Search

Abstract: Erosion, in particular gully erosion, is a widespread problem. Its mapping is crucial for erosion monitoring and remediation of degraded areas. In addition, mapping of areas with high potential for future gully erosion can be used to assist prevention strategies. Good relations with topographic variables collected from the field are appropriate for determining areas susceptible to gullying. Image analysis of high resolution remotely sensed imagery (HRI) in combination with field verification has proven to be a good approach, although dependent on expensive imagery. Automatic and semi-automatic methods, such as object-oriented analysis (OOA), are rapid and reproducible. However, HRI data are not always available. We therefore attempted to identify gully systems using statistical modeling of image features from medium resolution imagery, here ASTER. These data were used for determining areas within gully system boundaries (GSB) using a semi-automatic method based on OOA. We assess if the selection of useful object features can be done in an objective and transferable way, using Random Forests (RF) for prediction of gully systems at regional scale, here in the Sehoul region, near Rabat, Morocco. Moderate success was achieved using a semi-automatic object-based RF model (out-of-bag error of 18.8%). Besides compensating for the imbalance between gully and non-gully classes, the procedure followed in this study enabled us to balance the classification error rates. The user's and producer's accuracy of the data with a balanced set of class showed an improved accuracy of the spatial estimates of gully systems, when compared to the data with imbalanced class. The model over-predicted the area within the GSB (13–27%), but its overall performance demonstrated that medium resolution satellite images contain sufficient information to identify gully systems, so that large areas can be mapped with relatively little effort and acceptable accuracy. [Copyright &y& Elsevier]

Abstract: In contrast to the many studies that use expert-based analysis of LiDAR derivatives for landslide mapping in forested terrain, only few studies have attempted to develop (semi-)automatic methods for extracting landslides from LiDAR derivatives. While all these studies are pixel-based, it has not yet been tested whether object-oriented analysis (OOA) could be an alternative. This study investigates the potential of OOA using only single-pulse LiDAR derivatives, such as slope gradient, roughness and curvature to map landslides. More specifically, the focus is on both LiDAR data segmentation and classification of slow-moving landslides in densely vegetated areas, where spectral data do not allow accurate landslide identification. A multistage procedure has been developed and tested in the Flemish Ardennes (Belgium). The procedure consists of (1) image binarization and multiresolution segmentation, (2) classification of landslide parts (main scarps and landslide body segments) and non-landslide features (i.e. earth banks and cropland fields) with supervised support vector machines at the appropriate scale, (3) delineation of landslide flanks, (4) growing of a landslide body starting from its main scarp, and (5) final cleaning of the inventory map. The results obtained show that OOA using LiDAR derivatives allows recognition and characterization of profound morphologic properties of forested deep-seated landslides on soil-covered hillslopes, because more than 90% of the main scarps and 70% of the landslide bodies of an expert-based inventory were accurately identified with OOA. For mountainous areas with bedrock, on the other hand, creation of a transferable model is expected to be more difficult. [Copyright &y& Elsevier]

Abstract: Object-oriented analysis (OOA) has been demonstrated to produce more accurate results than pixel-based image processing. Studies carried out by previous researchers have shown how landslide inventories can be prepared from multispectral satellite images using OOA. However, panchromatic images are frequently the only data available after a landslide event. Furthermore, preparation of historical inventories relies on the analysis of satellite images and aerial photographs acquired over past few decades that are also mostly only available in black and white. In such cases the methodology developed using multispectral data cannot be used directly due to limited spectral information, in particular in near-infrared bands. In this paper we present a new methodology that addresses some of these issues. Using high resolution panchromatic images from Cartosat-1 (2.5m) and IRS-1D (5.8m), and a 10m gridded DTM extracted from Cartosat-1, we developed a new approach which uses change detection techniques and a global contextual criteria in an object-based environment to detect and classify landslides into five different types. Continuous time series images from 1998 to 2006 were used to prepare annual landslide inventories in a highly rugged Himalayan terrain. The maximum and minimum detection percentages achieved for all landslides are 96.7% and 71.5%, respectively, with corresponding quality percentages of 88.1% and 55.3%, respectively. However, the lack of spectral information proved to be a hurdle resulting in a high branching factor that indicates that further work is required to eliminate false positives. Nevertheless, the method was able to create much needed historical landslide inventories, which are critical for landslide hazard and risk assessment studies. [Copyright &y& Elsevier]

Abstract: Gully erosion is responsible for a substantial amount of soil loss and is generally considered an indicator of desertification. Hence, mapping these gully features provides essential information needed on sediment production, identification of vulnerable areas for gully formation, land degradation, and environmental and socio-economical effects. This paper investigates the use of object-oriented image analysis (OOA) to extract gully erosion features from satellite imagery, using a combination of topographic, spectral, shape (geometric) and contextual information obtained from IKONOS and GEOEYE-1 data. A rule-set was developed and tested for a semi-arid to sub-humid region in Morocco. The percentage of gully system area indicated negligible overestimations between the reference area and the OOA area in two sub-watersheds (0.03% and 1.77%). We also observed that finer gully-related edges within the complex gully systems were better identified semi-automatically than was possible by manual digitization, suggesting higher detection accuracy. OOA-based gully mapping is quicker and more objective than traditional methods, and is thus better suited to provide essential information for land managers to support their decision making processes, and for the erosion research community. [Copyright &y& Elsevier]

Abstract: Landslide inventory mapping is an indispensable prerequisite for reliable hazard and risk analysis, and with the increasing availability of very high resolution (VHR) remote sensing imagery the creation and updating of such inventories on regular bases and directly after major events is becoming possible. The diversity of landslide processes and spectral similarities of affected areas with other landscape elements pose major challenges for automated image processing, and time-consuming manual image interpretation and field surveys are still the most commonly applied mapping techniques. Taking advantage of recent advances in object-oriented image analysis (OOA) and machine learning algorithms, a supervised workflow is proposed in this study to reduce manual labor and objectify the choice of significant object features and classification thresholds. A sequence of image segmentation, feature selection, object classification and error balancing was developed and tested on a variety of sample datasets (Quickbird, IKONOS, Geoeye-1, aerial photographs) of four sites in the northern hemisphere recently affected by landslides (Haiti, Italy, China, France). Besides object metrics, such as band ratios and slope, newly introduced topographically-guided texture measures were found to enhance significantly the classification, and also feature selection revealed positive influence on the overall performance. With an iterative procedure to examine the class-imbalance within the training sample it was furthermore possible to compensate spurious effects of class-imbalance and class-overlap on the balance of the error rates. Employing approximately 20% of the data for training, the proposed workflow resulted in accuracies between 73% and 87% for the affected areas, and approximately balanced commission and omission errors. [Copyright &y& Elsevier]

