation project focuses on the question whether the combination of water vapour measurements and wind measurements can increase the quality of these water vapour maps. The increase in quality is expected as wind is moving the water vapour, also called advecting. Therefore in addition, 3D wind data obtained from HIRLAM is incorporated in the map making. HIRLAM is the abbreviation for High Resolution Limited Area Model for Numerical Weather Prediction, and this model is created by a European consortium of meteorological institutes. A modified Kriging interpolation approach is used to combine both current measurements and measurements from the past advected by the wind. By using a case study it is investigated whether the IWV prediction in between the GPS ground stations could be improved. For this case study, and using this method, we can conclude that on average the IWV maps are not improved by incorporating advected measurements. However, there are also locations for which the IWV predictions are improved.

PDF icon

Integrated Water Vapour mapping using vertical Wind Profiles
Edwin Jongkind
Master thesis Aerospace Engineering, Delft University of Technology, 2011

DTM special assignment - Martin Valk ICESat logo

The GLAS laser range instrument on board of the ICESat satellite obtained along track elevations between 2003 and 2009. In this case study, changes in elevation over Tibet for four different terrain types are considered. Lake (Nam Tso), City (Lhasa), River (Brahmaputra) and Mountains. Martins conclusions are that elevation changes over lakes and rivers are in this case well interpretable. Over the city and moutains, results are less clear.

PDF icon

Interpreting elevation changes over Tibet
Martin Valk
Case Study Digital Terrain Modeling, Geomatics course, 2011

Master thesis project - Stijn Verlaar ALOS logo

Pavescan is a low cost mobile system for road survey. Because of the absence of navigation sensors it has several practical drawbacks compared to most other mobile mapping systems, but those systems are expensive and do not fulfil all accuracy requirements. Pavescan will be more attractive if some of the practical drawbacks are reduced. Pavescan measures road profiles across the road by laser scanning at a series of positions. The separate scans are linked via control points, which have to be measured seperately. The accuracy in height of the scans is most important (e.g. for volume calculations) and has to be in the range of millimetres. It is evaluated if integrating close range photogrammetry can achieve millimeter accuracy of the scan and reduce the number of control points. A test survey was conducted with a downward looking camera mounted close to the scanner at 3.6 metres above the road. The position and orientation of the camera at each exposure is retrieved by bundle adjustment. The accuracy of the bundle adjustment depended strongly on the configuration of the control points. Due to the narrow shape of a strip of images, this configuration will always be weak. For a sequence of 28 images ( 43 metres), around five control points are needed to achieve sub-centimetre accuracy of the object points. If more than 36 images or fewer than four control points were used, no bundle adjustment solution could be obtained. A reduction of control points is therefore not possible. The practical feasibility for integrating close range photogrammetry into Pavescan is low, since too many control points are needed that should spatially be well distributed and measured with tachymetry. Therefore it is not recommended to integrate photogrammetry in Pavescan.

PDF icon

Evaluation of close range photogrammetric support for Pavescan
Stijn Verlaar
Master thesis Geomatics, Delft University of Technology, 2010
( Warning: PDF file = 16 MB)

Master thesis project - Hang Yu ALOS logo

The glaciers on the Tibetan plateau feed main Asian rivers on which hundreds of millions of people depend. As the glaciers are suspect to melting, it is important to monitor them. Over the remote Tibetan glaciers only satellite remote sensing techniques are potentially able to provide full coverage. In her research, Hang considered the use of ALOS/PRISM stereo satellite photos for constructing a Digital Elevation Model of a glaciated area. If repeated elevation models of sufficient quality could be constructed, their difference could give insight in local changes in glacial mass balance. Hangs results indicate that using standard software it is possible to create a DEM but that the quality is still hampered by i) unsufficient ground control points, ii) lack of texture on glaciers and iii) lack of visibility and shadow effects in mountain valleys.

PDF icon

Monitoring Glacier Elevation Changes Over the Tibetan Plateau Using ALOS PRISM and ICESat
Hang Yu
Master thesis Geomatics, Delft University of Technology, 2010
( Warning: PDF file = 7 MB)

Master thesis project - Maja Bitenc Rijkswaterstaat logo

The Dutch coast typically consists of a sandy beach followed by dunes, partly covered by marram grass. Monitoring and maintenance of the coast is essential in protecting the Dutch hinterland from the sea, but the beaches are also intensely used for recreation. Allthough it is known that the beaches especially suffer erosion during heavy storm events, still a good method for a fast damage inventory does not yet exist. In this project Maja has investigated if a laser mobile mapping system (LMMS) is suited for this job. A LMMS consists of a number of line scanners mounted on the roof of a car, together with a GPS for positioning, and additional sensors to correct for the motion of the car. Using a LMMS it is possible to quickly obtain georefereced xyz data of a large stretch of beach. In her thesis Maja describes an analysis of data from a pilot project organized by The Dutch Ministery of Public Works. Both the quality of individual laser points and of a derived Digital Elevation Model is analysed.

