Solar photovoltaic module detection using laboratory and airborne imaging spectroscopy data.

Over the past decades, solar panels have been widely used to harvest solar energy owing to the decreased cost of silicon-based photovoltaic (PV) modules, and therefore it is essential to remotely map and monitor the presence of solar PV modules. Many studies have explored on PV module detection based on color aerial photography and manual photo interpretation. Imaging spectroscopy data are capable of providing detailed spectral information to identify the spectral features of PV, and thus potentially become a promising resource for automated and operational PV detection. However, PV detection with imaging spectroscopy data must cope with the vast spectral diversity of surface materials, which is commonly divided into spectral intra-class variability and inter-class similarity. We have developed an approach to detect PV modules based on their physical absorption and reflection characteristics using airborne imaging spectroscopy data. A large database was implemented for training and validating the approach, including spectra-goniometric measurements of PV modules and other materials, a HyMap image spectral library containing 31 materials with 5627 spectra, and HySpex imaging spectroscopy data sets covering Oldenburg, Germany. By normalizing the widely used Hydrocarbon Index (HI), we solved the intra-class variability caused by different detection angles, and validated it against the spectra-goniometric measurements. Knowing that PV modules are composed of materials with different transparencies, we used a group of spectral indices and investigated their interdependencies for PV detection with implementing the image spectral library. Finally, six well-trained spectral indices were applied to HySpex data acquired in Oldenburg, Germany, yielding an overall PV map. Four subsets were selected for validation and achieved overall accuracies, producer's accuracies and user's accuracies, respectively. This physics-based approach was validated against a large database collected from multiple platforms (laboratory measurements, airborne imaging spectroscopy data), thus providing a robust, transferable and applicable way to detect PV modules using imaging spectroscopy data. We aim to create greater awareness of the potential importance and applicability of airborne and spaceborne imaging spectroscopy data for PV modules identification.

Data Products, Quality and Validation of the DLR Earth Sensing Imaging Spectrometer (DESIS).

Imaging spectrometry from aerial or spaceborne platforms, also known as hyperspectral remote sensing, provides dense sampled and fine structured spectral information for each image pixel, allowing the user to identify and characterize Earth surface materials such as minerals in rocks and soils, vegetation types and stress indicators, and water constituents. The recently launched DLR Earth Sensing Imaging Spectrometer (DESIS) installed on the International Space Station (ISS) closes the long-term gap of sparsely available spaceborne imaging spectrometry data and will be part of the upcoming fleet of such new instruments in orbit. DESIS measures in the spectral range from 400 and 1000 nm with a spectral sampling distance of 2.55 nm and a Full Width Half Maximum (FWHM) of about 3.5 nm. The ground sample distance is 30 m with 1024 pixels across track. In this article, a detailed review is given on the applicability of DESIS data based on the specifics of the instrument, the characteristics of the ISS orbit, and the methods applied to generate products. The various DESIS data products available for users are described with the focus on specific processing steps. The results of the data quality and product validation studies show that top-of-atmosphere radiance, geometrically corrected, and bottom-of-atmosphere reflectance products meet the mission requirements. The limitations of the DESIS data products are also subject to a critical examination.

Detecting Unknown Artificial Urban Surface Materials Based on Spectral Dissimilarity Analysis.

High resolution imaging spectroscopy data have been recognised as a valuable data resource for augmenting detailed material inventories that serve as input for various urban applications. Image-specific urban spectral libraries are successfully used in urban imaging spectroscopy studies. However, the regional- and sensor-specific transferability of such libraries is limited due to the wide range of different surface materials. With the developed methodology, incomplete urban spectral libraries can be utilised by assuming that unknown surface material spectra are dissimilar to the known spectra in a basic spectral library (BSL). The similarity measure SID-SCA (Spectral Information Divergence-Spectral Correlation Angle) is applied to detect image-specific unknown urban surfaces while avoiding spectral mixtures. These detected unknown materials are categorised into distinct and identifiable material classes based on their spectral and spatial metrics. Experimental results demonstrate a successful redetection of material classes that had been previously erased in order to simulate an incomplete BSL. Additionally, completely new materials e.g., solar panels were identified in the data. It is further shown that the level of incompleteness of the BSL and the defined dissimilarity threshold are decisive for the detection of unknown material classes and the degree of spectral intra-class variability. A detailed accuracy assessment of the pre-classification results, aiming to separate natural and artificial materials, demonstrates spectral confusions between spectrally similar materials utilizing SID-SCA. However, most spectral confusions occur between natural or artificial materials which are not affecting the overall aim. The dissimilarity analysis overcomes the limitations of working with incomplete urban spectral libraries and enables the generation of image-specific training databases.

Endmember-Guided Unmixing Network (EGU-Net): A General Deep Learning Framework for Self-Supervised Hyperspectral Unmixing.

Over the past decades, enormous efforts have been made to improve the performance of linear or nonlinear mixing models for hyperspectral unmixing (HU), yet their ability to simultaneously generalize various spectral variabilities (SVs) and extract physically meaningful endmembers still remains limited due to the poor ability in data fitting and reconstruction and the sensitivity to various SVs. Inspired by the powerful learning ability of deep learning (DL), we attempt to develop a general DL approach for HU, by fully considering the properties of endmembers extracted from the hyperspectral imagery, called endmember-guided unmixing network (EGU-Net). Beyond the alone autoencoder-like architecture, EGU-Net is a two-stream Siamese deep network, which learns an additional network from the pure or nearly pure endmembers to correct the weights of another unmixing network by sharing network parameters and adding spectrally meaningful constraints (e.g., nonnegativity and sum-to-one) toward a more accurate and interpretable unmixing solution. Furthermore, the resulting general framework is not only limited to pixelwise spectral unmixing but also applicable to spatial information modeling with convolutional operators for spatial-spectral unmixing. Experimental results conducted on three different datasets with the ground truth of abundance maps corresponding to each material demonstrate the effectiveness and superiority of the EGU-Net over state-of-the-art unmixing algorithms. The codes will be available from the website: https://github.com/danfenghong/IEEE_TNNLS_EGU-Net.

Priority list of biodiversity metrics to observe from space.

Monitoring global biodiversity from space through remotely sensing geospatial patterns has high potential to add to our knowledge acquired by field observation. Although a framework of essential biodiversity variables (EBVs) is emerging for monitoring biodiversity, its poor alignment with remote sensing products hinders interpolation between field observations. This study compiles a comprehensive, prioritized list of remote sensing biodiversity products that can further improve the monitoring of geospatial biodiversity patterns, enhancing the EBV framework and its applicability. The ecosystem structure and ecosystem function EBV classes, which capture the biological effects of disturbance as well as habitat structure, are shown by an expert review process to be the most relevant, feasible, accurate and mature for direct monitoring of biodiversity from satellites. Biodiversity products that require satellite remote sensing of a finer resolution that is still under development are given lower priority (for example, for the EBV class species traits). Some EBVs are not directly measurable by remote sensing from space, specifically the EBV class genetic composition. Linking remote sensing products to EBVs will accelerate product generation, improving reporting on the state of biodiversity from local to global scales.