Bachelor students of Earth, Climate & Technology (EC&T) at the TUDelft have to do post-mining risk management assignments in the old mines in Valkenburg as part of second-year courses. They will do this by inspecting and assessing the stability of underground structures. To achieve this, a virtual reality application is created, which will be part of a larger project called VROCK. For this, 3D models have been obtained in the Valkenburggroeve and the Plankertgroeve. Selected points of interest are scanned, which are often large pillars. The points of interest are scanned using two LiDAR scanners on iPhones. An app called Scaniverse is used for scanning, which gave 3D mesh models as output. These scans result in models with a high mesh density, and thus they need to be optimized to be computationally light enough for virtual reality. This is solved using quadric error metrics using edge contraction. This method ranks every edge of the 3D mesh by the error caused by optimization, and then contracts these edges based on this ranking. The optimized models are then compared with the original versions to assess their quality as well as comparing how other scan ranges and processing modes change the optimized results. It is found that a scan range of 5 meters with the detail processing mode result in the most suitable models for virtual reality, as they show the highest texture quality with the lowest polygon count. The new optimized models are prepared for the final virtual reality application, which is made in the Unreal Engine 5.3 game engine. Multiple methods such as Level of Detail, as well as the types of levels used and lighting placement and lighting type are used to increase performance as much as possible. The XRZone at the TUDelft Library will create the full application with my processed models; however, to assess performance, a small virtual reality demo is made for players to get a feeling of the final result. A theoretical limit of 1,125,000 rendered polygons is assumed based on the virtual reality hardware; however, the results show that at most 465,478 polygons were rendered. Based on this, the performance is very well, as older hardware should also be able to work with the amount of polygons being rendered. Additionally, a survey is held to assess the quality of the game demo. Players are asked about different light intensities, motion sickness as well as if they prefer three smaller levels instead of one large level. The results of this are overall very positive, with almost all players preferring three levels instead of one.

Coastal environments are vital for ecological stability, human activity, and climate resilience, yet they are increasingly affected by anthropogenic activities. Particularly, construction machinery such as bulldozers plays a critical role in altering the beach environment through beach nourishment purposes and coastal engineering, but their presence and movement are rarely tracked systematically. This research addresses that gap by developing a method to automatically identify bulldozers and other large dynamic objects from multiple epochs of permanent terrestrial laser scanning (TLS) point cloud data using multidimensional feature analysis and supervised machine learning.

This study presents a robust framework for automatically classifying bulldozers and other large dynamic objects from terrestrial laser scanning (TLS) point clouds, achieving a test accuracy of 92.5% with a k-Nearest Neighbours (k-NN) classifier. This framework integrates both 3D point cloud descriptors with 2D projection-based features. Starting from raw TLS data, object clusters are extracted and described using geometric features. For each object, 3D features including linearity, planarity, and verticality are computed; 2D raster-based descriptors, including footprint spread and height variation, are computed from XY and XZ plane projections. These features are aggregated to build an object-level dataset. At the same time, global descriptors of the horizontal and vertical extents are computed to capture the dimension information. Together, these features are then standardised at the object level to train supervised classifiers capable of distinguishing four object classes: 'large bulldozer', 'other bulldozer', 'tractor-trailer', and 'other'.

The evaluation reveals that the instance-based k-NN model consistently outperformed a Support Vector Machine (SVM), which proves less robust to class imbalance and dataset shift due to its reliance on a fixed global decision boundary. Feature importance analysis confirms that a combination of 3D descriptors capturing structural complexity (e.g., eigenentropy, omnivariance) and 2D projection features quantifying vertical profiles is the most discriminative. The framework's real-world applicability is validated on an independently and automatically segmented dataset, where the k-NN model maintains a high overall accuracy of 90.9%. This validation also highlights the classification performance's sensitivity to segmentation quality, as incomplete object data from partial occlusion predictably decreases accuracy

In conclusion, this study establishes a practical and reliable feature-based methodology for monitoring anthropogenic activity in dynamic coastal zones. Future work should focus on enhancing this framework by expanding the training dataset to improve robustness, implementing adaptive binning for 2D feature extraction to better handle scale variance, and integrating more advanced segmentation algorithms to enable a fully automated monitoring pipeline. Such improvements will further solidify the method's utility for long-term environmental monitoring and data-driven coastal management.

Identification of dynamic objects in sequential terrestrial Light Detection And Ranging (LiDAR) point cloud data is important for analyzing activity and usage of coastal environments. This research focuses on identifying non-geomorphological dynamic objects, such as people and bulldozers, in sequential terrestrial LiDAR point cloud data acquired by a permanent laser scanner installed in Noordwijk, the Netherlands. A workflow is proposed and demonstrated, consisting of three main components: ground and non-ground separation using a Cloth Simulation Filter, dynamic point detection through Cloud-to-Cloud comparison, and clustering of individual dynamic objects using Density-Based Spatial Clustering of Applications with Noise (DBSCAN). Parameter tuning is performed by evaluating all possible configurations
within a defined range and validated against manually identified dynamic objects. For a week-long dataset, the error in the number of detected large dynamic objects is relatively low at 6.9%, whereas the error for small dynamic objects is higher at 23.0%, attributed to their proximity to the ground and to each other. On a point-to-point basis, the optimized configuration results in an average error of 13.5% for large dynamic objects and 33.7% for small dynamic objects with respect to a reference set. A sensitivity analysis using a Monte Carlo simulation with normally distributed parameter variations around the tuned values demonstrates robustness to moderate parameter fluctuations, particularly for larger dynamic objects, which show a standard deviation of 0.07 detected objects, while smaller objects show greater variability with a standard deviation of 0.34 detected objects. The application of the Cloth Simulation Filter adds value by excluding geomorphological processes, contributing to a reduction in error rate of approximately 95% for large objects and 62% for small objects. Overall, the presented workflow offers a robust automated approach for detecting dynamic objects on sandy beaches in LiDAR point cloud data, with demonstrated potential for scalable, long-term monitoring.