Abstract: The Yogyakarta area in Indonesia suffered a devastating earthquake on 27 May 2006. There was an immediate international response, and the International Charter “Space and Major Disasters” was activated, leading to a rapid production of image-based damage maps and other assistance. Most of the acquired images were processed by UNOSAT and DLR-ZKI, while substantial damage mapping also occurred on the ground. This paper assesses the accuracy and completeness of the damage maps produced based on Charter data, using ground damage information collected during an extensive survey by Yogyakarta''s Gadjah Mada University in the weeks following the earthquake and that has recently become available. More than 54,000 buildings or their remains were surveyed, resulting in an exceptional validation database. The UNOSAT damage maps outlining clusters of severe damage are very accurate, while earlier, more detailed results underestimated damage and missed larger areas. Damage maps produced by DLR-ZKI, using a damage-grid approach, were found to underestimate the extent and severity of the devastation. Both mapping results also suffer from limited image coverage and extensive cloud contamination. The ground mapping gives a more accurate picture of the extent of the damage, but also illustrates the challenge of mapping a vast area. The paper concludes with a discussion on ways to improve Charter-based damage maps by integration of local knowledge, and to create a wider impact through generation of customised mapping products using web map services. [Copyright &y& Elsevier]

Abstract: Recognition and classification of landslides is a critical requirement in pre- and post-disaster hazard analysis. This has been primarily done through field mapping or manual image interpretation. However, image interpretation can also be done semi-automatically by creating a routine in object-based classification using the spectral, spatial and morphometric properties of landslides, and by incorporating expert knowledge. This is a difficult task since a fresh landslide has spectral properties that are nearly identical to those of other natural objects, such as river sand and rocky outcrops, and they also do not have unique shapes. This paper investigates the use of a combination of spectral, shape and contextual information to detect landslides. The algorithm is tested with a 5.8m multispectral data from Resourcesat-1 and a 10m digital terrain model generated from 2.5m Cartosat-1 imagery for an area in the rugged Himalayas in India. It uses objects derived from the segmentation of a multispectral image as classifying units for object-oriented analysis. Spectral information together with shape and morphometric characteristics was used initially to separate landslides from false positives. Objects recognised as landslides were subsequently classified based on material type and movement as debris slides, debris flows and rock slides, using adjacency and morphometric criteria. They were further classified for their failure mechanism using terrain curvature. The procedure was developed for a training catchment and then applied without further modification on an independent catchment. A total of five landslide types were detected by this method with 76.4% recognition and 69.1% classification accuracies. This method detects landslides relatively quickly, and hence has the potential to aid risk analysis, disaster management and decision making processes in the aftermath of an earthquake or an extreme rainfall event. [Copyright &y& Elsevier]

Vast archives of legacy maps exist for most parts of the world, containing analog information on a variety of environmental and socioeconomic parameters, and often dating back to the nineteenth century. The information contained in those data, which is potentially of great utility in environmental change or demographic studies, has traditionally only been accessible through digitizing or visual map analysis. In this paper, we show how object-oriented analysis can be used to unlock such information. We demonstrate this on a 1962 map of Kenya, which shows population distribution using differently sized dots. The procedure developed extracts those population signatures accurately, despite size and color variations, dot bleeding, and conglutination, well as overlap with other map elements. Over 39 000 dots were automatically extracted, corresponding to 99.6 % of the published population figure, with an accuracy between 94.6% and 98.5% for the different symbol sizes. We also discuss the utility of the derived geographic-information-system-ready information, for example, to assess malaria exposure and calculate population change figures, using the 1999 census data at a detailed sublocation level. [ABSTRACT FROM AUTHOR]

Coal fires worldwide are causing enormous economic loss and environmental pollution, prompting a need to develop the methodology to map, analyse, and monitor them. Coal fire-related land cover mapping yields the knowledge of the distribution of coal, which is crucial for the delineation of potential coal fires and coal fire risk areas. The conventional pixel-based method does not make use of spatial relationships, whereas object-based image analysis (OBIA) has the potential to consider not only objects' spectral but spatial information, in terms of size, shape, texture and relations to other image objects. This article reports on coal fire-related land cover mapping with OBIA using multi-resolution segmentation and hierarchical classification. Both spectral and spatial information were used in the classification. It yielded an overall accuracy of 12.3% higher than that by the pixel-based method (70.67%), and the difference is significant (p = 0.05) by McNemar's test. [ABSTRACT FROM AUTHOR]

At least 40,000 deaths have been attributed to historic lahars (volcanic mudflows). The most recent lahar disaster occurred in 1998 at Casita volcano, Nicaragua, claiming over 2,500 lives. Lahars can cover large areas and be highly destructive, and constitute a challenge for disaster management. With infrastructure affected and access frequently impeded, disaster management can benefit from the synoptic coverage provided by satellite imagery. This potential has been recognised for other types of natural disasters, but limitations are also known. Dedicated satellite constellations for disaster response and management have been proposed as one solution. Here we investigate the utility of currently available and forthcoming optical and radar sensors as tools in lahar disaster management. Applied to the Casita case, we find that imagery available at the time could not have significantly improved disaster response. However, forthcoming satellites, especially radar, will improve the situation, reducing the benefit of dedicated constellations. [ABSTRACT FROM AUTHOR]

Analyzes the morphology and structure at Casita volcano in Nicaragua and describes an actively deforming edifice containing a large gravity slide likely to collapse. Theory of volcano flank spreading; Geology of Casita; Data from topographic maps, aerial photographs, radar images and field observation; Analogue models of slope changes and fault patterns.

