scholarly journals Generating Elevation Surface from a Single RGB Remotely Sensed Image Using Deep Learning

2020 ◽  
Vol 12 (12) ◽  
pp. 2002 ◽  
Author(s):  
Emmanouil Panagiotou ◽  
Georgios Chochlakis ◽  
Lazaros Grammatikopoulos ◽  
Eleni Charou
Keyword(s):  
Deep Learning ◽  
Satellite Imagery ◽  
Image Data ◽  
Autonomous Driving ◽  
Generative Adversarial Networks ◽  
Electromagnetic Spectrum ◽  
Digital Elevation ◽  
Elevation Data ◽  
Specialized Equipment ◽  
Available Information

Generating Digital Elevation Models (DEM) from satellite imagery or other data sources constitutes an essential tool for a plethora of applications and disciplines, ranging from 3D flight planning and simulation, autonomous driving and satellite navigation, such as GPS, to modeling water flow, precision farming and forestry. The task of extracting this 3D geometry from a given surface hitherto requires a combination of appropriately collected corresponding samples and/or specialized equipment, as inferring the elevation from single image data is out of reach for contemporary approaches. On the other hand, Artificial Intelligence (AI) and Machine Learning (ML) algorithms have experienced unprecedented growth in recent years as they can extrapolate rules in a data-driven manner and retrieve convoluted, nonlinear one-to-one mappings, such as an approximate mapping from satellite imagery to DEMs. Therefore, we propose an end-to-end Deep Learning (DL) approach to construct this mapping and to generate an absolute or relative point cloud estimation of a DEM given a single RGB satellite (Sentinel-2 imagery in this work) or drone image. The model has been readily extended to incorporate available information from the non-visible electromagnetic spectrum. Unlike existing methods, we only exploit one image for the production of the elevation data, rendering our approach less restrictive and constrained, but suboptimal compared to them at the same time. Moreover, recent advances in software and hardware allow us to make the inference and the generation extremely fast, even on moderate hardware. We deploy Conditional Generative Adversarial networks (CGAN), which are the state-of-the-art approach to image-to-image translation. We expect our work to serve as a springboard for further development in this field and to foster the integration of such methods in the process of generating, updating and analyzing DEMs.

scholarly journals Making Radiomics More Reproducible across Scanner and Imaging Protocol Variations: A Review of Harmonization Methods

2021 ◽  
Vol 11 (9) ◽  
pp. 842
Author(s):  
Shruti Atul Mali ◽  
Abdalla Ibrahim ◽  
Henry C. Woodruff ◽  
Vincent Andrearczyk ◽  
Henning Müller ◽  
...  
Keyword(s):  
Deep Learning ◽  
Domain Knowledge ◽  
Data Augmentation ◽  
Image Data ◽  
Clinical Decision ◽  
Generative Adversarial Networks ◽  
Imaging Protocol ◽  
Image Domain ◽  
Style Transfer ◽  
Augmentation Techniques

Radiomics converts medical images into mineable data via a high-throughput extraction of quantitative features used for clinical decision support. However, these radiomic features are susceptible to variation across scanners, acquisition protocols, and reconstruction settings. Various investigations have assessed the reproducibility and validation of radiomic features across these discrepancies. In this narrative review, we combine systematic keyword searches with prior domain knowledge to discuss various harmonization solutions to make the radiomic features more reproducible across various scanners and protocol settings. Different harmonization solutions are discussed and divided into two main categories: image domain and feature domain. The image domain category comprises methods such as the standardization of image acquisition, post-processing of raw sensor-level image data, data augmentation techniques, and style transfer. The feature domain category consists of methods such as the identification of reproducible features and normalization techniques such as statistical normalization, intensity harmonization, ComBat and its derivatives, and normalization using deep learning. We also reflect upon the importance of deep learning solutions for addressing variability across multi-centric radiomic studies especially using generative adversarial networks (GANs), neural style transfer (NST) techniques, or a combination of both. We cover a broader range of methods especially GANs and NST methods in more detail than previous reviews.

scholarly journals Post-Disaster Building Damage Detection from Earth Observation Imagery Using Unsupervised and Transferable Anomaly Detecting Generative Adversarial Networks

2020 ◽  
Vol 12 (24) ◽  
pp. 4193
Author(s):  
Sofia Tilon ◽  
Francesco Nex ◽  
Norman Kerle ◽  
George Vosselman
Keyword(s):  
Deep Learning ◽  
Damage Detection ◽  
Satellite Imagery ◽  
Building Damage ◽  
Training Data ◽  
Training Dataset ◽  
Generative Adversarial Networks ◽  
Generative Adversarial Network ◽  
Custom Made ◽  
Post Disaster