Arsenic contamination in groundwater is a major public health concern in the Ganges-Brahmaputra Basin, where millions rely on shallow aaquifers for drinking water. Naturally occurring arsenic is mobilised under specific sedimentological and geochemical conditions, particularly in Holocene alluvial deposits. Although extensively studied, arsenic distribution remains highly variable and difficult to predict. This study investigates how geomorphological features, specifically oxbow lakes and point bars, can be used to improve arsenic risk prediction and mapping using machine learning. The approach offers a targeted and scalable method for identifying high-risk zones, particularly in data-scarce environments. The divergence between theoretical assumptions and dataset trends illustrates the challenges of generalising risk models without high-precision, ground-validated input data. As a proof of concept, a two-stage workflow was implemented. In the first stage, a You Only Look Once object detection model was trained to locate oxbow lakes and point bars using satellite imagery. These landforms are key indicators of arsenic-prone zones due to their depositional history. The model performed well on well-isolated oxbow lakes and their associated point bars but struggled with hydrologically connected oxbow lakes and heavily vegetated areas, highlighting the need for more diverse training data and the potential value of false-colour imagery. A case study was conducted using historical arsenic well measurements to evaluate model assumptions. A supervised classification with the eXtreme Gradient Boosting algorithm confirmed the predictive value of geomorphological variables, with sand content, elevation, and soil organic carbon emerging as dominant predictors. Vegetation and precipitation data were excluded due to low relevance and poor temporal alignment. In the second stage, a Gaussian Mixture Model was applied to classify arsenic risk using the same geospatial variables. The model produced spatially coherent and interpretable risk zones, with high probability in most predictions. Areas of low probability were primarily located at transition zones between risk classes, indicating regions where higher-resolution or more precise input data may be necessary to reduce uncertainty and improve model reliability. This study provides a practical and semi-automated framework for geospatial arsenic risk assessment. While the risk classification is relative, future work should incorporate population-weighted exposure metrics to better guide mitigation. The method developed here supports more efficient fieldwork planning and decision-making in complex fluvial environments.

Few rare, circular, concentric enclosure ditches called rondels were discovered in Slovakia, a country in Europe; within the ditches, material traces of Neolithic European culture can be excavated. For exca- vations to happen, archaeologists must first locate these rare structures. Most rondels were spotted on agricultural fields or searched for manually during aerial surveys in a time-consuming manner. With the release of a high-resolution multispectral aerial orthophotomosaic data set of Slovakia, detailed sites containing rondels may be discovered using machine learning techniques.

Machine learning techniques using convolutional neural network (CNN) models can be applied to the field of archaeology to search for excavation sites automatically. An obstacle remains: CNN models require a lot of training image data to be efficient at their task, which is to classify whether areas contain rondels or otherwise. There are only 20 visible rondels on the orthophotomosaic that can be used as training images for model input, creating an imbalanced data set of a class with a minority class of aerial images of rondels and a large majority class of aerial images without rondels. Sketches and recorded characteristics of rondels from current images and from archaeological publications were used to automatically and randomly replicate rondel appearances from above, resulting in a created balanced data set with sufficient rondel examples for CNN training.

Multiple ResNet-34 models and a ConvNeXt model with differing hyperparameters were trained. The most promising model, a modified ResNet-34, was selected based on validation loss from the cross- entropy loss function and on the number of correctly and incorrectly classified labeled images from a test set. The selected model is used to classify data from the orthophotomosaic for rondels using a sliding window technique. Over 9510 square kilometers of agricultural land cover in western and eastern Slovakia was selected from the CORINE land cover map for classification. 7 suspected rondel sites were found, and 2 were determined to likely be rondels, based on their circular ditch-like appearance in 4 sets of multispectral images and in LiDAR elevation data. Results indicate that exact rondel layouts can be delineated with high-resolution orthophotomasics, however identifying circular elevation patterns of ditches proves to be challenging without using additional LiDAR data.

