Abstract
Satellite air pollution data sources are presented in this article. Satellite missions are listed and described, and corresponding instruments measuring the concentration of greenhouse gases ...or other parameters that affect air pollution. Main data hubs are enumerated, and other platforms providing access to the data in different formats. Some platforms contain tools and services that have the capabilities to process and store the satellite data further. Most of them also provide the visualisation of the data, and other simplifications for data selection.
This paper considers a new method for the automatic generation of digital terrain models from LiDAR data. The method iterates a thin plate spline interpolated surface towards the ground, while ...points’ residuals from the surface are inspected at each iteration, with a gradually decreasing window size. Top-hat transformation is used to enhance discontinuities caused by surface objects. Finally, parameter-free ground point filtering is achieved by automatic thresholding based on standard deviation. The experiments show that this method correctly determines DTM even in those cases of more difficult terrain features. The expected accuracy of ground point determination on those datasets commonly used in practice today is over 96%, while the average total error produced on the ISPRS benchmark dataset is under 6%.
This paper proposes a new method for 3D delineation of single tree-crowns in LiDAR data by exploiting the complementaries of treetop and tree trunk detections. A unified mathematical framework is ...provided based on the graph theory, allowing for all the segmentations to be achieved using marker-controlled watersheds. Treetops are defined by detecting concave neighbourhoods within the canopy height model using locally fitted surfaces. These serve as markers for watershed segmentation of the canopy layer where possible oversegmentation is reduced by merging the regions based on their heights, areas, and shapes. Additional tree crowns are delineated from mid- and under-storey layers based on tree trunk detection. A new approach for estimating the verticalities of the points’ distributions is proposed for this purpose. The watershed segmentation is then applied on a density function within the voxel space, while boundaries of delineated trees from the canopy layer are used to prevent the overspreading of regions. The experiments show an approximately 6% increase in the efficiency of the proposed treetop definition based on locally fitted surfaces in comparison with the traditionally used local maxima of the smoothed canopy height model. In addition, 4% increase in the efficiency is achieved by the proposed tree trunk detection. Although the tree trunk detection alone is dependent on the data density, supplementing it with the treetop detection the proposed approach is efficient even when dealing with low density point-clouds.
Community detection is a key to understanding the structure of complex networks. Many community detection approaches have been proposed based on the modularity optimization. Algorithms that optimize ...one initial solution often get into local optima, but algorithms that simultaneously optimize a population of solutions have high computational complexity. To solve these problems, genetic algorithms improved by a local learning procedure known as memetic algorithms can be applied. We propose a memetic algorithm for community detection in networks, that exploits node entropy for local learning. Node entropy is easy to use to speed up the convergence of an evolutionary algorithm and to increase the quality of partitions, while it uses only the node’s neighborhood and does not require any threshold value. Moreover, this algorithm is slightly modified in order to avoid modularity function which suffers a resolution limit and, therefore, it may fail to detect small communities. We propose and use an entropy function as an optimization function and as criteria in grouping crossover operator. Experiments on real-world and synthetic networks illustrate that the proposed method can find natural partitions effectively.
This paper proposes a new framework for ground extraction and building detection in LiDAR data. The proposed approach constructs the connectivity of a grid over the LiDAR point-cloud in order to ...perform multi-scale data decomposition. This is realised by forming a top-hat scale-space using differential morphological profiles (DMPs) on points’ residuals from the approximated surface. The geometric attributes of the contained features are estimated by mapping characteristic values from DMPs. Ground definition is achieved by using features’ geometry, whilst their surface and regional attributes are additionally considered for building detection. A new algorithm for local fitting surfaces (LoFS) is proposed for extracting planar points. Finally, transitions between planar ground and non-ground regions are observed in order to separate regions of similar geometrical and surface properties but different contexts (i.e. bridges and buildings). The methods were evaluated using ISPRS benchmark datasets and show superior results in comparison to the current state-of-the-art.
