References of "Teferle, Felix Norman 50003185"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailRobust Techniques for Building Footprint Extraction in Aerial Laser Scanning 3D Point Clouds
Nurunnabi, Abdul Awal Md UL; Teferle, Felix Norman UL; Balado, Jesus et al

in Robust Techniques for Building Footprint Extraction in Aerial Laser Scanning 3D Point Clouds (2022, November)

The building footprint is crucial for a volumetric 3D representation of a building that is applied in urban planning, 3D city modeling, cadastral and topographic map generation. Aerial laser scanning (ALS ... [more ▼]

The building footprint is crucial for a volumetric 3D representation of a building that is applied in urban planning, 3D city modeling, cadastral and topographic map generation. Aerial laser scanning (ALS) has been recognized as the most suitable means of large-scale 3D point cloud data (PCD) acquisition. PCD can produce geometric detail of a scanned surface. However, it is almost impossible to get point clouds without noise and outliers. Besides, data incompleteness and occlusions are two common phenomena for PCD. Most of the existing methods for building footprint extraction employ classification, segmentation, voting techniques (e.g., Hough-Transform or RANSAC), or Principal Component Analysis (PCA) based methods. It is known that classical PCA is highly sensitive to outliers, even RANSAC which is known as a robust technique for shape detection is not free from outlier effects. This paper presents a novel algorithm that employs MCMD (maximum consistency within minimum distance), MSAC (a robust variant of RANSAC) and a robust regression to extract reliable building footprints in the presence of outliers, missing points and irregular data distributions. The algorithm is successfully demonstrated through two sets of ALS PCD. [less ▲]

Detailed reference viewed: 35 (0 UL)
Full Text
Peer Reviewed
See detailkCV-B: Bootstrap with Cross-Validation for Deep Learning Model Development, Assessment and Selection
Nurunnabi, Abdul Awal Md UL; Teferle, Felix Norman UL; Laefer, Debra et al

in kCV-B: Bootstrap with Cross-Validation for Deep Learning Model Development, Assessment and Selection (2022, October)

This study investigates the inability of two popular data splitting techniques: train/test split and k-fold cross-validation that are to create training and validation data sets, and to achieve sufficient ... [more ▼]

This study investigates the inability of two popular data splitting techniques: train/test split and k-fold cross-validation that are to create training and validation data sets, and to achieve sufficient generality for supervised deep learning (DL) methods. This failure is mainly caused by their limited ability of new data creation. In response, the bootstrap is a computer based statistical resampling method that has been used efficiently for estimating the distribution of a sample estimator and to assess a model without having knowledge about the population. This paper couples cross-validation and bootstrap to have their respective advantages in view of data generation strategy and to achieve better generalization of a DL model. This paper contributes by: (i) developing an algorithm for better selection of training and validation data sets, (ii) exploring the potential of bootstrap for drawing statistical inference on the necessary performance metrics (e.g., mean square error), and (iii) introducing a method that can assess and improve the efficiency of a DL model. The proposed method is applied for semantic segmentation and is demonstrated via a DL based classification algorithm, PointNet, through aerial laser scanning point cloud data. [less ▲]

Detailed reference viewed: 41 (2 UL)
Full Text
Peer Reviewed
See detailRobust Approach for Urban Road Surface Extraction Using Mobile Laser Scanning Data
Nurunnabi, Abdul Awal Md UL; Teferle, Felix Norman UL; Lindenbergh, Roderik et al

in Robust Approach for Urban Road Surface Extraction Using Mobile Laser Scanning Data (2022, June)

Road surface extraction is crucial for 3D city analysis. Mobile laser scanning (MLS) is the most appropriate data acquisition system for the road environment because of its efficient vehicle-based on-road ... [more ▼]

Road surface extraction is crucial for 3D city analysis. Mobile laser scanning (MLS) is the most appropriate data acquisition system for the road environment because of its efficient vehicle-based on-road scanning opportunity. Many methods are available for road pavement, curb and roadside way extraction. Most of them use classical approaches that do not mitigate problems caused by the presence of noise and outliers. In practice, however, laser scanning point clouds are not free from noise and outliers, and it is apparent that the presence of a very small portion of outliers and noise can produce unreliable and non-robust results. A road surface usually consists of three key parts: road pavement, curb and roadside way. This paper investigates the problem of road surface extraction in the presence of noise and outliers, and proposes a robust algorithm for road pavement, curb, road divider/islands, and roadside way extraction using MLS point clouds. The proposed algorithm employs robust statistical approaches to remove the consequences of the presence of noise and outliers. It consists of five sequential steps for road ground and non-ground surface separation, and road related components determination. Demonstration on two different MLS data sets shows that the new algorithm is efficient for road surface extraction and for classifying road pavement, curb, road divider/island and roadside way. The success can be rated in one experiment in this paper, where we extract curb points; the results achieve 97.28%, 100% and 0.986 of precision, recall and Matthews correlation coefficient, respectively. [less ▲]

