Article
Environmental Sciences
Daniele Ventura, Luca Grosso, Davide Pensa, Edoardo Casoli, Gianluca Mancini, Tommaso Valente, Michele Scardi, Arnold Rakaj
Summary: This study evaluated an integrated approach using low-cost unmanned aerial and surface vehicles to collect detailed remote sensing data and accurately map shallow benthic communities. Photogrammetric outputs from UAV and USV were classified using OBIA approach and achieved overall classification accuracies over 70%. The results demonstrated the practicality and feasibility of using aerial and underwater ultra-high spatial resolution imagery for detailed analysis.
FRONTIERS IN MARINE SCIENCE
(2023)
Article
Environmental Sciences
Chuanyun Wang, Yang Su, Jingjing Wang, Tian Wang, Qian Gao
Summary: With the development of unmanned aerial vehicle (UAV) technology and swarm intelligence technology, the study focuses on the threat and challenge brought by UAV swarms to low-altitude airspace defense. A dataset named UAVSwarm is manually annotated for UAV swarm detection and tracking, which includes various scenes and types of UAVs. Advanced detection and multi-object tracking models are used for comprehensive testing and performance verification. The experimental results show the dataset's availability and usability for training and testing various UAV detection and swarm tracking tasks.
Article
Computer Science, Information Systems
Daniele Palossi, Nicky Zimmerman, Alessio Burrello, Francesco Conti, Hanna Mueller, Luca Maria Gambardella, Luca Benini, Alessandro Giusti, Jerome Guzzi
Summary: This research focuses on achieving complex tasks with nano-sized unmanned aerial vehicles, specifically in estimating and maintaining the relative 3-D pose of the UAV with respect to a person. The study utilizes a vision-based deep neural network and ultra-low power processor for real-time autonomous navigation, demonstrating excellent performance.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Marine
Francisco Contreras-de-Villar, Francisco J. Garcia, Juan J. Munoz-Perez, Antonio Contreras-de-Villar, Veronica Ruiz-Ortiz, Patricia Lopez, Santiago Garcia-Lopez, Bismarck Jigena
Summary: Research has shown that various factors such as flight time, frame overlap, and the number of GCPs can significantly impact the accuracy of beach area mapping with RPAS. In general, conducting flights in the early morning can help reduce errors and improve accuracy in surveys.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2021)
Article
Engineering, Multidisciplinary
Ismail Elkhrachy
Summary: This study aimed to produce accurate geospatial 3D data from UAV images. The solution generated met the 2015 ASPRS accuracy standards, with horizontal RMSE values of 4-6 cm and vertical accuracy of 5-6 cm, which were twice and three times the Ground Sample Distance (GSD), respectively.
ALEXANDRIA ENGINEERING JOURNAL
(2021)
Article
Computer Science, Interdisciplinary Applications
Carlos Alberto Villarreal, Carlos Guillermo Garzon, Jose Pedro Mora, Julian David Rojas, Carlos Alberto Rios
Summary: This paper presents a methodological approach for capturing difficult-to-access geological outcrops using unmanned aerial vehicle-based digital photogrammetric data. The obtained data can be used for geomodelling of mineral deposits and oil and gas geological structures.
JOURNAL OF INDUSTRIAL INFORMATION INTEGRATION
(2022)
Article
Green & Sustainable Science & Technology
Patricia Arranz, Fredrik Christiansen, Maria Glarou, Shane Gero, Fleur Visser, Machiel G. Oudejans, Natacha Aguilar de Soto, Kate Sprogis
Summary: This study examined the body shape, allometric relationships, and body condition of short-finned pilot whales in the North Atlantic. The researchers used unmanned aerial vehicles to measure the body length, width, and height of the whales. They found that there was no difference in body condition among reproductive classes or locations.
Article
Environmental Sciences
Luka Jurjevic, Mateo Gasparovic, Xinlian Liang, Ivan Balenovic
Summary: This study evaluated the accuracy of close-range remote sensing techniques for DTM data collection in forest areas, and found that these techniques can achieve higher accuracy compared to airborne laser scanning and digital aerial photogrammetry data.
Article
Forestry
Facundo Pessacg, Francisco Gomez-Fernandez, Matias Nitsche, Nicolas Chamo, Sebastian Torrella, Ruben Ginzburg, Pablo De Cristoforis
Summary: Forestry aerial photogrammetry using Unmanned Aerial Systems (UAS) helps bridge the gap between detailed fieldwork and broad-range satellite imagery-based analysis. However, optical sensors face limitations in collecting and classifying ground points in woodlands. This study proposes a novel method to generate accurate Digital Terrain Models (DTMs) in forested environments and develops a realistic simulator for controlled experimentation.
