Using semivariogram indices to analyse heterogeneity in spatial patterns in remotely sensed images
(Elsevier, 2013-01) Balaguer Beser, Ángel Antonio; Ruiz Fernández, Luis Ángel; Hermosilla, T.; Recio Recio, Jorge Abel; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección; Ministerio de Ciencia e Innovación
he benchmark problem proposed in this paper is to identify regions in aerial or satellite images with geometric patterns and describe the geometric properties of the constituent elements of the pattern and their spatial distribution. This is a relevant topic in image analysis processes where spatial regular patterns are studied. This paper first presents two approaches based on multi-directional semivariograms for reducing the processing time required to compute omnidirectional semivariograms. A set of parameters for describing the structure of a semivariogram, introduced by Balaguer et al. (2010), is extracted from an experimental semivariogram and analysed to quantify the heterogeneity of the distribution of elements (trees) with periodic patterns in images of orchards. An assessment is made using four image datasets. The first dataset is composed of synthetic images that simulate regularly spaced tree crops and real images, and is used to evaluate the influence that the orientation of elements (regularly spaced trees) in the objects (crop plots) has in the descriptive parameter values. This dataset is also used to compare different semivariogram computational approaches. The other three are also composed of synthetic images and are used to evaluate the semivariogram parameters under different spatial heterogeneity conditions, and are generated by varying patterns and tree characteristics, i.e., existence or absence of faults, regular/irregular distributions, and size of the elements. Finally, the proposed methodology is applied to real aerial orthoimages of orchard plots.
Using street based metrics to characterize urban typologies
(Elsevier, 2014-03) Hermosilla, T.; Palomar Vázquez, Jesús Manuel; Balaguer Beser, Ángel Antonio; Balsa Barreiro, José; Ruiz Fernández, Luis Ángel; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección; Ministerio de Ciencia e Innovación
[EN] Urban spatial structures reflect local particularities produced during the development of a city. High spatial resolution imagery and LiDAR data are currently used to derive numerical attributes to describe in detail intra-urban structures and morphologies. Urban block boundaries have been frequently used to define the units for extracting metrics from remotely sensed data. In this paper, we propose to complement these metrics with a set of novel descriptors of the streets surrounding the urban blocks under consideration. These metrics numerically describe geometrical properties in addition to other distinctive aspects, such as presence and properties of vegetation and the relationship between the streets and buildings. For this purpose, we also introduce a methodology for partitioning the street area related to an urban block into polygons from which the street urban metrics are derived. We achieve the assessment of these metrics through application of a one-way ANOVA procedure, the winnowing technique, and a decision tree classifier. Our results suggest that street metrics, and particularly those describing the street geometry, are suitable for enhancing the discrimination of complex urban typologies and help to reduce the confusion between certain typologies. The overall classification accuracy increased from 72.7% to 81.1% after the addition street of descriptors. The results of this study demonstrate the usefulness of these metrics for describing street properties and complementing information derived from urban blocks to improve the description of urban areas. Street metrics are of particular use for the characterization of urban typologies and to study the dynamics of cities.
Description and validation of a new set of object-based temporal geostatistical features for land-use/land-cover change detection
(Elsevier, 2016-11) Gil Yepes, José Luis; Ruiz Fernández, Luis Ángel; Recio Recio, Jorge Abel; Balaguer Beser, Ángel Antonio; Hermosilla Gómez, Txomin; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección; Generalitat Valenciana; Comunidad Autónoma de la Región de Murcia
[EN] A new set of temporal features derived from codispersion and cross-semivariogram geostatistical functions is proposed, described, extracted, and evaluated for object-based land-use/land-cover change detection using high resolution images. Five features were extracted from the codispersion function and another six from the cross-semivariogram. The set of features describes the temporal behaviour of the internal structure of the image objects defined in a cadastral database. The set of extracted features was combined with spectral information and a feature selection study was performed using forward stepwise discriminant analysis, principal component analysis, as well as correlation and feature interpretation analysis. The temporal feature set was validated using high resolution aerial images from an agricultural area located in south-east Spain, in order to solve a tree crop change detection problem. Direct classification using decision tree classifier was used as change detection method. Different classifications were performed comparing various feature group combinations in order to obtain the most suitable features for this study. Results showed that the new sets of cross-semivariogram and codispersion features provided high global accuracy classification results (83.55% and 85.71% respectively), showing high potential for detecting changes related to the internal structure of agricultural tree crop parcels. A significant increase in accuracy value was obtained when combining features from both groups with spectral information (94.59%).