Detailed reference viewed: 20 (0 UL)
Full Text
See detailDeep Learning for Ground and Non-ground Surface Separation: A Feature-based Semantic Segmentation Algorithm for Point Cloud Classification
Nurunnabi, Abdul Awal Md UL; Lindenbergh, Roderik; Teferle, Felix Norman UL

E-print/Working paper (2022)

Precise ground surface topography is crucial for 3D city analysis, digital terrain modeling, natural disaster monitoring, high-density map generation, and autonomous navigation to name a few. Deep ... [more ▼]

Precise ground surface topography is crucial for 3D city analysis, digital terrain modeling, natural disaster monitoring, high-density map generation, and autonomous navigation to name a few. Deep learning (DL; LeCun, et al., 2015), a division of machine learning (ML), has been achieving unparalleled success in image processing, and recently demonstrated a huge potential for point cloud analysis. This article presents a feature-based DL algorithm that classifies ground and non-ground points in aerial laser scanning point clouds. Recent advancements of remote sensing technologies make it possible digitizing the real world in a near automated fashion. LiDAR (Light Detection and Ranging) based point clouds that are a type of remotely sensed georeferenced data, providing detailed 3D information on objects and environment have been recognized as one of the most powerful means of digitization. Unlike imagery, point clouds are unstructured, sparse and of irregular data format which creates many challenges, but also provides huge opportunities for capturing geometric details of scanned surfaces with millimeter accuracy. Classifying and separating non-ground points from ground points largely reduce data volumes for consecutive analyses of either ground or non-ground surfaces, which consequently saves cost and labor, and simplifies further analysis. [less ▲]

Detailed reference viewed: 55 (1 UL)
Full Text
Peer Reviewed
See detailResampling methods for a reliable validation set in deep learning based point cloud classification
Nurunnabi, Abdul Awal Md UL; Teferle, Felix Norman UL

in Resampling methods for a reliable validation set in deep learning based point cloud classification (2022, June)

A validation data set plays a pivotal role in tweaking a machine learning model trained in a supervised manner. Many existing algorithms select a part of available data by using random sampling to produce ... [more ▼]

A validation data set plays a pivotal role in tweaking a machine learning model trained in a supervised manner. Many existing algorithms select a part of available data by using random sampling to produce a validation set. However, this approach can be prone to overfitting. One should follow careful data splitting to have reliable training and validation sets that can produce a generalized model with a good performance for the unseen (test) data. Data splitting based on resampling techniques involves repeatedly drawing samples from the available data. Hence, resampling methods can give better generalization power to a model, because they can produce and use many training and/or validation sets. These techniques are computationally expensive, but with increasingly available high-performance computing facilities, one can exploit them. Though a multitude of resampling methods exist, investigation of their influence on the generality of deep learning (DL) algorithms is limited due to its non-linear black-box nature. This paper contributes by: (1) investigating the generalization capability of the four most popular resampling methods: k-fold cross-validation (k-CV), repeated k-CV (Rk-CV), Monte Carlo CV (MC-CV) and bootstrap for creating training and validation data sets used for developing, training and validating DL based point cloud classifiers (e.g., PointNet; Qi et al., 2017a), (2) justifying Mean Square Error (MSE) as a statistically consistent estimator, and (3) exploring the use of MSE as a reliable performance metric for supervised DL. Experiments in this paper are performed on both synthetic and real-world aerial laser scanning (ALS) point clouds. [less ▲]

Detailed reference viewed: 35 (1 UL)
Full Text
Peer Reviewed
See detailA TWO-STEP FEATURE EXTRACTION ALGORITHM: APPLICATION TO DEEP LEARNING FOR POINT CLOUD CLASSIFICATION
Nurunnabi, Abdul Awal Md UL; Teferle, Felix Norman UL; Laefer, Debra et al

in A TWO-STEP FEATURE EXTRACTION ALGORITHM: APPLICATION TO DEEP LEARNING FOR POINT CLOUD CLASSIFICATION (2022, March)