PDF icon

Evaluation of a laser Land-based Mobile Mapping System for measuring sandy coast morphology
Maja Bitenc
Master thesis Geomatics, Delft University of Technology, 2010
( Warning: PDF file = 44 MB)

Master thesis project - Bas van Goor Gemeentewerken logo

The city of Rotterdam aims at improving the accessibility of its city centre by public transport. Part of this plan is a new light rail connection between The Hague and Rotterdam that will be linked to the existing Rotterdam metro net below Rotterdam Central Station. To achieve this, the number of tracks is increased from two to three and the metro station itself will be rebuild. The effect of the stress of the construction work indiced on the metro staton is carefully monitored using tachyometry and inclino- and joint meters. In his master thesis Bas considers how in addition terrestrial laser scanning can be used to monitor the effect of the construction works. He had three data sets available from different times, and acquired a fourth data set himself. Main focus was the design of a procedure to make the scan data from the different epochs comparable. For this purpose identical objects, like parts of walls, in different epochs are identified. If identification is indeed possible, it can be considered if small deformations took place; Failed indentification, on the other hand, is caused by changes, occlusions or edge effects. The new procedure is demonstrated on the tunnel data.

PDF icon

Change detection and deformation analysis using Terrestrial Laser Scanning - Case study of the metro tunnel at Rotterdam Central Station.
Bas van Goor
Master thesis Geomatics, Delft University of Technology, 2011
( Warning: PDF file = 55 MB)

Case Study project Space - Simon Billemont ICESat logo ASTER

In 2009 the global digital terrain model ASTER GDEM was released. In this case study Simon evaluates the official quality claims by comparing GDEM elevations to elevations obtained from the GLAS lasers onboard of the ICESat satellite. This validation takes place over Tibet and both flat and high relief terrain is considered.

PDF icon

Validation of ASTER GDEM over Tibet
Simon Billemont
Case Study Report BSc Space, Fact. of Aerospace Engineering, 2010

Master thesis project - Bas van Son

Coastal monitoring: Bathymetric surveys of an area of 1 square kilometer of the shoreface near Ter Heijde have been perfomed using the NEMO Jetski. Instruments installed are a Single Beam Echo Sounder for depth measurements and accurate GPS for positioning. The quality of the Jetski as a platform for bathymetric surveys has been assessed. The local morphology shows a shoreparallel subtidal bar, situated 500m offshore and around -4m NAP. The position of the bar is monitored within weeks before and after storm events between September and December 2008. The bar has migrated approximately 10m onshore during a calm weather period. In a stormy month the bar migrated 25m - 30m offshore.
Coastal modeling: The numerical area model Delft3D is set up to reproduce the morphological changes as monitored in the field. Wave conditions from the offshore Europlatform buoy were used. Model results show offshore bar migration during storm as well as some flattening of the bar crest.
Synthesis: Theory and observations described in literature show agreement with the field observations at Ter Heijde. They show onshore bar migration during fair weather conditions, whereas the bar moves offshore during storm conditions. The main onshore driving component is wave asymmetry whereas the main offshore driving component is undertow.

PDF icon

Monitoring and modeling nearshore morphodynamic behaviour on storm time scales
Bas van Son
Master thesis Civil Engineering, Delft University of Technology, 2009

Master thesis project - Ramses Molijn

ICESat CSR Since 2003 the Ice, Cloud and Elevation Satellite (ICESat) is obtaining elevations of unprecedented accuracy and coverage over notably Antarctica and Greenland. ICESat does so by recording the time of flight of laser pulses send from the satellite, reflected by the Earth's surface and received back by the satellite. In fact, it not just records one moment, but captures the full return signal as a function of time, the so-called full waveform. For his thesis project Ramses has investigated how ICESat full waveform parameters can be used to obtain information on the terrain hit by a laser pulse. It turns out that to some extend the full waveform signal can be used to discriminate between pulses reflected by ice, rock, water and snow. His master thesis project included a three month internship at the Centre for Space Research of the University of Texas, where he got a view in the `kitchen` of ICESat and obtained insight in the many different parameters that are all influencing an ICESat elevation measurement.