Current urban tree inventories rely heavily on time-consuming manual work and often fail to capture all trees. To effectively monitor the impact of urban trees on their environment and vice versa, an automated method for detecting and grouping trees based on their characteristics is crucial. This research aims to expand current urban tree inventories and cluster trees based on their characteristics. Existing inventories typically include species, age, and height, but trees of the same species and age can vary significantly due to environmental factors. This study focuses on a 500x600 meter area in Delft, encompassing 641 recorded trees from the municipal inventory. An automatic tree detection method using airborne LiDAR (AHN4) point cloud data combined with Random Forest classification was implemented, achieving an accuracy of 85-90%. Individual trees were identified using an existing tree segmentation algorithm, detecting 70% of recorded trees with a mean location difference of 0.71 meters and identifying an additional 460 trees, including those on private land. Despite promising results, limitations include the undetection of small trees (below 3 meters) and classification errors leading to missed detections, multiple identifications for single trees, and false positives. Geometric and reflectance features were extracted. Highresolution, 30cm, spectral images from the SuperView Neo satellites, acquired across three seasons, provided spectral features like tree color and NDVI. Overall resulting in a total of 50 features. This comprehensive inventory allows for clustering of individual trees based on geometric, reflectance, and spectral similarities using a K-means algorithm. The approach enhances urban tree inventories by incorporating new features from airborne Li- DAR and spectral images, such as tree height distribution, crown sphericity, and density. Seasonal changes from spectral images provide insights into tree behavior. These detailed features significantly improve clustering, effectively grouping similar trees together. The findings reveal that trees of the same species in seemingly similar environments exhibit significantly different characteristics. This research offers a method to enhance urban tree inventories and supports long-term studies to reveal how trees respond to urban development, climate change, and ecological dynamics. Correlating tree growth and health with specific locations and climatic conditions can aid in developing sustainable urban planning and conservation strategies.

LIght Detection And Ranging (LiDAR) imaging technology has advanced over the past two decades, being used for applications such as Digital Terrain Models (DTM) and Building Integration Modeling (BIM) integration. The resulting products, point clouds, serve diverse purposes, each demanding specific quality standards. Failing to meet the standards risks rendering the data ineffective or even unusable. Contractors, including Rijkswaterstaat, therefore, specify adherence to set requirements or standards. Current manual sampling for assessing the quality highlights the need for an automated tool.

This study proposes a workflow for automating quality validation of LiDAR infrastructure point clouds. The workflow assesses point cloud quality based on three primary components: coverage, relative accuracy, and absolute accuracy. The methodology includes:
• Point Cloud Density assessment: involves analyzing 2D horizontal cells and partial-3D spaces to ensure compliance with density requirements.
• Overlapping Regions Alignment: identifies and compares surfaces in overlapping areas of point clouds to determine relative accuracy.
• Benchmark alignment: extracts points corresponding to spherical targets, estimates the center coordinates, and evaluates adherence to absolute accuracy standards.

Quality assessments were conducted on static, mobile, and airborne point clouds. The static point cloud analysis revealed non-compliance with density requirements in 2D, with approximately half the points failing to meet standards. In 3D analysis, compliance was observed for 1 m2 horizontal cells, but individual 1-meter sections often fell short upon closer inspection. These findings highlight the need for tailored quality standards: detailed 3D analysis is crucial for complex environments like tunnels, while road environments can be effectively evaluated in 2D. Relative accuracy assessments for static and mobile datasets showed compliance with scanner specifications, with RMSE values meeting the specified requirements. Absolute accuracy assessments on static point clouds met requirements with minimal deviations in both XY and Z directions.

Recommendations include defining density requirements for different environments, establishing acceptance criteria, and defining allowable deviations for all requirements.

In 2003, the Hondsbossche and Pettemer Zeewering, located from Camperduin to Petten, was classified to not be within safety margins of the Dutch Coast. In order to protect the Dutch Coast from sea level rise, the Hondsbossche dunes were created in 2015. This project included the building of an artificial lagoon for recreational purposes. This lagoon is protected from the sea by three sand dunes. Without these dunes, the lagoon will not remain to exist. The aim of this study is to detect changes of the shape and the height of the dunes protecting the lagoon at Camperduin from the sea using annual lidar data.
JARKUS data is obtained from airborne laser scanning for the Dutch coastal areas which is done yearly between January and March. Eight datasets of annual lidar data which include the dunes are used for this study. The datasets consists of point clouds for the years 2016 to 2023.
To obtain information about the changes occurring, several methods are used. The workflow recommended is the C2M method for 3D changes. These changes can be clustered by K-means clustering. 2D changes can be observed using cross sections, contour lines and the volume changes.
The seaward side of the dunes decreases in height due to marine erosion such as storms. Around 4.5 meter erosion on some locations has been found due to large storms in 2022. The lagoon side of the dunes increases in height by aeolian transport. The rate of dune growth is found to be 0.5 meters per year.
The middle dune experiences more erosion than deposition and the volume decreases. The erosion is caused by less vegetation and human impacts. If the trend from before the storm in 2022 continues, the dune will shrink and disappear if no additional maintenance is done. The other two dunes do not experience big volume losses and are classified as stable.

This study explores the application of advanced photogrammetry techniques to enhance the accuracy of Digital Elevation Models (DEMs) derived from satellite imagery. It focuses on refining and integrating the photogrammetry pipeline to accommodate diverse satellite sources effectively. With recent advancements in photogrammetry and the expanding accessibility of satellite imagery, there is a growing opportunity for precise earth surface modeling. This research adapts a traditional photogrammetry pipeline to include state-of-the-art computer vision algorithms, specifically the DISK algorithm for feature detection and LightGlue for feature matching. These enhancements are complemented by tailored adjustments to camera models and projection matrices to suit the unique characteristics of satellite data.

The methodology emphasizes the modification of existing pipelines to optimize the handling of satellite images, incorporating sophisticated feature detection and matching technologies. The performance of these adaptations is rigorously evaluated through extensive analysis using satellite imagery across varied resolutions and environmental conditions. Results from the study indicate marked improvements in the fidelity and accuracy of the generated DEMs, which are substantiated by validation against high-resolution LiDAR ground truth data.