Article
Automation & Control Systems
Dongyu Li, Kun Cao, Linghuan Kong, Haoyong Yu
Summary: This article investigates cooperative circumnavigation for groups of networked UAVs under a directed interaction topology, proposing a structured control design based on affine transformations. Leader UAVs can fully manipulate follower UAVs without the need for global information, achieving efficient spatial formation without the requirement of information exchange.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2021)
Article
Environmental Sciences
Juan Pedro Carbonell-Rivera, Jesus Torralba, Javier Estornell, Luis Angel Ruiz, Pablo Crespo-Peremarch
Summary: This study investigates the use of UAV-DAP point clouds to classify tree and shrub species in Mediterranean forests. Machine learning methods are used to achieve accurate classification results, which provide important information for generating fuel variables in wildfire models.
Article
Robotics
Wooyong Park, Xiangyu Wu, Dongjae Lee, Seung Jae Lee
Summary: Existing multirotor-based cargo transportation lacks constant cargo attitude due to underactuation, while fragile payloads require a consistent posture. We propose a new fully-actuated multirotor UAV platform capable of translational motion while maintaining a constant attitude. To address the change in center-of-mass (CoM) position when loading cargo, we introduce a model-free CoM estimation method inspired by the extremum-seeking control (ESC) technique. Experimental results validate the effectiveness of the proposed estimation method in achieving satisfactory constant-attitude flight performance.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Geosciences, Multidisciplinary
Guoxun Zheng, Zhengang Jiang, Hua Zhang, Xuekun Yao
Summary: In the era of artificial intelligence and big data, semantic segmentation of images is crucial for various applications. Deep learning based semantic segmentation methods, specifically the fully convolutional neural network (FCN), have shown to be efficient and highly accurate. The author applied FCN-32s, FCN-16s, and FCN-8s to a UAV remote sensing image dataset and achieved stable accuracy rates of around 94% for vegetation recognition and over 88% for road recognition. The overall mean pixel accuracy rate was above 91%. This demonstrates that utilizing FCN in semantic segmentation of UAV remote sensing images significantly improves efficiency and accuracy.
FRONTIERS IN EARTH SCIENCE
(2023)
Article
Computer Science, Interdisciplinary Applications
Juan Pedro Carbonell-Rivera, Javier Estornell, Luis Angel Ruiz, Pablo Crespo-Peremarch, Jaime Almonacid-Caballer
Summary: This study presents Class3Dp, a software for classifying vegetation species in colored point clouds. The software utilizes geometric, spectral, and neighborhood features along with machine learning methods to classify the point cloud, allowing for the recognition of species composition in an ecosystem.
ENVIRONMENTAL MODELLING & SOFTWARE
(2024)
Article
Multidisciplinary Sciences
Amy P. Bogolin, Drew R. Davis, Richard J. Kline, Abdullah F. Rahman
Summary: The study investigated the use of drones and high-resolution cameras to survey large freshwater turtle populations in arid riverine landscapes, demonstrating the great potential of drone-based surveys.
Article
Geography, Physical
Zille Hussnain, Sander Oude Elberink, George Vosselman
Summary: The study focuses on improving trajectory accuracy in urban canyons by using corresponding points between aerial images and MLS data. Results indicate that this method can achieve near-decimetre level geo-referencing of point clouds in urban environments.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Editorial Material
Environmental Sciences
Monika Kuffer, Karin Pfeffer, Claudio Persello
Article
Remote Sensing
S. Briechle, P. Krzystek, G. Vosselman
Summary: The study introduces an approach, Silvi-Net, based on convolutional neural networks for 3D tree classification by fusing airborne lidar data and multispectral images. The method showed promising results on data from two natural forest areas, demonstrating its effectiveness in applications such as automated inventory and monitoring projects.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Geography, Physical
Yaping Lin, George Vosselman, Yanpeng Cao, Michael Ying Yang
Summary: In this paper, a local and global encoder network (LGENet) for semantic segmentation of ALS point clouds is introduced, which achieves good performance on the ISPRS benchmark dataset and DCF2019 dataset by using different feature extraction and global information utilization methods.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Ecology
Bassam Qarallah, Malik Al-Ajlouni, Ayman Al-Awasi, Mohammad Alkarmy, Emad Al-Qudah, Ahmad Bani Naser, Amani Al-Assaf, Caroline M. Gevaert, Yolla Al Asmar, Mariana Belgiu, Yahia A. Othman
Summary: By using remote sensing data and field measurements, we studied the fire severity and recovery process of the Latroon dry forest in Jordan after the 2003 fire, finding that about 65% of the forest was burned. UAV assessments showed that 90% of the burned area had recovered to pre-fire conditions by 2020, but tree heights in severely burned areas were significantly lower than those in moderately burned areas.