Non-linear fourth-order image interpolation for subpixel edge detection and localization
(Elsevier, 2008-09-01) Hermosilla, T.; Bermejo, E.; Balaguer Beser, Ángel Antonio; Ruiz Fernández, Luis Ángel; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección; Ministerio de Educación y Ciencia
[EN] A fourth-order non-linear interpolation procedure based on the ENO (Essentially Non-Oscillatory) methodology is presented and evaluated, with the purpose of increasing the geometric accuracy of edge detection in digital images. Two possible cases are considered one that considers that each pixel of the image represents a point value, the other that the pixel is an average value of a function. After image interpolation to obtain a finer grid of pixels, the Canny edge detection algorithm is applied, with the objective of improving the localization and geometry of the edges at a subpixel level. The results are compared with other schemes based on fourth order two-dimensional interpolation methods, such as a centered scheme based on a cubic convolution, a fourth order non-centered lineal scheme and a centered cubic convolution based on local gradient features. The evaluation is performed using visual and analytical techniques applied over aerial and satellite images, analyzing the positional errors of the detected edges, as well as the errors due to changes in scale and orientation. In addition to the subpixel edge detection, the quality of the interpolated images is tested. We conclude that the proposed methodology based on ENO interpolation improves the detection of edges in images as compared to other fourth-order methods
Precipitation modeling using multiple regression in R
(Universitat Politècnica de València, 2021-06-09T10:51:06Z) Riutort Mayol, Gabriel; Ruiz Fernández, Luis Ángel; Balaguer Beser, Ángel Antonio; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección
In this document we apply multiple regression methods (linear and polynomial estimators) to obtain the best estimation of the seasonal mean precipitation in the Comunitat Valenciana (Spain), using the altitude and cartographic coordinates (XUTM and YUTM), as well as their product, as multivariate estimators or independent variables. The methodology is applied and the visualization of results done using R code and functions.
Empirical Models for Spatio-Temporal Live Fuel Moisture Content Estimation in Mixed Mediterranean Vegetation Areas Using Sentinel-2 Indices and Meteorological Data
(MDPI AG, 2021-09) Costa-Saura, José M.; Balaguer Beser, Ángel Antonio; Ruiz Fernández, Luis Ángel; Pardo Pascual, Josep Eliseu; Soriano-Sancho, José L.; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección; Generalitat Valenciana
[EN] Live fuel moisture content (LFMC) is an input factor in fire behavior simulation models highly contributing to fire ignition and propagation. Developing models capable of accurately estimating spatio-temporal changes of LFMC in different forest species is needed for wildfire risk assessment. In this paper, an empirical model based on multivariate linear regression was constructed for the forest cover classified as shrublands in the central part of the Valencian region in the Eastern Mediterranean of Spain in the fire season. A sample of 15 non-monospecific shrubland sites was used to obtain a spatial representation of this type of forest cover in that area. A prediction model was created by combining spectral indices and meteorological variables. This study demonstrates that the Normalized Difference Moisture Index (NDMI) extracted from Sentinel-2 images and meteorological variables (mean surface temperature and mean wind speed) are a promising combination to derive cost-effective LFMC estimation models. The relationships between LFMC and spectral indices for all sites improved after using an additive site-specific index based on satellite information, reaching a R-adj(2) = 0.70, RMSE = 8.13%, and MAE = 6.33% when predicting the average of LFMC weighted by the canopy cover fraction of each species of all shrub species present in each sampling plot.
Selección de un modelo de regresión lineal múltiple para el cálculo de la precipitación media en verano
(Universitat Politècnica de València, 2021-06-09T10:15:33Z) Balaguer Beser, Ángel Antonio; Ruiz Fernández, Luis Ángel; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección
En este artículo se comparan distintos procedimientos para la selección de modelos de regresión lineal múltiple usando datos reales. Se aplican los métodos de selección paso a paso hacia adelante y selección paso a paso hacia atrás. Se trabaja con modelos cuyas variables son estadísticamente significativas con un Valor-P inferior a 0.05. Se selecciona el mejor modelo en función del coeficiente R-cuadrado ajustado, la raíz del error cuadrático medio (RMSE en las siglas en inglés) y el error absoluto medio (MAE en siglas en inglés), el criterio de información de Akaike, el criterio Bayesiano de Schwarz-Bayesian y el criterio de Hannan-Quinn. También se analizan los residuos de cada modelo para verificar si se cumplen las hipótesis de linealidad, homocedasticidad, independencia y normalidad de los residuos. Esta metodología se aplica para obtener modelos de regresión en la predicción de la precipitación media durante los meses del verano meteorológico en el territorio de la Comunitat Valenciana (España) y áreas adyacentes, usando algunas variables de carácter geográfico y topográfico descritas en Portalés et al. (2010). Para ello se utiliza el programa Statgraphics Centurion XVII.