The refined pipeline effectively manages multi-source satellite images and produces terrain models of significantly higher quality, vital for robust geospatial analysis. This work not only bridges the gap between remote sensing and computer vision but also lays the groundwork for future research aimed at improving DEM generation from satellite imagery. This study proposes potential transformative practices in geospatial analysis and supports continued progress in fields such as environmental monitoring, urban planning, and disaster management.

The city of Amsterdam faces the challenge of monitoring and assessing 200 kilometers of historic quay walls, of which much is deemed to be in poor condition. A key monitoring technique used is photogrammetry resulting in deformation testing. The fundamental data source forming the basis of this deformation analysis is a collection of overlapping images acquired of the masonry quay walls. Solely focusing on deformations overlooks a potential wealth of information which could be retrieved from this imagery, like the existence of cracks in the quay walls, a key sign of potential deformation of the structure.
As manual visual inspection of this imagery is very time-consuming, this work proposes a methodology based on fully-supervised deep learning-based segmentation techniques with the goal of detecting and localizing cracks in the masonry quay walls. For this purpose, two neural networks are trained, one for the segmentation of quay walls in images, and one for the segmentation of cracks.
The neural network architectures which are considered in this work are DeepLabV3+, FPN, MANet and LinkNet, together with different encoders and loss functions. For quay wall segmentation, we adopt transfer learning on a network trained on masonry walls and fine-tune it for quay walls specifically. Here, DeepLabV3+ with ResNeXt-50 was found to be most effective, achieving a F1-score of 96.3 % on the test set. For crack segmentation, FPN with ResNeSt-50 performed best, resulting in a test set F1-score of 78.8 %.
The inference of the crack network is done with a multi-level scheme to detect cracks at different image scales and increase output confidence.
The inherent photogrammetric properties of the imagery have proven to be vital for further post-processing steps, like aggregating overlapping predictions, resulting in more prediction confidence.
Photogrammetry also enables converting pixel-wise predictions to crack length and crack width in the units of meters and millimeters respectively. The methodology additionally proposes photogrammetric image processing methods to transform neural network predictions to a 3D representation and a true-to-scale orthographic 2D image.
Additionally a concise visual evaluation has been conducted to assess the prediction performance on an otherwise unlabelled dataset.
This thesis presents an engineering effort for fully-supervised crack localization within the context of photogrammetric processed images, with generalization in mind for automatic assessment.

The railway industry is constantly growing to meet the demand of society for stable, accessible and sustainable transportation. With this growth, the need for the railway to be reliable increases, requiring frequent surveying and maintenance. Fugro's RILA (Rail Infrastructure aLignment Acquisition) mobile mapping system contributes by making the surveying more accessible to the relevant railway network stakeholders. However, the system has its limitations in environments where the GNSS (Global Navigation Satelite System) signal is occluded, such as tunnels and underground stations. The geo-data collected by RILA in those areas is poorly georeferenced due to the poorly tracked trajectory of the system, which introduces spatial data misalignment up to a meter or more. The current methods to fix data misalignment rely on manual data corrections, which is not cost-effective, or on automatic solutions, which have limited applicability.

Thus, the aim of this research is to develop an improved trajectory optimization method, thereby ensuring accurate geo-referencing and alignment of the survey data. This thesis proposes a newly developed methodology to achieve this aim: features are extracted from point cloud surveys, matched and utilized by g2o optimizer and GNSS processing software to optimize the trajectory. The development is described and results are evaluated on two different scales - locally, within a point cloud tile and globally, within a sequence of tiles. It is done by using Glasgow's underground railway network as a test case.

Results from the implementation demonstrate significant improvements in trajectory accuracy - a misalignment of point cloud data was reduced from a 1.5 m to a cm level within an optimization time frame that took approximately 10 hours. This improvement in accuracy was present under different complex environments using both the local and global versions of the algorithm. However, the area near the railway tunnel entrance saw a limited benefit from the implementation of the proposed algorithm.

In conclusion, the developed trajectory optimization algorithm optimizes the trajectory and improves the alignment of the survey data. Moreover, the method outperforms the currently employed solutions by being automatic and applicable in different environments. However, further research is required to optimize the algorithm itself (accuracy and computationally speed of the algorithm) and to more accurately define its limitations in terms of the surveyed environments.

Historical aerial imagery serves as a valuable data source for observing Antarctica, facilitating an extended temporal scale of observation and enabling comparisons to deepen understanding of glacier dynamics. However, many historical aerial datasets, including the Antarctica Single Frames dataset utilized in this study, lack geo-referencing and orientation metadata essential for spatial analysis. One method of geo-referencing these historical images involves image matching to establish Ground Control Points (GCPs). This study focuses on the prerequisite for image matching: ensuring alignment between unreferenced historical images and already geo-referenced images in terms of scene and approximate resolution, a process termed 'geo-localization' herein.

Geo-localization is achieved by comparing the historical image with positions within a predefined geo-referenced Area of Interest (AoI). Two predefined remote sensing datasets are used: Sentinel-2 and Quantarctica Rock Outcrop Mask, from which AoIs are generated. Positions within the AoI exhibiting the highest similarity to the historical image are likely to correspond to the same ground area, thus providing the location of the historical imagery.