JOURNAL OF ARID ENVIRONMENTS
(2021)
Article
Environmental Sciences
Xiaoyu Sun, Wufan Zhao, Raian Maretto, Claudio Persello
Summary: This study investigates the combination of normalized digital surface models (nDSMs) with aerial images to optimize the extraction of building polygons using the frame field learning method, resulting in improved precision and regularity of building outlines. The method achieves a mean intersection over union (IoU) of 0.80 with the fused data (RGB + nDSM) compared to an IoU of 0.57 with the baseline (using RGB only), demonstrating reduced false positives and increased positional accuracy.
Article
Geography, Physical
Yaping Lin, George Vosselman, Michael Ying Yang
Summary: This paper proposes a weakly supervised approach for semantic segmentation of ALS point clouds, achieving efficient and accurate segmentation on large-scale datasets through the use of pseudo labels and improved network structures. Experimental results demonstrate that the method outperforms traditional approaches in terms of performance.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Ning Zhang, Francesco Nex, George Vosselman, Norman Kerle
Summary: This research focuses on using deep learning to detect victims in disaster debris, proposes a method to generate harmonious composite images for training, and significantly improves detection accuracy.
Article
Environmental Sciences
Nasir Farsad Layegh, Roshanak Darvishzadeh, Andrew K. Skidmore, Claudio Persello, Nina Krueger
Summary: This study aims to develop a classification method by integrating graph-based semi-supervised learning (SSL) and an expert system (ES) to improve the accuracy of vegetation classification. The results show that this method has higher accuracy compared to other classification methods, and using all red-edge spectral band combinations yields the best results.
Article
Engineering, Civil
Fashuai Li, Zhize Zhou, Jianhua Xiao, Ruizhi Chen, Matti Lehtomaki, Sander Oude Elberink, George Vosselman, Juha Hyyppa, Yuwei Chen, Antero Kukko
Summary: This paper presents an improved framework for instance-aware semantic segmentation of road furniture in mobile laser scanning data. The framework detects road furniture, decomposes them into poles and components, extracts instance information, and classifies the components using a classifier and DenseCRF. The combination of random forest with DenseCRF achieves high overall accuracies.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Editorial Material
Environmental Sciences
Mila Koeva, Rohan Bennett, Claudio Persello
Summary: Contemporary land administration systems incorporate cadastre and land registration concepts and play a crucial role in global land management. The use of innovative remote sensing techniques, such as unmanned aerial vehicles and satellite-based acquisitions, provides high-resolution spatial information for improved land management. This Special Issue aims to explore the usage and integration of emerging remote sensing techniques in the land administration domain.
Article
Computer Science, Information Systems
Abbas Najmi, Caroline M. M. Gevaert, Divyani Kohli, Monika Kuffer, Jati Pratomo
Summary: This research aims to integrate Remote Sensing Imagery (RSI) and Street View Images (SVI) for slum mapping. The experimental results demonstrate that combining RSI and SVI improves the accuracy of slum mapping, depending on how and at what level they are integrated in the network.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Review
Remote Sensing
Caroline M. Gevaert
Summary: This paper reviews examples of explainable machine learning and explainable artificial intelligence in the field of Earth Observation, classifying the methods and identifying limitations. The findings indicate a lack of clarity, with explanations often targeting domain experts and lacking testing for usefulness to the intended audience.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Proceedings Paper
Geography, Physical
A. Maiti, S. J. Oude Elberink, G. Vosselman
Summary: This paper investigates the impact of label noise on the performance of deep learning models in semantic segmentation. Experimental results show that label noise decreases the accuracy of the model, and different classes respond differently to label noise.
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION II
(2022)
Proceedings Paper
Geography, Physical
N. Zhang, F. Nex, G. Vosselman, N. Kerle
Summary: This paper addresses the issue of deep detection networks in detecting buried victims. By generating realistic images and using an unsupervised generative adversarial network for harmonization, the accuracy of victim detection can be effectively improved.
XXIV ISPRS CONGRESS: IMAGING TODAY, FORESEEING TOMORROW, COMMISSION III
(2022)