• An end-to-end deep learning framework for semantic segmentation in low-light indoor scenes outperforming the state-of-the-art DeepLab v3+. • A new data set contains both rendered and real normal-light/low-light image pairs for studying UAVs or robots understanding low-light indoor scenes. Semantic segmentation using convolutional neural networks (CNNs) achieves higher accuracy than traditional methods, but it fails to yield satisfactory results under illumination variants when the training set is limited. In this paper we present a new data set containing both real and rendered images and a novel cascade network to study semantic segmentation in low-light indoor environments. Specifically, the network decomposes a low-light image into illumination and reflectance components, and then a multi-tasking learning scheme is built. One branch learns to reduce noise and restore information on the reflectance (reflectance restoration branch). Another branch learns to segment the reflectance map (semantic segmentation branch). The CNN features from two tasks are concatenated together so as to improve the segmentation accuracy by embedding the illumination-invariant features. We compare our approach with other CNN-based segmentation frameworks, including the state-of-the-art DeepLab v3+, on the proposed real data set, and our approach achieves the highest mIoU (47.6 %). The experimental results also show that the semantic information supports the restoration of a sharper reflectance map, thus further improving the segmentation. Besides, we pre-train a model with the proposed large-scale rendered images and then fine-tune it on the real images. The pre-training results in an improvement of mIoU by 7.2 %. Our models and data set are publicly available for research. This research is part of the EU project INGENIOUS 1 1 https://ingenious-first-responders.eu/. Our data sets and models are available on our website 2 2 https://github.com/noahzn/LISU. [ABSTRACT FROM AUTHOR]

Structural disaster damage detection and characterization is one of the oldest remote sensing challenges, and the utility of virtually every type of active and passive sensor deployed on various air- and spaceborne platforms has been assessed. The proliferation and growing sophistication of unmanned aerial vehicles (UAVs) in recent years has opened up many new opportunities for damage mapping, due to the high spatial resolution, the resulting stereo images and derivatives, and the flexibility of the platform. This study provides a comprehensive review of how UAV-based damage mapping has evolved from providing simple descriptive overviews of a disaster science, to more sophisticated texture and segmentation-based approaches, and finally to studies using advanced deep learning approaches, as well as multi-temporal and multi-perspective imagery to provide comprehensive damage descriptions. The paper further reviews studies on the utility of the developed mapping strategies and image processing pipelines for first responders, focusing especially on outcomes of two recent European research projects, RECONASS (Reconstruction and Recovery Planning: Rapid and Continuously Updated Construction Damage, and Related Needs Assessment) and INACHUS (Technological and Methodological Solutions for Integrated Wide Area Situation Awareness and Survivor Localization to Support Search and Rescue Teams). Finally, recent and emerging developments are reviewed, such as recent improvements in machine learning, increasing mapping autonomy, damage mapping in interior, GPS-denied environments, the utility of UAVs for infrastructure mapping and maintenance, as well as the emergence of UAVs with robotic abilities. [ABSTRACT FROM AUTHOR]

Natural disasters are projected to increase in number and severity, in part due to climate change. At the same time a growing number of disaster risk reduction (DRR) and climate change adaptation measures are being implemented by governmental and non-governmental organizations, and substantial post-disaster donations are frequently pledged. At the same time there has been increasing demand for transparency and accountability, and thus evidence of those measures having a positive effect. We hypothesized that resilience-enhancing interventions should result in less damage during a hazard event, or at least quicker recovery. In this study we assessed recovery over a 3 year period of seven municipalities in the central Philippines devastated by Typhoon Haiyan in 2013. We used very high resolution optical images (<1 m), and created detailed land cover and land use maps for four epochs before and after the event, using a machine learning approach with extreme gradient boosting. The spatially and temporally highly variable recovery maps were then statistically related to detailed questionnaire data acquired by DEval in 2012 and 2016, whose principal aim was to assess the impact of a 10 year land-planning intervention program by the German agency for technical cooperation (GIZ). The survey data allowed very detailed insights into DRR-related perspectives, motivations and drivers of the affected population. To some extent they also helped to overcome the principal limitation of remote sensing, which can effectively describe but not explain the reasons for differential recovery. However, while a number of causal links between intervention parameters and reconstruction was found, the common notion that a resilient community should recover better and more quickly could not be confirmed. The study also revealed a number of methodological limitations, such as the high cost for commercial image data not matching the spatially extensive but also detailed scale of field evaluations, the remote sensing analysis likely overestimating damage and thus providing incorrect recovery metrics, and image data catalogues especially for more remote communities often being incomplete. Nevertheless, the study provides a valuable proof of concept for the synergies resulting from an integration of socio-economic survey data and remote sensing imagery for recovery assessment. [ABSTRACT FROM AUTHOR]

First responders and recovery planners need accurate and quickly derived information about the status of buildings as well as newly built ones to both help victims and to make decisions for reconstruction processes after a disaster. Deep learning and, in particular, convolutional neural network (CNN)-based approaches have recently become state-of-the-art methods to extract information from remote sensing images, in particular for image-based structural damage assessment. However, they are predominantly based on manually extracted training samples. In the present study, we use pre-disaster OpenStreetMap building data to automatically generate training samples to train the proposed deep learning approach after the co-registration of the map and the satellite images. The proposed deep learning framework is based on the U-net design with residual connections, which has been shown to be an effective method to increase the efficiency of CNN-based models. The ResUnet is followed by a Conditional Random Field (CRF) implementation to further refine the results. Experimental analysis was carried out on selected very high resolution (VHR) satellite images representing various scenarios after the 2013 Super Typhoon Haiyan in both the damage and the recovery phases in Tacloban, the Philippines. The results show the robustness of the proposed ResUnet-CRF framework in updating the building map after a disaster for both damage and recovery situations by producing an overall F1-score of 84.2%. [ABSTRACT FROM AUTHOR]

An introduction to the journal is presented which includes articles on the cross-section of the state-of-the art of geographic object-based image analysis (GEOBIA) research and reflect the diversity of work the conference aimed to present.