Most deep learning (DL) methods that are not end-to-end use several multi-scale and multi-type hand-crafted features that make the network challenging, more computationally intensive and vulnerable to ... [more ▼]

Most deep learning (DL) methods that are not end-to-end use several multi-scale and multi-type hand-crafted features that make the network challenging, more computationally intensive and vulnerable to overfitting. Furthermore, reliance on empirically-based feature dimensionality reduction may lead to misclassification. In contrast, efficient feature management can reduce storage and computational complexities, builds better classifiers, and improves overall performance. Principal Component Analysis (PCA) is a well-known dimension reduction technique that has been used for feature extraction. This paper presents a two-step PCA based feature extraction algorithm that employs a variant of feature-based PointNet (Qi et al., 2017a) for point cloud classification. This paper extends the PointNet framework for use on large-scale aerial LiDAR data, and contributes by (i) developing a new feature extraction algorithm, (ii) exploring the impact of dimensionality reduction in feature extraction, and (iii) introducing a non-end-to-end PointNet variant for per point classification in point clouds. This is demonstrated on aerial laser scanning (ALS) point clouds. The algorithm successfully reduces the dimension of the feature space without sacrificing performance, as benchmarked against the original PointNet algorithm. When tested on the well-known Vaihingen data set, the proposed algorithm achieves an Overall Accuracy (OA) of 74.64% by using 9 input vectors and 14 shape features, whereas with the same 9 input vectors and only 5PCs (principal components built by the 14 shape features) it actually achieves a higher OA of 75.36% which demonstrates the effect of efficient dimensionality reduction. [less ▲]

Detailed reference viewed: 47 (1 UL)
Full Text
Peer Reviewed
See detailMultiscale Integration of High-Resolution Spaceborne and Drone-Based Imagery for a High-Accuracy Digital Elevation Model Over Tristan da Cunha
Backes, Dietmar UL; Teferle, Felix Norman UL

in Frontiers in Earth Science (2020)

Very high-resolution (VHR) optical Earth observation (EO) satellites as well as low-altitude and easy-to-use unmanned aerial systems (UAS/drones) provide ever-improving data sources for the generation of ... [more ▼]

Very high-resolution (VHR) optical Earth observation (EO) satellites as well as low-altitude and easy-to-use unmanned aerial systems (UAS/drones) provide ever-improving data sources for the generation of detailed 3-dimensional (3D) data using digital photogrammetric methods with dense image matching. Today both data sources represent cost-effective alternatives to dedicated airborne sensors, especially for remote regions. The latest generation of EO satellites can collect VHR imagery up to 0.30 m ground sample distance (GSD) of even the most remote location from different viewing angles many times per year. Consequently, well-chosen scenes from growing image archives enable the generation of high-resolution digital elevation models (DEMs). Furthermore, low-cost and easy to use drones can be quickly deployed in remote regions to capture blocks of images of local areas. Dense point clouds derived from these methods provide an invaluable data source to fill the gap between globally available low-resolution DEMs and highly accurate terrestrial surveys. Here we investigate the use of archived VHR satellite imagery with approx. 0.5 m GSD as well as low-altitude drone-based imagery with average GSD of better than 0.03 m to generate high-quality DEMs using photogrammetric tools over Tristan da Cunha, a remote island in the South Atlantic Ocean which lies beyond the reach of current commercial manned airborne mapping platforms. This study explores the potentials and limitations to combine this heterogeneous data sources to generate improved DEMs in terms of accuracy and resolution. A cross-validation between low-altitude airborne and spaceborne data sets describes the fit between both optical data sets. No co-registration error, scale difference or distortions were detected, and a quantitative cloud-to-cloud comparison showed an average distance of 0.26 m between both point clouds. Both point clouds were merged applying a conventional georeferenced approach. The merged DEM preserves the rich detail from the drone-based survey and provides an accurate 3D representation of the entire study area. It provides the most detailed model of the island to date, suitable to support practical and scientific applications. This study demonstrates that combination archived VHR satellite and low-altitude drone-based imagery provide inexpensive alternatives to generate high-quality DEMs. [less ▲]

Detailed reference viewed: 99 (25 UL)
Full Text
See detailConventional EO Satellites vs. CubeSats; FDL - AI flood detection onboard a Nano Satellite
Backes, Dietmar UL; Schumann, Guy; Teferle, Felix Norman UL