PDF icon

ICESat full waveform signal analysis for the classification of land cover types over the Cryosphere
Ramses Molijn
Master thesis Geomatics, Delft University of Technology, 2009
( Warning: PDF file = 64 MB)

Master thesis project - Boudewijn Possel

Fugro logo Boudewijn will use airborne laser altimetry data as organised in the Dutch AHN database and as provided by Fugro FLI-MAP to assess properties of the shallow subsurface. His method consists of three steps: i) filtering, i.e. the removal of points representing roads, houses, ditches etc. ii) segmentation, i.e. the subdivision of remaining points in ancient channel points and other points, and iii) classification, which includes an attempt to classify the found channels according to age and/or type.

PDF icon

PDF icon

PDF icon

GIF icon
Automatic detection of buried channel deposits using high resolution laser altimetry data (FLI-MAP)
Boudewijn Possel
Master thesis Geomatics, Delft University of Technology, 2009
( Warning: PDF file = 18 MB)

Het karteren van oude rivierbeddingen
Eric van Rees
GiS Magazine, Oktober/November 2008-7

Mapping of ancient river channels
Eric van Rees
Geo Informatics, June 2008

Mapping ancient tidal and river channels with FLI-MAP and AHN laser altimetry
B.M.J. Possel, R.C. Lindenbergh, J.E.A. Storms and M.P. Kodde
Poster for the 9th Nederlands Aardwetenschappelijk Congres, Veldhoven, 2008

Geomatics Synthesis Project - 2008

Boskalis logo The Delfland shore near Monster-Ter Heijde is a hot spot in the Dutch coastal defence system. Only a narrow dune ridge is protecting the hinterland from potential flooding. Three projects are under development that aim at reinforcing the coastal system while maintaining and even extending the natural environment. Royal Boskalis Westminster NV, an important participant in these projects, asked the Geomatics students to measure the initial situation at the Monster-Ter Heijde coast and to propose a measurement strategy for monitoring the projects progress. For the purpose of mapping both topography and surface properties, the students tested kite areal photogrammetry, terrestrial laser scanning, echo sounding from a jet ski and grain size mapping based on natural radioactive background radiation. The results of the measurement campaign enabled the students to propose a monitoring strategy taking into account three types of coastal processes: storms, sand nourishments and gradual natural change.

PDF icon Final Report
Novel techniques for coastal monitoring, a case study near Monster - Ter Heijde
(Warning: PDF file = 64 MB!!!)

Design and Synthesis Project, 2008

ICESat Since 2003 the ICESat laser altimetry mission obtains elevations all over the world by its main payload, the GLAS full waveform laser system. ICESat has a near polar orbit and its data has lead to many new insights notably on ongoing surface processes on Greenland and Antarctica. To quantify ongoing changes in the mass balance of the major ice sheets it is important that the time series that were started by the ICESat measurements will be continued. This is the main purpose of the POLESat mission design. The design group has considered three mission concepts to improve both on the accuracy and on the spatial resolution of the ICESat mission: multi-satellite, split laser and multi-laser. It was decided by the group that a setup with five lasers in array gave the best results. In their final design it is shown that it is possible to design POLESat such that it obtains 40 times more elevations per campaign as ICESat. It should be noted however that the orbit configuration that is presented as having no eclipse in fact has an eclipse in practice.

PDF icon

PDF icon
Final Report
POLESat - Pole Oriented Laser Elevation Satellite
Group 13
Third years Design and Synthesis Project for Aerospace Engineering, Delft University of Technology, 2008


Minor Earth observation, case study, 2007

ICESat The GLAS instrument on board of the ICESat satellite obtains full waveform profiles over footprints on the Earth surface of 70m diameter. In this case study the waveform of a particular footprint in a forested area near IJmuiden, The Netherlands is compared to two other data sets obtained by laser ranging. First, the approximate footprint location was scanned with the FARO LS 880 terrestrial laser scanner (TLS) by the authors. After considerable processing, a synthetic TLS waveform could be constructed. Second, a synthetic waveform was constructed from airborne laser data from the AHN national Dutch laser project. The resulting waveforms are compared and discussed.