Post-disaster recovery (PDR) is a complex, long-lasting, resource intensive, and poorly understood process. PDR goes beyond physical reconstruction (physical recovery) and includes relevant processes such as economic and social (functional recovery) processes. Knowing the size and location of the places that positively or negatively recovered is important to effectively support policymakers to help readjust planning and resource allocation to rebuild better. Disasters and the subsequent recovery are mainly expressed through unique land cover and land use changes (LCLUCs). Although LCLUCs have been widely studied in remote sensing, their value for recovery assessment has not yet been explored, which is the focus of this paper. An RS-based methodology was created for PDR assessment based on multi-temporal, very high-resolution satellite images. Different trajectories of change were analyzed and evaluated, i.e., transition patterns (TPs) that signal positive or negative recovery. Experimental analysis was carried out on three WorldView-2 images acquired over Tacloban city, Philippines, which was heavily affected by Typhoon Haiyan in 2013. Support vector machine, a robust machine learning algorithm, was employed with texture features extracted from the grey level co-occurrence matrix and local binary patterns. Although classification results for the images before and four years after the typhoon show high accuracy, substantial uncertainties mark the results for the immediate post-event image. All land cover (LC) and land use (LU) classified maps were stacked, and only changes related to TPs were extracted. The final products are LC and LU recovery maps that quantify the PDR process at the pixel level. It was found that physical and functional recovery can be mainly explained through the LCLUC information. In addition, LC and LU-based recovery maps support a general and a detailed recovery understanding, respectively. It is therefore suggested to use the LC and LU-based recovery maps to monitor and support the short and the long-term recovery, respectively. [ABSTRACT FROM AUTHOR]

Rapid increase in population and growing concentration of capital in urban areas has escalated both the severity and longer-term impact of natural disasters. As a result, Disaster Risk Management (DRM) and reduction have been gaining increasing importance for urban areas. Remote sensing plays a key role in providing information for urban DRM analysis due to its agile data acquisition, synoptic perspective, growing range of data types, and instrument sophistication, as well as low cost. As a consequence numerous methods have been developed to extract information for various phases of DRM analysis. However, given the diverse information needs, only few of the parameters of interest are extracted directly, while the majority have to be elicited indirectly using proxies. This paper provides a comprehensive review of the proxies developed for two risk elements typically associated with pre-disaster situations (vulnerability and resilience), and two post-disaster elements (damage and recovery), while focusing on urban DRM. The proxies were reviewed in the context of four main environments and their corresponding sub-categories: built-up (buildings, transport, and others), economic (macro, regional and urban economics, and logistics), social (services and infrastructures, and socio-economic status), and natural. All environments and the corresponding proxies are discussed and analyzed in terms of their reliability and sufficiency in comprehensively addressing the selected DRM assessments. We highlight strength and identify gaps and limitations in current proxies, including inconsistencies in terminology for indirect measurements. We present a systematic overview for each group of the reviewed proxies that could simplify cross-fertilization across different DRM domains and may assist the further development of methods. While systemizing examples from the wider remote sensing domain and insights from social and economic sciences, we suggest a direction for developing new proxies, also potentially suitable for capturing functional recovery. [ABSTRACT FROM AUTHOR]

A letter to the editor is presented regarding the initiatives taken by Philippine government to reduce the damages caused by the Typhoon Hagupit tropical cyclone and states that cyclone forecasting was done and stormsurge inundation maps were prepared by Philippines government.

A letter to the editor is presented in response to an article about the collaboration of citizen scientists with the professional community to map the structural damage caused by disasters, which appeared in the 2014 issue.

Oblique aerial images offer views of both building roofs and façades, and thus have been recognized as a potential source to detect severe building damages caused by destructive disaster events such as earthquakes. Therefore, they represent an important source of information for first responders or other stakeholders involved in the post-disaster response process. Several automated methods based on supervised learning have already been demonstrated for damage detection using oblique airborne images. However, they often do not generalize well when data from new unseen sites need to be processed, hampering their practical use. Reasons for this limitation include image and scene characteristics, though the most prominent one relates to the image features being used for training the classifier. Recently features based on deep learning approaches, such as convolutional neural networks (CNNs), have been shown to be more effective than conventional hand-crafted features, and have become the state-of-the-art in many domains, including remote sensing. Moreover, often oblique images are captured with high block overlap, facilitating the generation of dense 3D point clouds – an ideal source to derive geometric characteristics. We hypothesized that the use of CNN features, either independently or in combination with 3D point cloud features, would yield improved performance in damage detection. To this end we used CNN and 3D features, both independently and in combination, using images from manned and unmanned aerial platforms over several geographic locations that vary significantly in terms of image and scene characteristics. A multiple-kernel-learning framework, an effective way for integrating features from different modalities, was used for combining the two sets of features for classification. The results are encouraging: while CNN features produced an average classification accuracy of about 91%, the integration of 3D point cloud features led to an additional improvement of about 3% (i.e. an average classification accuracy of 94%). The significance of 3D point cloud features becomes more evident in the model transferability scenario (i.e., training and testing samples from different sites that vary slightly in the aforementioned characteristics), where the integration of CNN and 3D point cloud features significantly improved the model transferability accuracy up to a maximum of 7% compared with the accuracy achieved by CNN features alone. Overall, an average accuracy of 85% was achieved for the model transferability scenario across all experiments. Our main conclusion is that such an approach qualifies for practical use. [ABSTRACT FROM AUTHOR]

Consumer-grade Unmanned Aerial Vehicles (UAVs) are poorly suited to monitor complex scenes where multiple analysis tasks need to be carried out in real-time and in parallel to fulfil time-critical requirements. Therefore, we developed an innovative UAV agnostic system that is able to carry out multiple road infrastructure monitoring tasks simultaneously and in real-time. The aim of the paper is to discuss the system design considerations and the performance of the processing pipeline in terms of computational strain and latency. The system was deployed on a unique typology of UAV and instantiated with realistic placeholder modules that are of importance for infrastructure inspection tasks, such as vehicle detection for traffic monitoring, scene segmentation for qualitative semantic reasoning, and 3D scene reconstruction for large-scale damage detection. The system was validated by carrying out a trial on a highway in Guadalajara, Spain. By utilizing edge computation and remote processing, the end-to-end pipeline, from image capture to information dissemination to drone operators on the ground, takes on average 2.9 s, which is sufficiently quick for road monitoring purposes. The system is dynamic and, therefore, can be extended with additional modules, while continuously accommodating developments in technologies, such as IoT or 5G. [ABSTRACT FROM AUTHOR]

Human detection in images using deep learning has been a popular research topic in recent years and has achieved remarkable performance. Training a human detection network is useful for first responders to search for trapped victims in debris after a disaster. In this paper, we focus on the detection of such victims using deep learning, and we find that state-of-the-art detection models pre-trained on the well-known COCO dataset fail to detect victims. This is because all the people in the training set are shown in photos of daily life or sports activities, while people in the debris after a disaster usually only have parts of their bodies exposed. In addition, because of the dust, the colors of their clothes or body parts are similar to those of the surrounding debris. Compared with collecting images of common objects, images of disaster victims are extremely difficult to obtain for training. Therefore, we propose a framework to generate harmonious composite images for training. We first paste body parts onto a debris background to generate composite victim images and then use a deep harmonization network to make the composite images look more harmonious. We select YOLOv5l as the most suitable model, and experiments show that using composite images for training improves the AP (average precision) by 19.4% ( 15.3 % → 34.7 % ). Furthermore, using the harmonious images is of great benefit to training a better victim detector, and the AP is further improved by 10.2% ( 34.7 % → 44.9 % ). This research is part of the EU project INGENIOUS. Our composite images and code are publicly available on our website. [ABSTRACT FROM AUTHOR]

