Search

This study introduces a cost-effective and automated methodology to swiftly and accurately gauge disaster-induced land cover changes, harnessing satellite imagery and machine learning. Using open-source platforms like Google Earth Engine, the approach streamlines training sample generation and assesses diverse machine learning algorithms on pre- and post-disaster Landsat images. The innovation lies in its semi-automatic training sample generation methodology, reducing time and manual effort for accurate land cover mapping and addressing the challenge of crafting training samples in disaster settings. The methodology was tested on the 2010 Chilean earthquake and tsunami, revealing a 45% increase in sediment cover and a 15% reduction in water cover, underlining vast landscape alterations. SVM stood out for its consistent accuracy. To validate the methodological robustness and broaden its applicability, we replicated the procedures using diverse land cover collection, including ESA WorldCover 10 m and Dynamic World. This comprehensive validation encompassed varied catastrophic events, extending the methodology's relevance beyond its original scope. The consistent findings across different scenarios underscored the method's robustness and underscored its potential for broader utilization. This streamlined, cost-effective approach suits multiple land cover products, setting the stage for constant Earth feature monitoring. In light of escalating natural disasters, the synergy of machine learning and geospatial technology spotlighted here offers promising avenues for improved disaster preparedness and response. [ABSTRACT FROM AUTHOR]

An introduction is presented in which the editors discuss the theme of the issue, Unmanned Aerial Systems (UAS) for environmental applications special issue preface.

Machine learning offers the potential for effective and efficient classification of remotely sensed imagery. The strengths of machine learning include the capacity to handle data of high dimensionality and to map classes with very complex characteristics. Nevertheless, implementing a machine-learning classification is not straightforward, and the literature provides conflicting advice regarding many key issues. This article therefore provides an overview of machine learning from an applied perspective. We focus on the relatively mature methods of support vector machines, single decision trees (DTs), Random Forests, boosted DTs, artificial neural networks, and k-nearest neighbours (k-NN). Issues considered include the choice of algorithm, training data requirements, user-defined parameter selection and optimization, feature space impacts and reduction, and computational costs. We illustrate these issues through applying machine-learning classification to two publically available remotely sensed data sets. [ABSTRACT FROM AUTHOR]

The WorldView-3 (WV-3) sensor, launched in 2014, is the first highspatial resolution scanner to acquire imagery in the shortwave infrared (SWIR). A spectral ratio of the SWIR combined with the nearinfrared (NIR) can potentially provide an effective differentiation of wildfire burn severity. Previous high spatial resolution sensors were limited to data fromthe visible and NIR for mapping burn severity, for example using the normalized difference vegetation index (NDVI). Drawing on a study site in the Pine Barrens of New Jersey, USA, we investigate optimal processing methods for analysing WV-3 data, with a focus on the pre-fire minus post-fire differenced normalized burn ratio (dNBR). Although the imagery, originally acquired with a 3.7 m instantaneous field of view, was aggregated to 7.5 m pixels by DigitalGlobe due to current licensing constraints, a slight additional smoothing of the data was nevertheless found to help reduce noise in the multi-temporal dNBR imagery. The highest coefficient of determination (R2) of the regressions of dNBR with the field-based composite burn index was obtained with a dNBR ratio produced with the NIR1 and SWIR6 bands. Only a very small increase in R2 was found when dNBR was calculated using the average of NIR1 and NIR2 for the NIR bands, and SWIR5 to SWIR8 for the SWIR bands. dNBR calculated using SWIR1 as the NIR band produced notably lower R2 values than when either NIR1 or NIR2 were used. Differenced NDVI data was found to produce models with a much lower R2 than dNBR, emphasizing the importance of the shortwave infrared region formonitoring fire severity. High spatial resolution dNBR data fromWV-3 can potentially provide valuable information on finer details regarding burn severity patterns than can be obtained from Landsat 30 m data. [ABSTRACT FROM AUTHOR]

The combination of RapidEye satellite imagery and light detection and ranging (lidar) derivatives was assessed for mapping land-cover within a mountaintop coal surface mine complex in the southern coalfields of West Virginia, USA. Support vector machines (SVM), random forests (RF), and boosted classification and regression trees (CART) algorithms were used. Incorporation of the lidar-derived data increased map accuracy in comparison to using only the five imagery bands, and SVM generally produced a more accurate classification than the ensemble tree algorithms based on overall map accuracy, Kappa statistics, allocation disagreement, quantity disagreement, and McNemar's test of statistical significance. Based on measures of predictor variable importance within the ensemble tree classifiers, the normalized digital surface model (nDSM) was found to be more useful than first return intensity data for differentiating the classes. [ABSTRACT FROM AUTHOR]

This letter evaluates the effectiveness of a new kernel-based extreme learning machine (ELM) algorithm for a land cover classification using both multi- and hyperspectral remote-sensing data. The results are compared with the most widely used algorithms – support vector machines (SVMs). The results are compared in terms of the ease of use (in terms of the number of user-defined parameters), classification accuracy and computation cost. A radial basis kernel function was used with both the SVM and the kernel-based extreme-learning machine algorithms to ensure compatibility in the comparison of the two algorithms. The results suggest that the new algorithm is similar to, or more accurate than, SVM in terms of classification accuracy, has notable lower computational cost and does not require the implementation of a multiclass strategy. [ABSTRACT FROM AUTHOR]