PDF icon
Reconstruction of an ICESat Full Waveform by Terrestrial and Airborne Laser Scanning
K.H. Spaans and A.L.A.B. Ronse
Case study, third year minor Earth Observation, Delft University of Technology, 2007

Master thesis project Cornelis Slobbe

ICESat Gravity The Greenland ice sheet is sensitive to climate change. Global heating is expected to result in ice mass losses that will contribute to global sea level rise. For this reason monitoring Greenland's ice mass balance is of utmost importance. Data of both the Ice, Cloud, and land Elevation Satellite (ICESat) laser altimetry mission and the Gravity Recovery and Climate Experiment (GRACE) gravity mission are used to create two independent estimates of Greenland's ice sheet mass balance over the full measurement period of about 2003 until 2007. Further improvements are possible when both datasets are combined in a joint inversion. In his master thesis, Cornelis provides the reader with a detailed overview of all the steps leading towards a joint inversion. Special attention is paid to the accuracy of the final results, given the uncertainty in the many assumptions that had to be made.

PDF icon Towards a combined estimation of Greenland's ice sheet mass balance using GRACE and ICESat data
Cornelis Slobbe
Master thesis Geomatics, Delft University of Technology, 2007.
(Warning: 25 MB)

Master thesis project Steven Pluymakers

Rijkswaterstaat logo Steven has analysed a unique Multibeam Echo Sounding data set of the Westerschelde near the village of Walsoorden. The data set consists of seventeen epochs of echo sounding data, sounded during and shortly after a large sand deposit. Steven used the Delft method of deformation analysis to outline the deposit location and to track the deposed sand over time. A classification method, based on underwater bedform parameter values, enabled Steven to divide the estuary floor in homogeneous regions. The research was performed in cooperation with the Dutch Institute for Coasts and Sea, RIKZ.

PDF icon A data analysis to bed dynamics in the Western Scheldt estuary New!
Steven Pluymaekers
Master thesis Geomatics, Delft University of Technology, 2007.

Case study, Benjamin Vis

Benjamin studies Archeology in Leiden. He is interested in incorporating remote sensing techniques in his archaeological studies. As a first introduction to data processing, Benjamin analysed AHN laser altimetry data. The purpose of the case study was to identify prehistoric burial mounts from the data in a forested area. Unfortunately this turned out to be very hard, probably because of inpenetratable forest canopy.

PDF icon
Report on the use of Matlab in analysing LiDAR data
Benjamin N. Vis
Case study, Delft University of Technology, 2007

Minor Earth observation, case study, 2006

ICESat ICESat laser altimetry data is a potential source for assessing land cover information. In this paper it is investigated how to obtain a database of crossover footprints from the ICESat laser altimetry data. The returned waveforms from these crossover footprints can then be used to detect land cover change. A method for finding these crossover footprints is introduced. Also waveform analysis is performed. A database has been obtained, using data from five different operation periods of ICESat containing 25847 pairs of crossover footprints in total. Nevertheless, there are still crossover footprints missing. Also there has been a contamination in the crossover footprints database with repeated track footprints.

PDF icon
ICESat full waveform database for land cover change detection over Europe
A.A. Kestila and T.S. van der Veen
Case study, third year minor Earth Observation, Delft University of Technology, 2006

PDF icon

Glacier Surface Analysis. Airborne Laser Scanning for monitoring glaciers and crevasses
Martin Kodde
MSc thesis, Geomatics, Delft University of Technology, 2006

Master thesis project Jane van Ree

Fugro logo Jane designed and executed four tests to compare the performance of terrestrial laser scanners. The tests took place in a long corridor in the basement of the Aerospace Engineering building. During her experiments she considered the influence of range, angle, colour and reflectivity on the quality of laser scan observations. It turned out that the influences of range and angle are as predicted, but that the influence of colour and reflectivity is more complicated.

PDF icon Determination of the precision and reliability parameters of terrestrial laser scanners
Jane van Ree
Master thesis Geomatics, Delft University of Technology, 2006.
(Warning: PDF file = 4 MB).

Design and Synthesis Project, 2006

In 2004 the Sumatra tsunami caused 260.000 casualties all around the Indian Ocean. A timely Tsunami warning could have decreased the impact of this disaster. GNSS-R is a technique for observing reflections from GPS signals from the Ocean's surface. In this project it is studied if it possible to design a Spaceborne Tsunami Warning System based on the analysis of the GNSS-R reflections that provides world wide coverage. In more detail a first design of a demonstrator satellite is given.