Scientific Conference (2019, December 11)

Detailed reference viewed: 71 (2 UL)
Full Text
See detailA comparison between conventional Earth Observation Satellites and CubeSats; Requirements, Capabilities and Data Quality
Backes, Dietmar UL; Hassani, Saif Alislam UL; Teferle, Felix Norman UL et al

Scientific Conference (2019, September 11)

From its early beginning as an educational tool in 1999, cubesats have evolved into a popular platform for technology demonstrations and scientific instruments. Ideas and innovations sparked from an ... [more ▼]

From its early beginning as an educational tool in 1999, cubesats have evolved into a popular platform for technology demonstrations and scientific instruments. Ideas and innovations sparked from an enthusiastic community led to the development of new Earth Observation (EO) technology concepts based on large constellations of satellites with high-resolution optical imagers previously considered as infeasible. Probably the most significant constellation today is deployed by Planet who are currently operating a fleet larger than 120 3U Dove satellites, which provide an imaging service with up to 3m Ground Sample Distance (GSD). The number of low-cost EO Cubesat systems is constantly increasing. However, for a number of reasons there still seems to be a reluctance to use such data for many EO applications. A better understanding of the capabilities of the current generation of small Cubesats compared to the traditional well-established bigger operational missions of high and medium resolution EO satellites is required. What are the critical capabilities and quality indicators? Due to the limited size and weight of Cubesats, critical system components, e.g. for navigation and communication, always compete with operational payloads such as optical camera/sensor systems. A functional EO system requires balanced payload, which provides adequate navigational capabilities, that match the requirements of the optical imagers (camera) deployed with the system. This study reviews the current performance and capabilities of Cubesats for optical EO and compares them to the capabilities of conventional, dedicated high and medium resolution EO systems. We summarise key performance parameters and quality indicators to evaluate the difference between the systems. An empirical study compares recent very high-resolution (VHR) imagery from big EO satellite missions with available images from Cubesats for the use case in disaster monitoring. Small and agile Nanosatellites or Cubesats already show remarkable performance. Although it is not expected that their performance and capability will match those of current bigger EO satellite missions, they are expected to provide a valuable tool for EO and remote sensing, in particular for downstream industry applications. [less ▲]

Detailed reference viewed: 320 (47 UL)
Full Text
Peer Reviewed
See detailTowards a high-resolution drone-based 3D mapping dataset to optimise flood hazard modelling
Backes, Dietmar UL; Schumann, Guy; Teferle, Felix Norman UL et al

in International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences (2019, June), XLII-2/W13

The occurrence of urban flooding following strong rainfall events may increase as a result of climate change. Urban expansion, ageing infrastructure and an increasing number of impervious surfaces are ... [more ▼]

The occurrence of urban flooding following strong rainfall events may increase as a result of climate change. Urban expansion, ageing infrastructure and an increasing number of impervious surfaces are further exacerbating flooding. To increase resilience and support flood mitigation, bespoke accurate flood modelling and reliable prediction is required. However, flooding in urban areas is most challenging. State-of-the-art flood inundation modelling is still often based on relatively low-resolution 2.5 D bare earth models with 2-5m GSD. Current systems suffer from a lack of precise input data and numerical instabilities and lack of other important data, such as drainage networks. Especially, the quality and resolution of the topographic input data represents a major source of uncertainty in urban flood modelling. A benchmark study is needed that defines the accuracy requirements for highly detailed urban flood modelling and to improve our understanding of important threshold processes and limitations of current methods and 3D mapping data alike. This paper presents the first steps in establishing a new, innovative multiscale data set suitable to benchmark urban flood modelling. The final data set will consist of high-resolution 3D mapping data acquired from different airborne platforms, focusing on the use of drones (optical and LiDAR). The case study includes residential as well as rural areas in Dudelange/Luxembourg, which have been prone to localized flash flooding following strong rainfall events in recent years. The project also represents a cross-disciplinary collaboration between the geospatial and flood modelling community. In this paper, we introduce the first steps to build up a new benchmark data set together with some initial flood modelling results. More detailed investigations will follow in the next phases of this project. [less ▲]

Detailed reference viewed: 140 (18 UL)
See detailMerging DEMs from VHR Optical Imagery with Drone Data - A High-resolution DEM for Tristan da Cunha
Backes, Dietmar UL; Teferle, Felix Norman UL

Scientific Conference (2018, December 12)