Automatic post-disaster mapping of building damage using remote sensing images is an important and time-critical element of disaster management. The characteristics of remote sensing images available immediately after the disaster are not certain, since they may vary in terms of capturing platform, sensor-view, image scale, and scene complexity. Therefore, a generalized method for damage detection that is impervious to the mentioned image characteristics is desirable. This study aims to develop a method to perform grid-level damage classification of remote sensing images by detecting the damage corresponding to debris, rubble piles, and heavy spalling within a defined grid, regardless of the aforementioned image characteristics. The Visual-Bag-of-Words (BoW) is one of the most widely used and proven frameworks for image classification in the field of computer vision. The framework adopts a kind of feature representation strategy that has been shown to be more efficient for image classification--regardless of the scale and clutter--than conventional global feature representations. In this study supervised models using various radiometric descriptors (histogram of gradient orientations (HoG) and Gabor wavelets) and classifiers (SVM, Random Forests, and Adaboost) were developed for damage classification based on both BoW and conventional global feature representations, and tested with four datasets. Those vary according to the aforementioned image characteristics. The BoW framework outperformed conventional global feature representation approaches in all scenarios (i.e., for all combinations of feature descriptors, classifiers, and datasets), and produced an average accuracy of approximately 90%. Particularly encouraging was an accuracy improvement by 14% (from 77% to 91%) produced by BoW over global representation for the most complex dataset, which was used to test the generalization capability. [ABSTRACT FROM AUTHOR]

Point clouds generated from airborne oblique images have become a suitable source for detailed building damage assessment after a disaster event, since they provide the essential geometric and radiometric features of both roof and façades of the building. However, they often contain gaps that result either from physical damage or from a range of image artefacts or data acquisition conditions. A clear understanding of those reasons, and accurate classification of gap-type, are critical for 3D geometry-based damage assessment. In this study, a methodology was developed to delineate buildings from a point cloud and classify the present gaps. The building delineation process was carried out by identifying and merging the roof segments of single buildings from the pre-segmented 3D point cloud. This approach detected 96% of the buildings from a point cloud generated using airborne oblique images. The gap detection and classification methods were tested using two other data sets obtained with Unmanned Aerial Vehicle (UAV) images with a ground resolution of around 1–2 cm. The methods detected all significant gaps and correctly identified the gaps due to damage. The gaps due to damage were identified based on the surrounding damage pattern, applying Gabor wavelets and a histogram of gradient orientation features. Two learning algorithms – SVM and Random Forests were tested for mapping the damaged regions based on radiometric descriptors. The learning model based on Gabor features with Random Forests performed best, identifying 95% of the damaged regions. The generalization performance of the supervised model, however, was less successful: quality measures decreased by around 15–30%. [ABSTRACT FROM AUTHOR]

Updated spatial information on the dynamics of slums can be helpful to measure and evaluate progress of policies. Earlier studies have shown that semi-automatic detection of slums using remote sensing can be challenging considering the large variability in definition and appearance. In this study, we explored the potential of an object-oriented image analysis (OOA) method to detect slums, using very high resolution (VHR) imagery. This method integrated expert knowledge in the form of a local slum ontology. A set of image-based parameters was identified that was used for differentiating slums from non-slum areas in an OOA environment. The method was implemented on three subsets of the city of Ahmedabad, India. Results show that textural features such as entropy and contrast derived from a grey level co-occurrence matrix (GLCM) and the size of image segments are stable parameters for classification of built-up areas and the identification of slums. Relation with classified slum objects, in terms of enclosed by slums and relative border with slums was used to refine classification. The analysis on three different subsets showed final accuracies ranging from 47% to 68%. We conclude that our method produces useful results as it allows including location specific adaptation, whereas generically applicable rulesets for slums are still to be developed. [ABSTRACT FROM AUTHOR]

Abstract: The development of surface fissures is an important indicator for understanding and forecasting slope movements. Landslide investigations therefore frequently include the elaboration and interpretation of maps representing their spatial distribution, typically comprising intensive field work and instrumentation. It is only recently that aerial photography with sub-decimetre spatial resolution is becoming more commonly available and opens a window to analyse such features from a remote sensing perspective. While these data are in principle helpful to elaborate maps from image interpretation techniques, there is still no image processing technique available to extract efficiently these geomorphological features. This work proposes a largely automated technique for the mapping of landslide surface fissures from very-high resolution aerial images. The processing chain includes the use of filtering algorithms and post-processing of the filtered images using object-oriented analysis. The accuracy of the resulting maps is assessed by comparisons with several expert maps in terms of affected area, fissure density and fissure orientation. Under homogenous illumination conditions, true positive rates up to 65% and false positive rates generally below 10% are achieved. The resulting fissure maps provide sufficient detail to infer mechanical processes at the slope scale and to prioritize areas for more detailed ground investigations or monitoring. [Copyright &y& Elsevier]

Abstract: Landslide inventories prepared manually from remote sensing data or through field surveys have shown to be useful for preparation of landslide susceptibility and hazard maps. Recent literatures show several studies have been carried out to prepare landslide inventories from satellite data by automatic methods. However, almost no attempt has been made to validate the effect of such inventories on landslide hazard and risk assessment. In this paper we have shown how landslide inventories prepared by semi-automatic methods from post-event satellite images can be used in the assessment of landslide susceptibility, hazard and risk in the High Himalayan terrain in India. A susceptibility map was made using the weights-of-evidence method, wherein weights were derived using the semi-automatically prepared historical landslide inventories combined with a series of pre-disposing factor maps. Seven evidence layers were used for the calculation of weights, selected in such a way that the majority could be derived from satellite data. Validation was done using the test data created through a temporal subsetting of the inventories. Temporal probability was calculated through Gumbel frequency distribution analysis using daily rainfall data of a 13year period for which landslide inventories were prepared from satellite data. Spatial probability was determined by calculating landslide density for the inventories per susceptibility class that represent a given return period. Elements at risk, such as buildings and roads, were interpreted from a high resolution Cartosat-1 (2.5m) image. Absolute vulnerability of the buildings and roads were multiplied with landslide spatial probability to derive the total loss for different return-period scenarios and shown in a risk curve. This study has shown that the inventories prepared semi-automatically can be used for landslide hazard and risk assessment in a data-poor environment. [Copyright &y& Elsevier]