This similarity assessment employs two Siamese Networks: SigNet and ResNet-50. SigNet, originally designed for signature verification tasks, consists of four convolutional layers. In contrast, ResNet-50, initially developed for image classification purposes, is characterized by its deep architecture comprising approximately 50 convolutional layers, as suggested by its name. In this study, these two models are initially pre-trained on cross-domain datasets and subsequently adaptively trained with task-specific datasets created in this study. The adaptive training datasets comprise triplets of similar and dissimilar images pre-processed using methods devised in this study. An evaluation methodology based on confidence level is developed to assess the model and workflow performance, which is then applied to 51 test historical image samples.

Overall, the results indicate that the ResNet-50 based network outperforms SigNet, achieving a 95.5% average confidence level. However, the method does not meet the initial expectation of directly providing the location of the historical image within the AoI. Instead, it identifies potential locations. Nevertheless, this outcome is valuable as it streamlines the search process for subsequent image matching steps. For instance, a 95.5% average confidence level for the ResNet-50 based network correlates with an approximate 95.5% reduction in processing time for geo-referencing when integrated with image matching in subsequent steps.

Sand waves can have a significant impact on offshore activities, and with the global push for sustainability and more renewable energy sources, these negative impacts will become more prevalent. Sand waves in the North Sea have a height up to a couple of meters and a wavelength between 100 and 1000 meters. The migration or change in asymmetry of sand waves can cause free spans in export cables leading to buckling or vibration causing failure. The migration can also cause cables to be exposed on the sea bed and in danger from offshore activities like anchors and fishing boats.

This research focuses on utilizing satellite data to determine sand wave characteristics. Satellites like Sentinel-1 and Sentinel-2 which are used in this research have global availability of data over multiple years at a 10 meter resolution. Sentinel-1 has a Synthetic Aperture Radar (SAR) which creates images by sending out microwave signals and recording the strength and the time delay of the returning signals. Sentinel-2 has an optical instrument which creates images by recording the reflected light from the sun on the earth's surface. Sand waves were detected through the change in Sea Surface Roughness due to the current interaction with the sea bed as described in the Alpers-Hennings Model. The difference in resulting sand wave characteristics between SAR and optical images were determined.

Three initial areas of interest off the coast of the Netherlands were chosen based on their different characteristics. Hoek van Holland contains short irregular sand waves. Holland Kust Zuid contains long regular sand waves. The third location is Alkmaar, which contains no sand waves and was used to determine the results of the methodology when a location has no sand waves present. The first step was to determine the environmental conditions necessary for sand waves to be visible in satellite images. Image collections for an area of interest over a set period of time were filtered based on the environmental parameters. These parameters include mean glint angle and cloud cover for optical images, and wind and current speed for both optical and SAR images. Threshold values were determined through literature. The wind speed should be between 3 and 12 m/s, the current speed should be greater than 0.4 m/s, and the mean glint angle for the latitude of the North Sea is less than 56 degrees. The image collection for the year 2021 for the three different areas of interest was filtered based on these parameters. The final collection after filtering was then manually checked for the visibility of sand waves. This number was lower than the initial filtered collection which means that there are additional factors that need to be accounted for.

A methodology was created to determine the sand wave characteristics over an area by applying a Fast Fourier Transformation (FFT) and calculating the sand wavelength and wave angle from the resulting signal. This was first applied to the three different areas of interest within the North Sea, Hoek van Holland, Holland Kust Zuid, and Alkmaar. Using the resulting image collection at these locations the sand wave characteristics were calculated. Although there are no sand waves at Alkmaar, due to the methodology, sand wave characteristics are still calculated. Also, the methodology works better at Holland Kust Zuid with long regular sand waves compared to at Hoek van Holland where sand waves are shorter and irregular.

To determine at which scale the satellite images should be viewed at to obtain information on the sand wave characteristics, the data was split into different area sizes and a FFT performed. The resulting average values and spread of the sand wavelength and wave angle were compared at different area sizes for optical and SAR images, as well as Multi-Beam Sonar (MBES). This resulted in an area of 5 by 5 kilometers which allows for the correct signals to be read across different possible sand wavelengths. Also, the effect of different sources of noise on the FFT and calculation of the sand wave characteristics for optical and SAR images was determined. From this it was seen that ships within a satellite image containing sand waves introduce a very strong signal within the FFT. This prevents the sand wave characteristics from being calculated correctly. Wind farms within a satellite image affect SAR images, and not optical images. Suspended sediment transport and clouds, which only occur within optical images, have different effects. Optical images are not affected by sediment transport because the current direction will always be perpendicular to sand wave crests and therefore does not influence the calculation of sand wave characteristics. Clouds, similarly to ships, introduce a strong signal within the FFT which blocks the signal of the sand waves. Dark patches (like rain cells), which effect SAR images, block the visibility of sand waves within satellite images and therefore there no signal corresponding to sand waves within the FFT can be found.

Then, the area of interest was increased to the entire southern North Sea. Satellite images over the North Sea were downloaded in 5 by 5 km tiles for a single satellite pass over per year where the environmental parameters met the conditions necessary to view sand waves. As for the local scale cases, a FFT was applied to each tile and the sand wave characteristics were calculated including sand wavelength, wave angle, density, and spatial frequency. The differences in characteristics were compared from year to year. Typically, the difference is less than 200 meters in sand wavelength, however there are points where the difference is much larger. This is due to noise affecting the calculated value. Additionally, the data for wind speed and current speed for the date and time of the satellite image was determined. From this it is seen that optical images can contain sand waves at lower wind and current speeds, at the lower end of the boundary conditions. Sand waves are only visible in SAR images when the current speed is very high. Also, the results were compared to the water depth of the North Sea. At depths less than 40 meters which occurs on the Dutch Continental Shelf sand waves are almost always visible with the correct environmental parameters. At depths greater than 40 meters, which occurs on the West side of the sand wave field, sand waves are not visible consistently. The calculated sand wavelength over the North Sea was also compared to wavelength obtained from MBES. This resulted in low correlation values for both optical and SAR. As a result, noise present in satellite images should be taken into account.

