We propose a novel approach for rapid segmentation of flooded buildings by fusing multiresolution, multisensor, and multitemporal satellite imagery in a convolutional neural network. Our model significantly expedites the generation of satellite imagery-based flood maps, crucial for first responders and local authorities in the early stages of flood events. By incorporating multitemporal satellite imagery, our model allows for rapid and accurate post-disaster damage assessment and can be used by governments to better coordinate medium- and long-term financial assistance programs for affected areas. The network consists of multiple streams of encoder-decoder architectures that extract spatiotemporal information from medium-resolution images and spatial information from high-resolution images before fusing the resulting representations into a single medium-resolution segmentation map of flooded buildings. We compare our model to state-of-the-art methods for building footprint segmentation as well as to alternative fusion approaches for the segmentation of flooded buildings and find that our model performs best on both tasks. We also demonstrate that our model produces highly accurate segmentation maps of flooded buildings using only publicly available medium-resolution data instead of significantly more detailed but sparsely available very high-resolution data. We release the first open-source dataset of fully preprocessed and labeled multiresolution, multispectral, and multitemporal satellite images of disaster sites along with our source code.

Informal settlements are home to the most socially and economically vulnerable people on the planet. In order to deliver effective economic and social aid, non-government organizations (NGOs), such as the United Nations Children's Fund (UNICEF), require detailed maps of the locations of informal settlements. However, data regarding informal and formal settlements is primarily unavailable and if available is often incomplete. This is due, in part, to the cost and complexity of gathering data on a large scale. To address these challenges, we, in this work, provide three contributions. 1) A brand new machine learning dataset purposely developed for informal settlement detection. 2) We show that it is possible to detect informal settlements using freely available low-resolution (LR) data, in contrast to previous studies that use very-high resolution~(VHR) satellite and aerial imagery, something that is cost-prohibitive for NGOs. 3) We demonstrate two effective classification schemes on our curated data set, one that is cost-efficient for NGOs and another that is cost-prohibitive for NGOs, but has additional utility. We integrate these schemes into a semi-automated pipeline that converts either a LR or VHR satellite image into a binary map that encodes the locations of informal settlements.

Inadequate characterization of AMD-generating mine waste is a major obstacle to remediation of post-mining sites, on the other hand a complete inventory and assessment of their environmental impacts are far from complete. This lack of update records calls for new fast and time and cost efficient monitoring techniques. Imaging spectroscopy represents an alternative to conventional methods (chemical analyses-based assessment tools) and efficient way to characterize mines and assess the AMD discharge. In this study, a new index for Sentinel-2 data (S-2) is introduced (DS-2(AMD)) allowing mapping AMD and sufficient differentiation between jarosite minerals and other oxy-hydroxides. It is also demonstrated that S-2 data has a potential to map seasonal variations of AMD.

Independent spectral analysis is usually employed to analyse hyperspectral optical (visible: VIS, near infrared: NIR, shortwave infrared: SWIR) and thermal (longwave infrared: LWIR) data. The integration of the spectral information provided by different wavelength ranges and the subsequent complex classification still remains challenging. In this paper we will demonstrate the benefits of mineral classification employed to optical and thermal hyperspectral data (CASI and SASI: 0.4-2.5 μ m; TASI: 8.6-11.5 μm) when using new tools (QUANTools) developed at the Czech Geological Survey (CGS). In this case study, the same approach was employed as published by Kopačková and Koucká (2017) [1], however, previously hyperspectral optical data were used together with multispectral long-wave infrared (LWIR) data. In this paper we demonstrate that the new concept, which was recently introduced, can be directly employed to CASI/SASI/TASI hyperspectral datasets and provide results in terms of mineral classification in a quick and easy way.

The growing demand for mineral and energy resources over the last decade has placed the extractive industry under increasing pressure to monitor and reduce the environmental and societal impact throughout the life-cycle of mining operations. Despite the mounting pressure, the industry is still facing the challenge of how to define targets for, and monitor, the impact of mining. In 2010, the EU-funded EO-MINERS project (Earth Observation for Monitoring and Observing Environmental and Societal Impacts of Mineral Resources Exploration and Exploitation) was set up in an effort to help address this issue, specifically through the application of Earth Observation (EO) data. Furthermore, the aim was to help facilitate and improve interaction and dialogue between the mineral extractive industry and society in view of its sustainable development, while improving its societal acceptability. One of the primary project objectives was to develop novel yet objective EO products contributing to a constructive “trialogue” involving stakeholders such as industrialists (mining companies), regulatory bodies and the civil society. EOMINERS is scheduled to run until October 2013.

The pH is one of the major chemical parameter affecting results of remediation programs carried on abandoned mines and dumps and one of the major parameters controlling heavy metal mobilization and speciation. Two approaches that contribute to pH mapping were compared. Firstly, high-altitude spectroradiometry (ASTER - Advanced Spaceborne Thermal Emission and Reflection Radiometer satellite data) together with ground-based spectroradiometry (ASD Filedspec spectroradiometer) were employed in order to identify the locations of the most significant sources of AMD discharge at the Sokolov lignite open-pit mines, Czech Republic. As a result, a map with delineated low-pH zones was created. Secondly, the feasibility of mapping surface pH on the basis of airborne multi-flight-line HyMap hyperspectral (HS) datasets was tested. The geochemical and mineral property was linked with emissivity spectra acquired in the field and laboratory. The spectral property of both AMD indicating minerals and the organic component occurring as part of mixtures proved to be identified using spectroscopy and spectral feature fitting (SFF) technique was employed to map the defined endmembers indicating a certain pH ranges in the HS image datasets.

Contents of biochemical compounds such as chlorophyll, nitrogen, cellulose or lignin in foliage can be used as indicators of the actual tree physiological status and the tree previsible damage. Imaging spectroscopy has been often applied for estimation of relations between foliage spectral and biochemical properties of chlorophyll or nitrogen contents ([4], [10], [15]). But it can also be used for the estimation of other important biochemical compounds in foliage such as lignin and cellulose ([6]; [12]), tannin [14] or polyphenolic compounds [13].