Abstract: Landslide susceptibility mapping (LSM) along road corridors in the Indian Himalayas is an essential exercise that helps planners and decision makers in determining the severity of probable slope failure areas. Logistic regression is commonly applied for this purpose, as it is a robust and straightforward technique that is relatively easy to handle. Ordinary logistic regression as a data-driven technique, however, does not allow inclusion of prior information. This study presents Bayesian logistic regression (BLR) for landslide susceptibility assessment along road corridors. The methodology is tested in a landslide-prone area in the Bhagirathi river valley in the Indian Himalayas. Parameter estimates from BLR are compared with those obtained from ordinary logistic regression. By means of iterative Markov Chain Monte Carlo simulation, BLR provides a rich set of results on parameter estimation. We assessed model performance by the receiver operator characteristics curve analysis, and validated the model using 50% of the landslide cells kept apart for testing and validation. The study concludes that BLR performs better in posterior parameter estimation in general and the uncertainty estimation in particular. [Copyright &y& Elsevier]

Abstract: Information about rapidly changing slum areas may support the development of appropriate interventions by concerned authorities. Often, however, traditional data collection methods lack information on the spatial distribution of slum-dwellers. Remote sensing based methods could be used for a rapid inventory of the location and physical composition of slums. (Semi-)automatic detection of slums in image data is challenging, owing to the high variability in appearance and definitions across different contexts. This paper develops an ontological framework to conceptualize slums using input from 50 domain-experts covering 16 different countries. This generic slum ontology (GSO) comprises concepts identified at three levels that refer to the morphology of the built environment: the environs level, the settlement level and the object level. It serves as a comprehensive basis for image-based classification of slums, in particular, using object-oriented image analysis (OOA) techniques. This is demonstrated by with an example of local adaptation of GSO and OOA parameterization for a study area in Kisumu, Kenya. At the object level, building and road characteristics are major components of the ontology. At the settlement level, texture measures can be potentially used to represent the contrast between planned and unplanned settlements. At the environs level, factors which extend beyond the site itself are important indicators, e.g. hazards due to floods plains and marshy conditions. The GSO provides a comprehensive framework that includes all potentially relevant indicators that can be used for image-based slum identification. These characteristics may be different for other study areas, but show the applicability of the developed framework. [Copyright &y& Elsevier]

The increased socio-economic significance of landslides has resulted in the application of statistical methods to assess their hazard, particularly at medium scales. These models evaluate where, when and what size landslides are expected. The method presented in this study evaluates the landslide hazard on the basis of homogenous susceptible units (HSU). HSU are derived from a landslide susceptibility map that is a combination of landslide occurrences and geo-environmental factors, using an automated segmentation procedure. To divide the landslide susceptibility map into HSU, we apply a region-growing segmentation algorithm that results in segments with statistically independent spatial probability values. Independence is tested using Moran's I and a weighted variance method. For each HSU, we obtain the landslide frequency from the multi-temporal data. Temporal and size probabilities are calculated using a Poisson model and an inverse-gamma model, respectively. The methodology is tested in a landslide-prone national highway corridor in the northern Himalayas, India. Our study demonstrates that HSU can replace the commonly used terrain mapping units for combining three probabilities for landslide hazard assessment. A quantitative estimate of landslide hazard is obtained as a joint probability of landslide size, of landslide temporal occurrence for each HSU for different time periods and for different sizes. [ABSTRACT FROM AUTHOR]

Image segmentation is a preliminary and critical step in object-based image classification. Its proper evaluation ensures that the best segmentation is used in image classification. In this article, image segmentations with nine different parameter settings were carried out with a multi-spectral Landsat imagery and the segmentation results were evaluated with an objective function that aims at maximizing homogeneity within segments and separability between neighbouring segments. The segmented images were classified into eight land-cover classes and the classifications were evaluated with independent ground data comprising 600 randomly distributed points. The accuracy assessment results presented similar distribution as that of the objective function values, that is segmentations with the highest objective function values also resulted in the highest classification accuracies. This result shows that image segmentation has a direct effect on the classification accuracy; the objective function not only worked on a single band image as proved by (Espindola, G.M., Camara, G., Reis, I.A., Bins, L.S. and Monteiro, A.M., 2006, Parameter selection for region-growing image segmentation algorithms using spatial autocorrelation. International Journal of Remote Sensing, 27, pp. 3035-3040.) but also on multi-spectral imagery as tested in this, and is indeed an effective way to determine the optimal segmentation parameters. McNemar's test (z2 = 10.27) shows that with the optimal segmentation, object-based classification achieved accuracy significantly higher than that of the pixel-based classification, with 99% significance level. [ABSTRACT FROM AUTHOR]

Abstract: The impact of topographic attributes on the uneven distribution of seismic response and associated devastation has frequently been observed and documented during seismic events, but has rarely been investigated at a regional scale. Existing numerical and experimental techniques applied to explore the impact of topographic attributes in the aggravation of seismic response, have been limited to isolated and/or synthetic hills and ridges. Predicting the realistic regional impact of topographic seismic response is strongly dependent on the resolution and accuracy of regional topographic information. This study evaluates the topographic attributes and seismic parameters computed from multi-resolution and source DEMs, to investigate the impact of data source and resolution on the derived topographic seismic response. Methodologies are developed to readily derive the spatial distribution of relevant topographic attributes and seismic parameters, utilizing the multi-resolution and source DEMs. The impact of DEM source and resolution on slope gradient, relative height of terrain and shear wave velocity () are addressed. It is observed that, even though, relatively coarse resolution DEMs underestimate the critical sites of steep slope gradient and the lower zones, this has limited impact on the derived normalized topographic aggravation factor. The free and easily accessible DEMs provide an opportunity for reasonable prediction of topographic seismic response, especially in near-real time. The slope gradient is observed to be the most sensitive topographic attribute to amplified seismic response, followed by the relative height. [Copyright &y& Elsevier]