By both applying the methodology to smaller scale cases and to the entire North Sea, it is possible to detect sand waves in different environments. Using the change in sea surface roughness due to sand waves it is possible to see sand waves in both optical and SAR satellite images. By applying the Fourier transformation the average sand wave characteristics over an area can be calculated. The methodology is limited by environmental parameters and noise that can be present in the satellite images. SAR and optical images require specific wind and current conditions, while optical images also requires specific mean glint angles and a low cloud cover. Different sources of noise also have a negative impact, adding signals to the FFT that do not correspond to the sand waves. Although the methodology proposed in this thesis is successful in determining the average sand wave characteristics over an area, more research for this topic would increase the possibilities. This includes using additional conditions for filtering image collections for sand waves, removing sources of noise, utilizing higher resolution satellite data, and testing the methodology on different sand wave fields that have different characteristics, such as symmetry. There are still many possibilities that can be explored in using satellite images to determine sand wave characteristics.

Coastal areas are sensitive ecosystems, which are important as natural habitat, recreational areas and for protection of the hinterland. Coastal monitoring is essential in the analysis and prediction of coastal development. Coastal monitoring has become more urgent due to climate change and its effects on sea level and frequency of storm events. Permanent laser scanning (PLS) provides a tool to acquire 3-dimensional point clouds of an area nearly continuously over extended periods of time. PLS bridges the gap between incidental, in-situ measurement campaigns with high spatial detail, and frequent monitoring via satellite data at lower spatial resolution. Methods to process the complex high-resolution permanent laser scanning data are needed to find and analyse the effects of geomorphological processes over extended periods of time and with high spatial detail. This dissertation deals with the development of methods for spatio-temporal data mining of large 4D data sets from permanent laser scanning for the application of coastal monitoring on sandy beaches.

Two data sets of hourly 3-dimensional point clouds, acquired over periods of six months and three years at two different locations on the Dutch coast were analysed, to identify and assess geomorphological dynamics at the sediment surface. Each data set contains up to 20 000 epochs capturing the dynamics of the sandy beaches in Kijkduin and Noordwijk.

In this thesis two methods are developed: the application of multiple hypothesis testing for the estimation of minimal detectable bias and the generation of a so-called inventory of trends, and the application of clustering algorithms for grouping elevation time series. The first method using multiple hypothesis testing provides a means to define the minimal detectable bias for an expected model behaviour of time series from permanent laser scanning. This method provides a new way to detect small but persistent and statistically significant changes in longer time series derived from 3-dimensional point clouds. Using multiple hypothesis testing allows to identify linear changes with slopes of 0.032 m/day and sudden changes in elevation of 0.031 m with a given discriminatory power of 80% and significance level of 5% in 24-hour time series.

In an additional step, multiple hypothesis testing is used to reduce the complex permanent laser scanning data set to an inventory of trends, which consists of linear pieces of time series, matching the predefined statistical models and corresponding parameters. This method is applied to find and analyse times and areas where specific processes such as storms, aeolian sand transport or bulldozer works occur. The inventory of trends is particularly effective for the detection of aeolian sand transport, which has been difficult to identify using other coastal observations because it causes small, gradual deformations at the sediment surface.

The second method uses clustering algorithms to identify areas which are subject to similar change patterns. These change patterns are then easily associated with underlying physical and anthropogenic processes, mostly tidal induced changes and bulldozer works.

In summary, the developed methods allow to effectively detect deformations on sandy beaches and establish their origins, such as storms, tides, anthropogenic activities or aeolian sand transport, with a resolution and detail that has not been achieved until now. These results allow further analysis and interpretation of geomorphological coastal processes. For instance, the analysis of bulldozer works in our study area leads to the conclusion, that not only buildings themselves, but also the associated human interventions on the sandy beach around each building have a significant impact on coastal morphology and possibly lead to increased erosion.

Semantic segmentation, a task vital in the creation of 3D point cloud models for buildings, is aimed at assigning meaning to individual points. However, due to the vast volume of unstructured point cloud data, precise semantic segmentation remains challenging. Significant progress has been observed in recent years with the application of deep learning techniques to point cloud segmentation, and the effectiveness of Dynamic Graph Convolutional Neural Network (DGCNN) and K-Nearest Neighbors (K-NN) in handling point cloud data has been recognized.
In this study, Dynamic Graph Convolutional Neural Network (DGCNN) was utilized for semantic segmentation on a building's point cloud scene. We adopted K-Nearest Neighbors (K-NN) as a crucial component of our methodology to optimize the segmentation process. By varying 'k' values in K-NN and exploring different block sizes, we aimed to obtain various segmentation results for comparison. When a block size of 1 meter was employed and 'k' was set to 20, an overall accuracy of 90.32%, mean accuracy of 87.64%, and IoU of 80.71% were achieved. However, the most favorable segmentation outcomes were observed when the block size remained 1 meter, and 'k' was set to 30, resulting in an overall accuracy of 93.86%, mean accuracy of 90.68%, and IoU of 84.97%.
These experiments underscore the significance of parameter selection in optimizing the performance of DGCNN for point cloud segmentation. The findings reveal that adjustments to 'k' values and block sizes can significantly influence segmentation accuracy and quality, emphasizing the importance of parameter optimization in the context of semantic segmentation for building point clouds using deep learning techniques. The utilization of K-NN played a crucial role in achieving these improvements by allowing us to adapt to the inherent variability in point cloud data.