The paper presents a new algorithm for reflection symmetry detection, which is specialized to detect maximal symmetric patterns in an Earth observation (EO) dataset. First, we stress the ...particularities that make symmetry detection in EO data different from detection in other geometric sets. The EO data acquisition cannot provide exact pairs of symmetric elements and, therefore, the approximate symmetry must be addressed, which is accomplished by voxelization. Besides this, the EO data symmetric patterns in the top view usually contain the most useful information for further processing and, thus, it suffices to detect symmetries with vertical symmetry planes. The algorithm first extracts the so-called interesting voxels and then finds symmetric pairs of line segments, separately for each horizontal voxel slice. The results with the same symmetry plane are then merged, first in individual slices and then through all the slices. The detected maximal symmetric patterns represent the so-called partial symmetries, which can be further processed to identify global and local symmetries. LiDAR datasets of six urban and natural attractions in Slovenia of different scales and in different voxel resolutions were analyzed in this paper, demonstrating high detection speed and quality of solutions.
Trees are natural objects, where deviations through the branches amplify geometric data for 3D representation and bring challenges to various applications dealing with 3D models, such as compression, ...visualization, symmetry detection, and radiative transfer simulation. This data article describes dataset of approximately symmetric 3D tree models with manually identified predominant symmetry plane in each tree model. Parameters for procedural tree synthesis were manually adjusted to produce approximately bilaterally symmetric trees which are grouped into species with distinct features. In the last step, each tree was manually annotated with approximate symmetry plane. This dataset contains geometric data of branches, manually defined parameters for tree synthesis method, point clouds, and a division plane with a score of bilateral symmetry strength. The generated trees can be used as benchmark data for verification of approximate reflectional symmetry detection methods. Additionally, generated 3D tree models can be used for other applications requiring pregenerated trees, such as compression of tree models, instancing, decimation methods, and radiative transfer simulation and modeling.
The Sentinel satellite constellation series, developed and operated by the European Space Agency, represents a dedicated space component of the European Copernicus Programme, committed to long-term ...operational services in the environment, climate and security. A huge amount of acquired data allow us different surveys. The paper considers the detection of changes in water levels in Lake Cerknica. The multispectral index has been calculated from Sentinel-2 data and transformed to a 3D point cloud. As shown by the results, symmetry measures of 3D point clouds could be used for the detection of water levels. Prediction functions using a genetic algorithm have been fitted, and the best result achieved was RMSE = 0.9824.
Multispectral satellite imagery offers a new perspective for spatial modelling, change detection and land cover classification. The increased demand for accurate classification of geographically ...diverse regions led to advances in object-based methods. A novel spatiotemporal method is presented for object-based land cover classification of satellite imagery using a Graph Neural Network. This paper introduces innovative representation of sequential satellite images as a directed graph by connecting segmented land region through time. The method's novel modular node classification pipeline utilises the Convolutional Neural Network as a multispectral image feature extraction network, and the Graph Neural Network as a node classification model. To evaluate the performance of the proposed method, we utilised EfficientNetV2-S for feature extraction and the GraphSAGE algorithm with Long Short-Term Memory aggregation for node classification. This innovative application on Sentinel-2 L2A imagery produced complete 4-year intermonthly land cover classification maps for two regions: Graz in Austria, and the region of Portorož, Izola and Koper in Slovenia. The regions were classified with Corine Land Cover classes. In the level 2 classification of the Graz region, the method outperformed the state-of-the-art UNet model, achieving an average F1-score of 0.841 and an accuracy of 0.831, as opposed to UNet's 0.824 and 0.818, respectively. Similarly, the method demonstrated superior performance over UNet in both regions under the level 1 classification, which contains fewer classes. Individual classes have been classified with accuracies up to 99.17%.
This paper introduces a novel approach to increase the spatiotemporal resolution of an arbitrary environmental variable. This is achieved by utilizing machine learning algorithms to construct a ...satellite-like image at any given time moment, based on the measurements from IoT sensors. The target variables are calculated by an ensemble of regression models. The observed area is gridded, and partitioned into Voronoi cells based on the IoT sensors, whose measurements are available at the considered time. The pixels in each cell have a separate regression model, and take into account the measurements of the central and neighboring IoT sensors. The proposed approach was used to assess NO2 data, which were obtained from the Sentinel-5 Precursor satellite and IoT ground sensors. The approach was tested with three different machine learning algorithms: 1-nearest neighbor, linear regression and a feed-forward neural network. The highest accuracy yield was from the prediction models built with the feed-forward neural network, with an RMSE of 15.49 ×10−6 mol/m2.