Limited awareness of environmental remote sensing's potential ability to support environmental policy development constrains the technology's utilization. This paper reviews the potential of earth observation from the perspective of environmental policy. A literature review of "remote sensing and policy" revealed that while the number of publications in this field increased almost twice as rapidly as that of remote sensing literature as a whole (15.3 versus 8.8% yr-1), there is apparently little academic interest in the societal contribution of environmental remote sensing. This is because none of the more than 300 peer reviewed papers described actual policy support. This paper describes and discusses the potential, actual support, and limitations of earth observation with respect to supporting the various stages of environmental policy development. Examples are given of the use of remote sensing in problem identification and policy formulation, policy implementation, and policy control and evaluation. While initially, remote sensing contributed primarily to the identification of environmental problems and policy implementation, more recently, interest expanded to applications in policy control and evaluation. The paper concludes that the potential of earth observation to control and evaluate, and thus assess the efficiency and effectiveness of policy, offers the possibility of strengthening governance. [ABSTRACT FROM AUTHOR]

Abstract: A flank collapse occurred at the Casita volcano in Nicaragua on 30 October 1998 during Hurricane Mitch. The collapse transformed into a disastrous lahar that completely buried two small towns 6 km downstream (killing about 2500 people), destroyed small settlements and disrupted the Pan American Highway. Based on the knowledge acquired during previous investigations with additional unpublished field data and observations, this study provides a comprehensive review of all previous studies (published and unpublished) on the 1998 Casita lahar and new insights into the initial flank collapse and the current stability. The knowledge on pre- and post-failure geometry, geology, lithology, tectonics and stratigraphy of the scarp area is improved and a summary of available geotechnical data is provided. Lithological characteristics that were significant in the initial flank collapse and failure mechanism were identified. The location of the failure surface was more precisely defined and the number and sequence of stages in the initial failure presented in previous studies have been confirmed. Slope stability analyses were carried out using limit-equilibrium methods. Geological interpretations, analysis of digital elevation models and geotechnical back-analyses confirmed that the flank collapse took place in three stages involving both the northern and southern areas of the scarp and occurred continuously during a time interval of seconds to a few minutes. In the first stage, failure initiated in a highly fractured and altered volcanic breccia in the northern area of the scarp which released a volume of 260000 m3. The flow that developed from this failure removed colluvium deposits at the toe of the slope in the southern part in not more than 40 s. This rapid removal of the colluvium triggered a second stage which comprised 640000 m3 and consisted in the failure of the southern part of the scarp by the sliding of a fractured volcanic breccia over a unit of clay-rich pyroclastic deposits. The third and final stage consisted in a failure of the remaining breccia and the overlying fractured lavas in the northern area and involved a volume of 690000 m3. The fact that future events can affect some remaining settlements and the segment of the Pan American Highway, between Chinandega and León, motivated an analysis of the stability of the remaining slope using parameters calibrated in the back-analyses of the 1998 flank failure. The results indicated that the remaining slope is stable as long as groundwater levels in the northern and southern area are deeper than 65 m and 90 m respectively (relative to the points of maximum elevation on the analysed section) and the colluvium deposits in the southern area are not removed from the toe of the slope. [Copyright &y& Elsevier]

The sensitivity of synthetic aperture radar (SAR) and interferometric SAR (InSAR) to surface properties, especially changes in height and roughness, combined with an all-weather capability, makes radar remote sensing a potential tool for mapping urban damage caused by earthquakes. With InSAR, surface displacement has been mapped successfully and in detail, but for urban-damage mapping, results have so far been less conclusive. ENVISAT Advanced SAR images of Barn, Iran, that were acquired before and after the 2003 earthquake were used. Between preseismic and coseismic image pairs, coherence decreased with increasing damage levels. However, contrary to previous studies, earthquake damage caused both increases and decreases in amplitude. Therefore, its absolute value, which correlated with damage level, was used. Individually, coherence difference led to better results than absolute amplitude change, although still with limited accuracy. The combination of both resulted in an overall accuracy of 52%. Since vegetation causes decorrelation, a predisaster Advanced Spaceborne Thermal Emission and Reflection Radiometer image was used to mask out vegetated areas, which improved the accuracy by 4%. The ground-truth map showed damage levels per district instead of per pixel. Therefore, the ratio of pixels classified as "damage" and "nondamage" was calculated for each damage class of the map, and a clear relation was found. This shows that the aggregation level of the map partly explains the low-accuracy figures of the pixel-based evaluation. Although improvement was made, InSAR techniques for urban-damage mapping do not yet provide the accuracy levels needed for disaster mitigation. However, substantial improvements can be expected from instruments with higher spatial resolution, such as the recently launched Phased Array type L-band SAR and Terra SAR-X. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

• Different landslide recording methods including pixel-based, centroid, crown, and toe. • Landslide spatial modeling using Mahalanobis distance and random sampling methods. • Negative balance scenariosproposed for landslide susceptibility assessment. The Ziarat Watershed, located in the south of the Golestan Province, Iran, has witnessed several destructive landslide episodes, prompting a number of researchers to aspire to improve landslide susceptibility modeling (LSM) techniques. We constructed three scenarios focusing on landslide positioning techniques (pixel-based, centroid, crown, and toe), training/test sampling strategies (Mahalanobis distance (MD), and random sampling (RS)), with alternative landslide/non-landslide data balances (1:1, 1:2, and 1:3). The data mining boosted regression trees (BRT) model was used for the landslide susceptibility modeling, using landslide data and 13 landslide controlling factors for the Ziarat Watershed. The performance of the scenarios was assessed using the areas under the success and prediction rate curves (AUSRC and AUPRC). A combination of pixel-based–MD–1:2 showed the highest learning capability and goodness-of-fit with an AUSRC value of 0.87, and the highest predictive power and generalization capacity with an AUPRC value of 0.79. Conversely, centroid-based–1:3–RS, crown-based–1:3–RS, and toe-based–1:3–RS performed less well. Comparatively, the pixel-based, MD, and 1:2 data balance scenarios surpassed their counterparts and outperformed the other models. The results indicated a high spatial differentiation with a significant chi-square value of 4549.46 at 95% confidence level. Moreover, 15.21% of the study area, containing almost 50% of the landslides, was found to have a high susceptibility to landslides. According to the premier scenario (pixel-based–MD–1:2), lithological formation, distance from roads, and NDVI, with respective contributions of 31.4%, 12.9%, and 12%, are the main spatial controlling factors leading to landslide occurrences in the study area. [ABSTRACT FROM AUTHOR]