Analyzing the role of pulse density and voxelization parameters on full-waveform LiDAR-derived metrics
(Elsevier, 2018) Crespo Peremarch, Pablo; Ruiz Fernández, Luis Ángel; Balaguer Beser, Ángel Antonio; Estornell Cremades, Javier; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Escuela Politécnica Superior de Gandia; Grupo de Cartografía Geoambiental y Teledetección; Ministerio de Economía y Competitividad
[EN] LiDAR full-waveform (LFW) pulse density is not homogeneous along study areas due to overlap between contiguous flight stripes and, to a lesser extent, variations in height, velocity and altitude of the platform. As a result, LFW-derived metrics extracted at the same spot but at different pulse densities differ, which is called ¿side-lap effect¿. Moreover, this effect is reflected in forest stand estimates, since they are predicted from LFW-derived metrics. This study was undertaken to analyze LFW-derived metric variations according to pulse density, voxel size and value assignation method in order to reduce the side-lap effect. Thirty LiDAR samples with a minimum density of 16 pulses.m¿2 were selected from the testing area and randomly reduced to 2 pulses.m¿2 with an interval of 1 pulse.m¿2, then metrics were extracted and compared for each sample and pulse density at different voxel sizes and assignation values. Results show that LFW-derived metric variations as a function of pulse density follow a negative exponential model similar to the exponential semivariogram curve, increasing sharply until they reach a certain pulse density, where they become stable. This value represents the minimum pulse density (MPD) in the study area to optimally minimize the side-lap effect. This effect can also be reduced with pulse densities lower than the MPD modifying LFW parameters (i.e. voxel size and assignation value). Results show that LFW-derived metrics are not equally influenced by pulse density, such as number of peaks (NP) and ROUGHness of the outermost canopy (ROUGH) that may be discarded for further analyses at large voxel sizes, given that they are highly influenced by pulse density. In addition, side-lap effect can be reduced by either increasing pulse density or voxel size, or modifying the assignation value. In practice, this leads to a proper estimate of forest stand variables using LFW data.
Modelos empíricos de predicción del contenido de humedad del combustible vivo mediante índices espectrales de Sentinel-2 y datos meteorológicos
(Editorial Universitat Politècnica de València, 2021-10-01) Arcos, María; Balaguer Beser, Ángel Antonio; Ruiz Fernández, Luis Ángel; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección; Universitat Politècnica de València
[EN] The water content of the vegetation affects the flammability of the vegetation and fire behavior. A standard measure of this parameter is the live fuel moisture content (LFMC), calculated as the percentage of humidity of the vegetation relative to its dry weight. The aim of this work was to predict LFMC values of Rosmarinus officinalis in forest areas of the Valencian Community (Spain) using spectral indices obtained from Sentinel-2 satellite images and meteorological data. For this, LFMC values of this species were obtained from field samples collected biweekly from June to October in years 2019 and 2020 in three forest plots in the province of Valencia (Spain). The meteorological data (precipitation, temperature, relative humidity and wind speed) were obtained from observatories of the State Meteorological Agency (AEMET) of Spain. Multiple linear regression models were applied to estimate LFMC, using as predictor variables different spectral indices generated from Sentinel-2 images, calculated using Google Earth Engine and R programming. The results obtained using smoothed spectral data with the Savitzky-Golay filter were compared with data without such smoothing, also considering the differential contribution of the meteorological variables in each of the interpolated dates for each plot with data from the study area.
Definition of a comprehensive set of texture semivariogram features and their evaluation for object-oriented image classification
(Elsevier, 2010-02) Balaguer Beser, Ángel Antonio; Ruiz Fernández, Luis Ángel; Hermosilla, T.; Recio Recio, Jorge Abel; Dpto. de Matemática Aplicada; Dpto. de Ingeniería Cartográfica Geodesia y Fotogrametría; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Grupo de Cartografía Geoambiental y Teledetección; Ministerio de Educación y Ciencia
In this paper, a comprehensive set of texture features extracted from the experimental semivariogram of specific image objects is proposed and described, and their usefulness for land use classification of high resolution images is evaluated. Fourteen features are defined and categorized into three different groups, according to the location of their respective parameters in the semivariogram curve: (i) features that use parameters close to the origin of the semivariogram, (ii) the parameters employed extend to the first maximum, and (iii) the parameters employed are extracted from the first to the second maximum. A selection of the most relevant features has been performed, combining the analysis and interpretation of redundancies, and using statistical discriminant analysis methods. The suitability of the proposed features for object-based image classification has been evaluated using digital aerial images from an agricultural area on the Mediterranean coast of Spain. The performance of the selected semivariogram features has been compared with two different sets of texture features: those derived from the grey level co-occurrence matrix, and the values of raw semivariance directly extracted from the semivariogram at different positions. As a result of the tests, the classification accuracies obtained using the proposed semivariogram features are, in general, higher and more balanced than those obtained using the other two sets of standard texture features.