The 3D point cloud representation of building holds significant importance in developing a comprehensive smart city model. However, due to the substantial volume of point cloud data and its inherently unstructured nature, accomplishing semantic segmentation of the point cloud poses a formidable challenge. Recent years have witnessed substantial advancements in employing deep learning techniques for point cloud segmentation. Both PointNet and PointNet++ have demonstrated their prowess in managing point cloud data. In the context of this study, we harness the potential of DGCNN for segmenting building point clouds. DGCNN plays a pivotal role in partitioning the input point cloud, thereby influencing the sensor domain's network range. Consequently, this research delves into the role of parameter k in K-NN (K-Nearest Neighbors) and its implications. The findings illustrate that augmenting the 'k' value contributes to enhancing the overall precision of DGCNN. Nonetheless, complete precision in segmenting every module cannot be assured. The research objectives in the study is building façade, and the targets are windows, walls, balconies and doors. In this study, the key parameters are the values of 'k' and the 'block size.' Experiments were conducted by varying the values of these two parameters to obtain different segmentation results of building facade point clouds. Notably, setting k to 20, block size to 1m achieves an overall accuracy rate of 89.75%, a mean accuracy rate of 85.03%, an IoU of 76.24%, while elevating k to 30 and block size to 1m results in a heightened accuracy rate of 95.74%, a mean accuracy rate of 89.37%, an IoU of 81.52%.

Unequal deformation of the soil can cause deformation or damage to buildings, like tilted facades or cracks in walls. This research investigates how deformation of a building can be analyzed using Light Detection And Ranging (LiDAR) data. Cyclomedia captures LiDAR data yearly in the Netherlands making it possible to analyze either one or multiple epochs of data. If deformation is monitored, failure can be predicted, and repairs can be performed in time.

A subsiding area is found using the Dutch surface motion map. The data points of a single building are fetched from the LiDAR point cloud to analyze the following types of deformation: a difference in the torsion and tilt angle of the facade between two epochs of data; the tilt angle of the facade for a single epoch of data; and local deformation patterns on the facade. The data is segmented before computing the angles and analyzing the local deformation patterns. During segmentation, points that do not correspond to the façade (like a sunshade or windows) are removed. After segmentation, the facade is modeled by fitting a plane using Principal Component Analysis (PCA). The plane parameters (A, B, C, D) are used to determine the torsion and tilt angles. If the torsion or tilt angles are large (above a degree), it is expected the facade has deformed. Local deformation patterns can be analyzed by using a raster containing the distance of the segmented points with an accuracy of 8 centimeters. Besides the point clouds, Cyclomedia also captures 360° panoramic images (Cycloramas). These Cycloramas can be used to explain patterns that are visible in the rasters. For example, objects in front of the building like a bench or a sunshade.

This research uses Random Sample Consensus (RANSAC) to segment the data. From the results, it can be concluded RANSAC is not very predictable because random points are taken as input. So, points corresponding to the facade can be removed instead of points corresponding to windows or doors. Therefore, it is recommended to use another segmentation method for future research instead of RANSAC. Machine learning could be a good alternative to remove objects like windows and other unwanted points from the data.

Coastal areas are appealing for human beings and are the most populated areas of the world. Therefore, coastal processes have large impacts on economy and on millions of people's lives. Not only processes such as erosion, extreme events, and flood events must be understood and monitored, but the uncertainty given by the rapid climate change on these processes and on coastal systems pose a bigger issue on these populated areas. One of the key parameters of coastal processes, which is notably required for modelling coastal floods, is the beach face slope. Unfortunately, there is a lack of a reliable global dataset of the steepness of the beach face, which has been indicated to be key limitation for developing operational coastal inundation forecasting systems, as well as for quantifying the run-up and set-up contribution at the shoreline relative to global sea-level-rise.

In this report, a methodology for providing this parameter at a global level by using a LiDAR-based satellite (ICESat-2) is presented. Here, the accuracy of the beach face slope estimation, its dependency on the angle and distance between ICESat-2 measurements (passes) and the transect perpendicular to the coastline they are projected into (transect), a spatial and temporal exploration of ICESat-2 measurements on world's coasts, as well as the limitations and opportunities of this methodology are presented.

The findings of this study reveal strong visual agreement between the beach profiles and beach face profiles obtained from the validation datasets and ICESat-2. Moreover, the computed metrics, including the Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE), present low values, indicating a high level of agreement. The MAE values, ranging from 9 to 34 cm, not only demonstrate good agreement within the beach and beach face profiles but also indicate that the errors fall within the inherent variability of the beach. For assessing the impact of the angle and distance between passes and transect on the beach face slope extraction, synthetic data is generated using Digital Elevation Models (DEMs). The results demonstrate that long and spatially homogeneous beaches withstand larger α, whereas smaller, more variable, and pocket-shaped beaches are more sensitive to large angles due to the rapid decrease of spatial information alongshore. Nevertheless, regardless of the beach type, the closer the synthetic ICESat-2 measurements intersect over the beach face, the more likely the extracted beach face slope is to match the actual one. For both analysed cases, the beach face slope can be extracted with a tolerable level of uncertainty up to 60 degrees, and if the synthetic data is close to the beach face, the angle that it bears can be increased up to 70-80 degrees.