This study investigated the use of a geographic object-based image analysis (GEOBIA) approach with the incorporation of object-specific grey-level co-occurrence matrix (GLCM) texture measures from a multispectral Ikonos image for delineation of deciduous, evergreen, and mixed forest types in Guilford Courthouse National Military Park, North Carolina. A series of automated segmentations was produced at a range of scales, each resulting in an associated range of number and size of objects (or segments). Prior to classification, the spatial autocorrelation of each segmentation was evaluated by calculating Moran's I using the average image digital numbers (DNs) per segment. An initial assumption was made that the optimal segmentation scales would have the lowest spatial autocorrelation, and conversely, that over-and under-segmentation would result in higher autocorrelation between segments. At these optimal segmentation scales, the automated segmentation was found to yield information comparable to manually interpreted stand-level forest maps in terms of the size and number of segments. A series of object-based classifications was carried out on the image at the entire range of segmentation scales. The results demonstrated that the scale of segmentation directly influenced the object-based forest type classification results. The accuracies were higher for classification of images identified from a spatial autocorrelation analysis to have an optimal segmentation, compared to those determined to have over- and under-segmentation. An overall accuracy of 79 percent with a Kappa of 0.65 was obtained at the optimal segmentation scale of 19. The addition of object-specific GLCM multiple texture analysis improved classification accuracies up to a value of 83 percent overall accuracy and a Kappa of 0.71 by reducing the confusion between evergreen and mixed forest types. Although some misclassification still remained because of local segmentation quality, a visual assessment of the texture-enhanced GEOBIA classification generally agreeable with manually interpreted forest types. [ABSTRACT FROM AUTHOR]

This study investigated the potential value of integrating hyperspectral visible, near-infrared, and short-wave infrared imagery with multispectral thermal data for geological mapping. Two coregistered aerial data sets of Cuprite, Nevada were used: Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) hyperspectral data, and MODIS/ASTER Airborne Simulator (MASTER) multispectral thermal data. Four classification methods were each applied to AVIRIS, MASTER, and a combined set. Confusion matrices were used to assess the classification accuracy. The assessment showed, in terms of kappa coefficient, that most classification methods applied to the combined data achieved a marked improvement compared to the results using either AVIRIS or MASTER thermal infrared (TIR) data alone. Spectral angle mapper (SAM) showed the best overall classification performance. Minimum distance classification had the second best accuracy, followed by spectral feature fitting (SFF) and maximum likelihood classification. The results of the study showed that SFF applied to the combination of AVIRIS with MASTER TIR data are especially valuable for identification of silicified alteration and quartzite, both of which exhibit distinctive features in the TIR region. SAM showed some advantages over SFF in dealing with multispectral TIR data, obtaining higher accuracy in discriminating low albedo volcanic rocks and limestone which do not have unique, distinguishing features in the TIR region. [Copyright &y& Elsevier]

Spatially accurate and timely information on tree health is an essential component of maintaining sustainability in a species-rich urban setting. Our objective was to evaluate the potential of WorldView-3 (WV-3) satellite images from June 11th, July 30th and August 30th, 2017 to detect field-measured health condition classes of 2308 trees within the District of Columbia Department of Transportation's Street Tree Spatial Database. For each street tree in each image, we measured six vegetation indices (VIs), and find that NDVI1 (defined as the normalized ratio of the red and the first of the near infrared bands) on the July image shows the strongest statistical difference among the VI values of trees classified in the field during 2017 as in good, fair, or poor health condition. This result confirms that high spatial resolution remote sensing images provided the opportunity for a timely tree health assessment at individual tree crown level. Notably, the variability in VI attributable to health condition classes is smaller than the large declines in VIs between the June and August image dates. This greendown phenological pattern occurs similarly for trees in all health condition classes, and is thus an essential consideration when comparing VIs from different years or months for tree health assessment. The tree health condition discrimination was overwhelmed by this greendown phenology and the spectral variability in these species-rich urban settings. Based on these findings, we propose useful strategies of using these high-resolution WV-3 data for street tree health management in other cities. [ABSTRACT FROM AUTHOR]

With the promise of transformative changes for the management of rural and urban forests, the discrimination of tree species from satellite imagery has been a long-standing goal of remote sensing. For the species-rich urban setting of Washington, D.C. USA, we evaluate current prospects toward this goal by combining a Random Forest (RF) object-based tree species classification method with two large datasets 1) A suite of 12 very high resolution (VHR) WorldView-3 images (WV-3), whose image acquisition date cover each pheno-phase of the growing season from April to November; and 2) the 16,496 street trees from Washington D.C. Department of Transportation's (DDOT) field inventory. We classify the 19 most abundant tree species with an overall accuracy of 61.3% and classify the ten most abundant genera with an overall accuracy of 73.7%. We observe that (1) there are larger declines in accuracy when attempting to classify species in the same genus, and (2) the most valuable phenological period is fall senescence for classification at different taxonomic levels. Especially if satellite data can be matched to the key pheno-phases, our study highlights that current VHR satellite sensors now have the radiometric, spectral, and spatial resolution to potentially help manage species-rich urban forests. • High spatial resolution images to classify trees at two different taxonomic levels • Overall accuracy is 61.3% for 19 species, 73.7% for 10 genera. • Most valuable phenological period is fall senescence at two taxonomic levels. [ABSTRACT FROM AUTHOR]