PDF icon

PDF icon

PDF icon
Final Report
Space-Borne Tsunami Warning System
P.A.I. Brouwer, M. Visser, M. Baldee, J.C.T. Gunneman, B. Masselink, R.A. Molijn, H.M. Jara Orue, B.J.A. van Marwijk, T.C.K. Bermon
Third years Design and Synthesis Project for Aerospace Engineering, Delft University of Technology, 2006


Artikel in Algemeen Dagblad, d.d. 23 juni 2006 PDF icon
Tsunami speuren vanuit de ruimte

Internship Niels van der Zon

Gemeentewerken logo Niels worked at the survey department of the City of Rotterdam on increasing the possibilities to integrate terrestrial laser scanning data in the daily workflow. For this purpose he wrote a practical manual on the use of the Cyclone software for processing laser point clouds and contributed to a first laser scan measurement of the metro-tunnel below the Rotterdam Central Station.

PDF icon Verbeteren van het gebruik van Cyclone voor de verwerking van laserscandata
Niels van der Zon
Internship Geomatics, Delft University of Technology, 2006.
(Warning: PDF file = 6.5 MB).

Case study Geostatistics Astrid Humme

Vestigia logo Celtic fields are prehistoric agricultural field systems that are recognisable as small rectangular patches of land surrounded by low earth walls. Vestigia found some Celtic field systems in The Netherlands by illuminating airborne laser altimetry data in a suited way. In her case study, Astrid analyses whether it is possible to enhance such visualisations by means of filtering the large scale topography from the data set at hand.

PDF icon Revealing Celtic Fields near Doorwerth with Ordinary Kriging
Astrid J. M. Humme
Case study Geostatistics for Earth Observation course, 2006

AE-002, 2nd year project `Flight Mechanics, Earth Observation and Orbital Mechanics'

Waterstaat logo The seafloor of the Westerschelde estuarium is covered by bedforms like shoals, sand waves and mega ripples of different size and characteristic. Goal of the project is to develop a method for automatic determination and visualisation of the orientation, amplitude and asymmetry of the bedforms near the village of Walsoorden. For this purpose a MBES dataset was used, made available via the Rijks Instituut of Kust en Zee.

PDF icon Visualiseren van wandelende zandgolven in de Westerschelde
S. Bakker, D. Cleij, M. Groen, B. Hummelink, J. Leloux, D. Rogiest, R. Vandeberg en H. Yildizturan
Eindverslag tweedejaarsproject Lucht- en Ruimtevaart, TU Delft, 2006

Master thesis project Rinske van Gosliga

Gemeentewerken logo At the end of 2005 the build of a new tunnel has started, connecting the Rotterdam central railroad station to the Hofplein railroad line. Gemeentewerken Rotterdam is considering to monitor the construction process by means of terrestrial laser scanning. In her master thesis Rinske has investigated whether the use of laser scanning is feasible for tunnel deformation monitoring. For this purpose she has developed a scanning and deformation analysis method for detecting changes in the interior tunnel surface. She tested her method on scan data of artificial deformations in the second Van Heinenoordtunnel.

PDF icon Deformatieanalyse van een geboorde tunnel met behulp van terrestrische laserscanning
Rinske van Gosliga
Master thesis Geomatics, Delft University of Technology, 2005.
(Warning: PDF file = 8 MB).

3rd year internship Lidia van Halderen

Waterstaat logo Bij Rijkswaterstaat worden zowel de interpolatiemethodes DIGIPOL als SURFIS ingezet voor het interpoleren van dieptemetingen. Een andere veel gebruikte interpolatiemethode is Kriging. In dit onderzoek voor de Adviesdienst Geo Informatie heeft Lidia de interpolatie resultaten en nauwkeurigheden van deze drie methodes vergeleken. Daarvoor heeft ze een MBES dataset van de Petroleumhaven in Amsterdam gebruikt, waaruit ze verschillende configuraties van raaien heeft geselecteerd, die vervolgens dienen als input voor de drie methodes. Op deze manier kunnen de interpolatie resultaten gevalideerd worden met de oorspronkelijke MBES data set.

PDF icon Vergelijking tussen de interpolatiemethodes DIGIPOL, SURFIS en KRIGING
Lidia van Halderen
Stageverslag, Technische Universiteit Delft, Rapportnummer AGI-2005-GPMP-015, 2005.
(Warning: PDF file = 16 MB).

Geostatistics for Earth Observation

During spring 2005 I taught the master course Geostatistics for Earth Observation for master students in Geodesy and in Earth Observation. The students were divided in groups of two and had to weekly hand in worked out assignments. For the assignments every group worked with two `private' data sets, one provided by me and one found by the students themselves. For solving the assigments the students could sometimes use predefined Matlab command but often new programs had to be written.