We present an unsupervised deep learning approach for post-disaster building damage detection that can transfer to different typologies of damage or geographical locations. Previous advances in this direction were limited by insufficient qualitative training data. We propose to use a state-of-the-art Anomaly Detecting Generative Adversarial Network (ADGAN) because it only requires pre-event imagery of buildings in their undamaged state. This approach aids the post-disaster response phase because the model can be developed in the pre-event phase and rapidly deployed in the post-event phase. We used the xBD dataset, containing pre- and post- event satellite imagery of several disaster-types, and a custom made Unmanned Aerial Vehicle (UAV) dataset, containing post-earthquake imagery. Results showed that models trained on UAV-imagery were capable of detecting earthquake-induced damage. The best performing model for European locations obtained a recall, precision and F1-score of 0.59, 0.97 and 0.74, respectively. Models trained on satellite imagery were capable of detecting damage on the condition that the training dataset was void of vegetation and shadows. In this manner, the best performing model for (wild)fire events yielded a recall, precision and F1-score of 0.78, 0.99 and 0.87, respectively. Compared to other supervised and/or multi-epoch approaches, our results are encouraging. Moreover, in addition to image classifications, we show how contextual information can be used to create detailed damage maps without the need of a dedicated multi-task deep learning framework. Finally, we formulate practical guidelines to apply this single-epoch and unsupervised method to real-world applications.

scholarly journals A study on Anomaly Detection GAN-based methods on image data

2019 ◽  
Author(s):  
Emanuel Silva ◽  
Johannes Lochter
Keyword(s):  
Deep Learning ◽  
Anomaly Detection ◽  
Image Data ◽  
Original Method ◽  
Detection Task ◽  
Generative Adversarial Networks ◽  
High Dimensional ◽  
Complex Data ◽  
New Techniques ◽  
Adversarial Networks

The anomaly detection task is a well know problem being researched among a variety of areas, including machine learning. The task is to identify data patterns that have a non expected behaviour, that can be a malicious data sent by an attacker or a unexpected valid behaviour, in both cases the anomaly need to be identified. With the advance of deep learning based techniques showing that this class of algorithms can learn high-dimensional and complex data patterns, naturally it became an option to the anomaly detection task. Recent researches in literature are using a sub-field of deep learning algorithms named Generative Adversarial Networks for predicting anomalous samples, since the original method can learn the data distribution. These new techniques make some changes for the anomaly detection task, and this work provides a briefly review on these methods and provides a comparison with well known methods.

scholarly journals A comparison of deep learning algorithms on image data for detecting floodwater on roadways

2021 ◽  
pp. 58-58
Author(s):  
Salih Sarp ◽  
Murat Kuzlu ◽  
Yanxiao Zhao ◽  
Mecit Cetin ◽  
Ozgur Guler
Keyword(s):  
Object Detection ◽  
Traffic Management ◽  
Image Data ◽  
Autonomous Driving ◽  
Management Systems ◽  
Generative Adversarial Networks ◽  
Segmentation Method ◽  
Adversarial Networks ◽  
Segmentation Algorithms ◽  
Natural Settings

Object detection and segmentation algorithms evolved significantly in the last decade. Simultaneous object detection and segmentation paved the way for real-time applications such as autonomous driving. Detection and segmentation of (partially) flooded roadways are essential inputs for vehicle routing and traffic management systems. This paper proposes an automatic floodwater detection and segmentation method utilizing the Mask Region-Based Convolutional Neural Networks (Mask-R-CNN) and Generative Adversarial Networks (GAN) algorithms. To train the model, manually labeled images with urban, suburban, and natural settings are used. The performances of the algorithms are assessed in accurately detecting the floodwater captured in images. The results show that the proposed Mask-R-CNN-based floodwater detection and segmentation outperform previous studies, whereas the GAN-based model has a straightforward implementation compared to other models.

scholarly journals Applying Airborne LiDAR to Map Salt Marsh Inland Boundaries

2021 ◽  
Vol 13 (21) ◽  
pp. 4245
Author(s):  
Lee B. van Ardenne ◽  
Gail L. Chmura
Keyword(s):  
Salt Marsh ◽  
Satellite Imagery ◽  
Salt Marshes ◽  
Airborne Lidar ◽  
Tidal Elevation ◽  
Digital Elevation ◽  
Elevation Data ◽  
Edge Location ◽  
The Difference ◽  
Edge Based