The extraction of high-resolution, Digital Elevation Models (DEM) from very high-resolution (VHR) optical satellite imagery, as well as low altitude drone images by Photogrammetric methods or modern ... [more ▼]

The extraction of high-resolution, Digital Elevation Models (DEM) from very high-resolution (VHR) optical satellite imagery, as well as low altitude drone images by Photogrammetric methods or modern Structure from Motion (SFM) engines, has rapidly matured. Today both data sources are representing cost-effective alternatives to dedicated airborne sensors, especially for remote and difficult to access regions. Ever-growing archives of high-resolution Satellite imagery, are providing a rich data source which covers even the most remote locations with high-resolution imagery up to 0.30m ground sample distance multiple times enabling the generation of high-resolution DEMS. Furthermore, low-cost, low weight and easy to use drones can easily be deployed in remote regions and capture limited areas with very high resolution. Dense point clouds derived from this method provide an invaluable data source to fill the gap between globally available low-resolution DEMs and highly accurate terrestrial surveys. The presented case study investigates the use of VHR archive imagery as well as low-cost drone imagery to generate high-quality DEMs using photogrammetric tools over a remote region which is difficult to access by manned airborne platforms. We highlight the potential and limitations of both data sources to provide high resolution, accurate elevation data. [less ▲]

Detailed reference viewed: 93 (3 UL)
Full Text
See detailTowards multiscale data fusion of high-resolution space borne and terrestrial datasets over Tristan da Cunha
Backes, Dietmar UL; Teferle, Felix Norman UL; Abraha, Kibrom Ebuy UL et al

Poster (2018, April 10)

Ever improving low cost, lightweight and easy to use sensing technologies are enabling the capture of rich 3D Datasets to support an unprecedented range of applications in Geosciences. Especially low-cost ... [more ▼]

Ever improving low cost, lightweight and easy to use sensing technologies are enabling the capture of rich 3D Datasets to support an unprecedented range of applications in Geosciences. Especially low-cost LiDAR systems as well as optical sensors, which can be deployed from terrestrial or low altitude aerial platforms, allow the collection of large datasets without detailed expert knowledge or training. Dense pointcloud derived from these technologies provide an invaluable source to fill the gap between highly precise and accurate terrestrial topographic surveys and large area Digital Surface Models (DSMs) derived from airborne and spaceborne sensors. However, the collection of reliable 3D pointclouds in remote and hazardous locations remains to be very difficult and costly. Establishing a reliable georeference, ensuring accuracy and data quality as well as merging such rich datasets with existing or space borne mapping provide additional challenges. The presented case study investigates the data quality and integration of a heterogeneous dataset collected over the remote island of Tristan da Cunha. High-resolution 3D pointclouds derived by TLS and drone Photogrammetry are merged with space borne imagery while preserving the accurate georeference provided by Ground Control derived from geodetic observations. The volcanic island of Tristan da Cunha located in the centre of the Southern Atlantic Ocean is one of the most remote and difficult to access locations on the planet. Its remote location, rough climatic conditions and consistent cloud coverage provides exceptional challenges for terrestrial, aerial as well as space borne data acquisition. Amongst many other scientific installations, the island also hosts a continuous GNSS station observation and monitoring facilities operated by the University of Luxembourg, which provided the opportunity to conduct a local terrestrial data acquisition campaign consistent with a terrestrial ground survey, Laserscanning and an image acquisition from a low-cost drone. The highly accurate Ground Control network, observed by GNSS and total station, provides a reliable georeference. Pointclouds were acquired around the area of the harbour using a Leica P20 terrestrial Laserscanner, as well as drone Photogrammetry based on images collected by a low-cost DJI Phantom3 drone. To produce a map of the complete island a comprehensive dataset of high-resolution space borne imagery based on the Digital Globe WorldView constellation was acquired which provided high resolution mapping information. The case study presents a cross-validation of terrestrial, low altitude airborne as well as spaceborne datasets in terms coregistration, absolute georeference, scale, resolution and overall data quality. Following the evaluation a practical approach to fuse this heterogeneous dataset is applied which aims to preserve overall data quality, local resolution and accurate georeference and avoid edge artefacts. The conclusions drawn from our preliminary results provide some good practice advice for similar projects. The final topographic dataset enables mapping and monitoring of local geohazards as, e.g. coastal erosion and recent landslides thus also supporting the local population. [less ▲]

Detailed reference viewed: 202 (35 UL)