PDF icon Geostatistics for Earth Observation, Master course Geodesy/Earth Observation
Roderik Lindenbergh and Ramon Hanssen.
Delft University of Technology, 2005.
Version: spring 2005.

AE-002, 2nd year project `Flight Mechanics, Earth Observation and Orbital Mechanics'

De Waddenzee bestaat uit modderplaten die doorsneden worden door kanalen. Tijdens de eb- en vloedcyclus transporteren de kanalen water tussen de Noord- en Waddenzee. Gedurende de jaren verandert de positie van de kanalen, bijvoorbeeld doordat de kracht van het water de kanaalranden afschuurt. Doel van dit project is om uit dieptedata van de Waddenzee de kanaalstructuur te bepalen om daarmee de kanaaldynamiek door de jaren heen te kunnen volgen.

PDF icon De kanaalstructuur van de Waddenzee
M. van den Berg, R. Brandt, A. van der Eijk, J. Lemmens, A. Oyen, R. Sopjes en H. Vrijkorte
Eindverslag tweedejaarsproject Lucht- en Ruimtevaart, TU Delft, 2005
(Opmerking: 1 plaatje verloren gegaan bij word→pdf conversie, PDF file = 3 MB ).

Master thesis project Peter Menting

Waterstaat logo Both the Dutch North Sea Directorate (Directie Noordzee) and the Hydographic Service of the Royal Netherlands Navy (Hydrografische Dienst) use time series of echo sounding measurements to predict changes of the sea bottom depth. But the methods used are at first sight quite different. The North Sea Directorate uses a Kalman filter approach, while the method of the Hydographic Service is based on geodetic deformation analysis. Hydrografische dienst logo Peter Menting compared the performance of both methods on equal data sets, some consisting of `real' data and other of simulated data. Based on these experiments he proposed a combined method profiting of the strengths of both approaches. The combined methods give promising results when applied on the test data sets.

PDF icon Detection and prediction of sea floor dynamics
Peter Menting
Master thesis Geodesy, Delft University of Technology, 2004.
(Warning: PDF file = 5 MB).

Eindrapport onderzoeksvaardigheden Rinske van Gosliga

Rinske van Gosliga heeft voor het vak `Onderzoeksvaardigheden' gekeken of het mogelijk is om locatie, tijdstip en volume van zandsuppleties te reconstrueren uit een tijdreeks van hoogtedata. Met behulp van een laser gemonteerd aan een vliegtuig is het zuid-westen van Texel herhaaldelijk in kaart gebracht in de periode 1996-2001. Het is bekend dat in deze periode ook minstens één zandsuppletie heeft plaatsgevonden op het strand om te compenseren voor de afslag door de zee. Rinske heeft de suppletie gemodelleerd en de laserdata getest met behulp van geostatistische methoden.

PDF icon Volumebepaling van zandsuppleties op Texel aan de hand van tijdreeksen van laseraltimetrie data.
Rinske van Gosliga.
Eindrapport Onderzoeksvaardigheden, Studie Geodesie,
Technische Universiteit Delft, 2004.

Master thesis project Natasha Hennis

Fugro logo Natasha Hennis worked for six months at Fugro Intersite B.V. on a method to automatically remove outliers from data sets of multibeam data. An existing method uses Kriging in one direction to obtain an interpolated value for a sounding to be tested. The difference of the interpolated value and the observed value has to meet a certain test criterium in order to be accepted. A main drawback of this 1D method is that it removes not only outliers but also small objects like pipes. Natasha investigated if a similar approach using Kriging in two directions would avoid this unwanted removal of pipes and found out that in most cases she could avoid at least 40% of the false outliers.

PDF icon Automatic Outlier Detection in Multibeam Data.
Natasha Hennis.
Master thesis Geodesy, Delft University of Technology, 2003.
(Warning: PDF file = 3 MB).

Mathematica notebooks for Utrecht University College

Matica logo In 1998, the Utrecht University College started, providing a general `Science' bachelor program. The Mathematical Institute was asked to organize a course `Mathematics in Context' for the first year Science students of the College. It was decided that the course would consist of both lectures and computer aided exercise classes. Worksheets for the exercise classes were prepared by and for the Computer Algebra System Mathematica. The latest worksheets I worked on date from Spring 2001 and can be found by following the link below.

PDF icon Mathematics in context, Mathematica notebooks for Utrecht University College
Odo Diekmann, Roderik Lindenbergh, Martijn van Manen et al.
Utrecht University 1998-2001.
Version: spring 2001.

Roderik Lindenbergh
January 2007