The determination of rates and stocks of carbon storage in salt marshes, as well as their protection, require that we know where they and their boundaries are. Marsh boundaries are conventionally mapped through recognition of plant communities using aerial photography or satellite imagery. We examined the possibility of substituting the use of 1 m resolution LiDAR-derived digital elevation models (DEMs) and tidal elevations to establish salt marsh upper boundaries on the New Brunswick coasts of the Gulf of St. Lawrence and the Bay of Fundy, testing this method at tidal ranges from ≤2 to ≥4 m. LiDAR-mapped marsh boundaries were verified with high spatial resolution satellite imagery and a subset through field mapping of the upland marsh edge based upon vegetation and soil characteristics, recording the edge location and elevation with a Differential Geographic Positioning System. The results show that the use of high-resolution LiDAR and tidal elevation data can successfully map the upper boundary of salt marshes without the need to first map plant species. The marsh map area resulting from our mapping was ~30% lower than that in the province’s aerial-photograph-based maps. However, the difference was not primarily due to the location of the upper marsh boundaries but more so because of the exclusion of mudflats and large creeks (features that are not valued as carbon sinks) using the LiDAR method that are often mapped as marsh areas in the provincial maps. Despite some minor limitations, the development of DEMs derived from LiDAR can be applied to update and correct existing salt marsh maps along extensive sections of coastlines in less time than required to manually trace from imagery. This is vital information for governments and NGOs seeking to conserve these environments, as accurate mapping of the location and area of these ecosystems is a necessary basis for conservation prioritization indices.

Deep Learning Approaches for Whiteboard Image Quality Enhancement

2019 ◽  
Vol 2019 (1) ◽  
pp. 360-368
Author(s):  
Mekides Assefa Abebe ◽  
Jon Yngve Hardeberg
Keyword(s):  
Deep Learning ◽  
Image Quality ◽  
Image Data ◽  
Quality Enhancement ◽  
Network Architectures ◽  
Learning Approaches ◽  
Data Set ◽  
Image Quality Enhancement ◽  
Processing Techniques ◽  
White Balancing

Different whiteboard image degradations highly reduce the legibility of pen-stroke content as well as the overall quality of the images. Consequently, different researchers addressed the problem through different image enhancement techniques. Most of the state-of-the-art approaches applied common image processing techniques such as background foreground segmentation, text extraction, contrast and color enhancements and white balancing. However, such types of conventional enhancement methods are incapable of recovering severely degraded pen-stroke contents and produce artifacts in the presence of complex pen-stroke illustrations. In order to surmount such problems, the authors have proposed a deep learning based solution. They have contributed a new whiteboard image data set and adopted two deep convolutional neural network architectures for whiteboard image quality enhancement applications. Their different evaluations of the trained models demonstrated their superior performances over the conventional methods.

F1_Tenth_Course_Report_Group_C

2018 ◽  
Author(s):  
Yi Chen ◽  
Sagar Manglani ◽  
Roberto Merco ◽  
Drew Bolduc
Keyword(s):  
Operating System ◽  
Autonomous Driving ◽  
Software Tools ◽  
Simulation Environment ◽  
Robot Operating System ◽  
Available Information

In this paper, we discuss several of major robot/vehicle platforms available and demonstrate the implementation of autonomous techniques on one such platform, the F1/10. Robot Operating System was chosen for its existing collection of software tools, libraries, and simulation environment. We build on the available information for the F1/10 vehicle and illustrate key tools that will help achieve properly functioning hardware. We provide methods to build algorithms and give examples of deploying these algorithms to complete autonomous driving tasks and build 2D maps using SLAM. Finally, we discuss the results of our findings and how they can be improved.

Urban Flood Mapping

2017 ◽  
Author(s):  
Indra Riyanto ◽  
Lestari Margatama
Keyword(s):  
Natural Disasters ◽  
Satisfying Condition ◽  
Three Dimensions ◽  
Urban Flood ◽  
Digital Elevation ◽  
Elevation Data ◽  
Recent Occurrence ◽  
Elevation Model ◽  
Digital Elevation Model Data ◽  
Potential Area

The recent degradation of environment quality becomes the prime cause of the recent occurrence of natural disasters. It also contributes in the increase of the area that is prone to natural disasters. Flood history data in Jakarta shows that flood occurred mainly during rainy season around January – February each year, but the flood area varies each year. This research is intended to map the flood potential area in DKI Jakarta by segmenting the Digital Elevation Model data. The data used in this research is contour data obtained from DPP–DKI with the resolution of 1 m. The data processing involved in this research is extracting the surface elevation data from the DEM, overlaying the river map of Jakarta with the elevation data. Subsequently, the data is then segmented using watershed segmentation method. The concept of watersheds is based on visualizing an image in three dimensions: two spatial coordinates versus gray levels, in which there are two specific points; that are points belonging to a regional minimum and points at which a drop of water, if placed at the location of any of those points, would fall with certainty to a single minimum. For a particular regional minimum, the set of points satisfying the latter condition is called the catchments basin or watershed of that minimum, while the points satisfying condition form more than one minima are termed divide lines or watershed lines. The objective of this segmentation is to find the watershed lines of the DEM image. The expected result of the research is the flood potential area information, especially along the Ciliwung river in DKI Jakarta.

Sign in / Sign up

Export Citation Format

Share Document