It remains elusive why there was only weak and limited ground shaking in Kathmandu valley during the 25 April 2015 Mw 7.8 Gorkha, Nepal, earthquake. Our spectral element numerical simulations show that, during this earthquake, surface topography restricted the propagation of seismic energy into the valley. The mountains diverted the incoming seismic wave mostly to the eastern and western margins of the valley. As a result, we find de-amplification of peak ground displacement in most of the valley interior. Modeling of alternative earthquake scenarios of the same magnitude occurring at different locations shows that these will affect the Kathmandu valley much more strongly, up to 2–3 times more, than the 2015 Gorkha earthquake did. This indicates that surface topography contributed to the reduced seismic shaking for this specific earthquake and lessened the earthquake impact within the valley. [ABSTRACT FROM AUTHOR]

Citizens are increasingly becoming an important source of geographic information, sometimes entering domains that had until recently been the exclusive realm of authoritative agencies. This activity has a very diverse character as it can, amongst other things, be active or passive, involve spatial or aspatial data and the data provided can be variable in terms of key attributes such as format, description and quality. Unsurprisingly, therefore, there are a variety of terms used to describe data arising from citizens. In this article, the expressions used to describe citizen sensing of geographic information are reviewed and their use over time explored, prior to categorizing them and highlighting key issues in the current state of the subject. The latter involved a review of ~100 Internet sites with particular focus on their thematic topic, the nature of the data and issues such as incentives for contributors. This review suggests that most sites involve active rather than passive contribution, with citizens typically motivated by the desire to aid a worthy cause, often receiving little training. As such, this article provides a snapshot of the role of citizens in crowdsourcing geographic information and a guide to the current status of this rapidly emerging and evolving subject. [ABSTRACT FROM AUTHOR]

Active learning (AL) algorithms have been proven useful in reducing the number of required training samples for remote sensing applications; however, most methods query samples pointwise without considering spatial constraints on their distribution. This may often lead to a spatially dispersed distribution of training points unfavorable for visual image interpretation or field surveys. The aim of this study is to develop region-based AL heuristics to guide user attention toward a limited number of compact spatial batches rather than distributed points. The proposed query functions are based on a tree ensemble classifier and combine criteria of sample uncertainty and diversity to select regions of interest. Class imbalance, which is inherent to many remote sensing applications, is addressed through stratified bootstrap sampling. Empirical tests of the proposed methods are performed with multitemporal and multisensor satellite images capturing, in particular, sites recently affected by large-scale landslide events. The assessment includes an experimental evaluation of the labeling time required by the user and the computational runtime, and a sensitivity analysis of the main algorithm parameters. Region-based heuristics that consider sample uncertainty and diversity are found to outperform pointwise sampling and region-based methods that consider only uncertainty. Reference landslide inventories from five different experts enable a detailed assessment of the spatial distribution of remaining errors and the uncertainty of the reference data. [ABSTRACT FROM AUTHOR]

Developing reliable methodologies of data quality assessment is of paramount importance for maximizing the exploitation of Earth observation (EO) products. Among the different factors influencing EO optical image quality, sharpness has a relevant role. When implementing on-orbit approaches of sharpness assessment, such as the edge method, a crucial step that strongly affects the final results is the selection of suitable edges to use for the analysis. Within this context, this paper aims at proposing a semi-automatic, statistically-based edge method (SaSbEM) that exploits edges extracted from natural targets easily and largely available on Earth: agricultural fields. For each image that is analyzed, SaSbEM detects numerous suitable edges (e.g., dozens-hundreds) characterized by specific geometrical and statistical criteria. This guarantees the repeatability and reliability of the analysis. Then, it implements a standard edge method to assess the sharpness level of each edge. Finally, it performs a statistical analysis of the results to have a robust characterization of the image sharpness level and its uncertainty. The method was validated by using Landsat 8 L1T products. Results proved that: SaSbEM is capable of performing a reliable and repeatable sharpness assessment; Landsat 8 L1T data are characterized by very good sharpness performance. [ABSTRACT FROM AUTHOR]

Remotely sensed data can provide the basis for timely and efficient building damage maps that are of fundamental importance to support the response activities following disaster events. However, the generation of these maps continues to be mainly based on the manual extraction of relevant information in operational frameworks. Considering the identification of visible structural damages caused by earthquakes and explosions, several recent works have shown that Convolutional Neural Networks (CNN) outperform traditional methods. However, the limited availability of publicly available image datasets depicting structural disaster damages, and the wide variety of sensors and spatial resolution used for these acquisitions (from space, aerial and UAV platforms), have limited the clarity of how these networks can effectively serve First Responder needs and emergency mapping service requirements. In this paper, an advanced CNN for visible structural damage detection is tested to shed some light on what deep learning networks can currently deliver, and its adoption in realistic operational conditions after earthquakes and explosions is critically discussed. The heterogeneous and large datasets collected by the authors covering different locations, spatial resolutions and platforms were used to assess the network performances in terms of transfer learning with specific regard to geographical transferability of the trained network to imagery acquired in different locations. The computational time needed to deliver these maps is also assessed. Results show that quality metrics are influenced by the composition of training samples used in the network. To promote their wider use, three pre-trained networks—optimized for satellite, airborne and UAV image spatial resolutions and viewing angles—are made freely available to the scientific community. [ABSTRACT FROM AUTHOR]

The timely and efficient generation of detailed damage maps is of fundamental importance following disaster events to speed up first responders' (FR) rescue activities and help trapped victims. Several works dealing with the automated detection of building damages have been published in the last decade. The increasingly widespread availability of inexpensive UAV platforms has also driven their recent adoption for rescue operations (i.e., search and rescue). Their deployment, however, remains largely limited to visual image inspection by skilled operators, limiting their applicability in time-constrained real conditions. This paper proposes a new solution to autonomously map building damages with a commercial UAV in near real-time. The solution integrates different components that allow the live streaming of the images on a laptop and their processing on the fly. Advanced photogrammetric techniques and deep learning algorithms are combined to deliver a true-orthophoto showing the position of building damages, which are already processed by the time the UAV returns to base. These algorithms have been customized to deliver fast results, fulfilling the near real-time requirements. The complete solution has been tested in different conditions, and received positive feedback by the FR involved in the EU funded project INACHUS. Two realistic pilot tests are described in the paper. The achieved results show the great potential of the presented approach, how close the proposed solution is to FR' expectations, and where more work is still needed. [ABSTRACT FROM AUTHOR]