The novel methodology for upscaling the beach face slope extraction is validated using the Dutch yearly coastal dataset, JARKUS. The results reaffirm the dependence of the beach face slope on the angle between pass and transect. It is concluded that even though further refinement of the algorithm for demarcating the beach face is necessary, the extraction of beach face slopes can be implemented on a global scale, if some challenges are overcome. Moreover, this methodology yields a valuable byproduct in the form of cross-shore profiles, which can be extracted globally with high accuracy. This additional output enhances the utility and versatility of the methodology. Additionally, a spatial and temporal exploration of ICESat-2 ATL03 data on the world's coasts, by globally sampling boxes of 20 km², is implemented. Then, by analysing the preliminary results from boxes already processed, it is observed that these boxes can encompass anywhere from 0 to 160 passes over the course of the available 5-year data. On average, approximately 63 passes are recorded within each box. From this analysis, it is concluded that particular coastal stretches, depending on their geometry and cloud coverage, can be effectively monitored for beach face slope changes using ICESat-2. Moreover, from the information extracted from these boxes, a map is being constructed for the purpose of revealing potential areas where the extraction of beach face slopes are possible given the beach geometry and availability of data. In summary, this study provides a novel methodology that sets the basis for a new era of coastal management and monitoring using LiDAR-based satellites at a global level.

Photogrammetry is a well-established technique that has a significant impact on the use of images for mapping purposes. Employing feature extraction, matching, and the bundle adjustment, a large number of images can be automatically processed. Ingenieursbureau Geodelta developed an application for measurements in processed (oblique) aerial images. However, the quality of these measurements and the added value of oblique images on the adjustment's quality were unclear. This will be assessed by means of the theoretical standard deviation resulting from the bundle adjustment. Upon estimating this quality, an adapted RANSAC method is proposed that incorporates this quality as weights within its algorithm to extract geometric features. The objective is to evaluate whether this enhances the RANSAC results and could be applied to 3D reconstructions.

The results indicate three key factors that influence the theoretical standard deviation: high tie point availability, larger observation angles, and image viewing direction. With this, the theoretical standard deviation for tie points in both Nadir and Oblique image sets separately approximates 3 centimeters in the horizontal direction and about 10 in the height direction. Combining the two sets enhances the results by nearly a factor of three in all directions because the Nadir images connect the Oblique images, combining the strong characteristics of both sets. This demonstrates the value of both Nadir and Oblique imagery. For image exteriors, the improvement is even more pronounced, yielding an improvement factor of 4 or 5. However, propagating this quality metric through a dense matching algorithm in an adapted, weighted RANSAC algorithm does not show significant improvements in the number of planes found or the percentage of points classified as inliers of those planes. Furthermore, the RANSAC method does not converge to a better result in fewer iterations using the proposed method.

There are multiple types of objects distributed along the rail trackside to support the operation of railroads. Damage to those objects will result in delay and cancellation of trains. Therefore, the maintenance of the objects is important to ensure the safety of railway operations, and the key to maintenance is to know the accurate position of the objects. RILA - a railway mobile mapping system developed by Fugro - provides continuous monitoring of the rail environment without interrupting the daily operation of rail tracks. The RILA dataset contains different types of data including video frames, point clouds, and measurement trajectories from the Global Navigation Satellite System and Inertial Measurement Unit (GNSS-IMU). This enables a remote and automatic object positioning process based on object recognition in video frames and positioning using photogrammetric techniques.

Most of the rail trackside objects are parts of the railway electrification system and the railway signaling system. The available ground truth data contains more information about signaling objects, so this thesis mainly focuses on the positioning of signaling objects. Three typical components: signal lights, railway cabinets, and railway side markers are selected to make case studies and develop the object detection and localization pipeline for them.

There are two main issues that existed in former research: One is object recognition only gives a 2D pixel position for an object, but we want to know its 3D position in the world; the other is the contextual information between frames is ignored in object positioning. Based on that, this thesis developed a new workflow: firstly, this thesis proposes the use of an existing 3D object detection model, Single-Stage Monocular 3D Object Detection via Keypoint Estimation (SMOKE), to recognize objects of interest from the RILA video frames. This model can not only predict an object's 2D pixel position on the image, but also its 3D position with respect to the camera. This enables the localization of an object in each of the frames it is detected in, which provides redundancy to define its final position. Secondly, this thesis developed a pipeline to clean the detection output of SMOKE so that most false detections can be removed. In addition, this thesis developed a method to use the contextual information in image sequences to position objects: first, extract the sequences by classifying all the predictions in the detection output through the Euclidean distances, then for each sequence, analyze the varying trend of predicted positions to select reliable positions. The mean of the selected positions is the final improved position of that object. By making tests on a part of the railway near Ely, UK, with 2983 frames, the workflow detected and positioned all of the 4 signal lights, all of the 10 markers, and 13 of the 14 cabinets in the testing dataset. Most signal lights and cabinets can be positioned within 1 meter to the ground truth, and the mean positioning offset for the railway side markers can be controlled around 2 meters. Complex scenarios could be solved, when up to three cabinets were located close together.