[go: up one dir, main page]
Next Issue
Volume 5, January
Previous Issue
Volume 4, September
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
6.9
2.8
Journals
IJGI
Volume 4
Issue 4
share announcement

ISPRS Int. J. Geo-Inf., Volume 4, Issue 4 (December 2015) – 53 articles , Pages 1774-2904

  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
1956 KiB  
Case Report
The Building Blocks of User-Focused 3D City Models
by Isabel Sargent, David Holland and Jenny Harding
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2890-2904; https://doi.org/10.3390/ijgi4042890 - 21 Dec 2015
Cited by 6 | Viewed by 5972
Abstract
At Ordnance Survey, GB, we have taken an incremental approach to creating our 3D geospatial database. Research at Ordnance Survey has focused not only on methods for deriving 3D data, but also on the needs of the user in terms of the actual [...] Read more.
At Ordnance Survey, GB, we have taken an incremental approach to creating our 3D geospatial database. Research at Ordnance Survey has focused not only on methods for deriving 3D data, but also on the needs of the user in terms of the actual tasks they perform. This provides insights into the type and quality of the data required and how its quality is conveyed. In 2007, using task analysis and user-centred design, we derived a set of geometric characteristics of building exteriors that are relevant to one or more use contexts. This work has been valuable for guiding which building data to collect and how to augment our products. In 2014, we began to supply building height attributes as an alpha-release enhancement to our 2D topography data, OS MasterMap® Topography Layer. This is the first in a series of enhancements of our 2D data that forms part of a road map that will ultimately lead to a full range of 3D products. This paper outlines our research journey from the understanding of the key 3D building characteristics to the development of geo-spatial products and the specification of research. There remains a rich seam of research into methods for capturing user-focused, geo-spatial data to enable visualisation and analysis in three dimensions. Because the process of informing and designing a product is necessarily focused on the practicalities of production, storage and distribution, this paper is presented as a case report, as we believe our journey will be of interest to others involved in the capture of 3D buildings at a national level. Full article
(This article belongs to the Special Issue Efficient Capturing of 3D Objects at a National Level: With a Focus on Buildings and Infrastructure)
►▼ Show Figures

Figure 1

Figure 1
<p>This figure uses synthetic 3D building models for illustrations of the building height characteristics of 3D buildings. (<b>a</b>) Characteristic 8: ground floor height; (<b>b</b>) Characteristic 7: height of building to base of roof; (<b>c</b>) Characteristic 6: maximum height of roof ridge; (<b>d</b>) Characteristic 5: highest point of the structure.</p> Full article ">Figure 2
<p>This figure uses synthetic 3D building data to illustrate the building height attributes that are currently being provided as part of the OS MasterMap Topography Layer product. (<b>a</b>) Within our data, Ordnance Survey distinguishes between structures that sit on the ground and those that are “non-obstructing” detail. The following attributes are illustrated for three types of building: a flat-roofed building with a balustrade, a ridge-roofed building with a chimney and an overhead canopy (in this case, connected to the ridge-roofed building). (<b>b</b>) AbsHmin: the absolute minimum height of the intersection of the external building walls and the underlying ground surface; (<b>c</b>) AbsH2: the absolute height of the base of the roof; (<b>d</b>) AbsHmax: the absolute height of the highest point on the building; (<b>e</b>) RelH2: AbsH2 − AbsHmin; (<b>f</b>) RelHmax: AbsHmax − AbsHmin</p> Full article ">Figure 3
<p>Illustrations of the building height attributes that are not currently available (<b>a</b>) AbsH1: the absolute height of the base of the building; (<b>b</b>) AbsH3: the absolute maximum height of the main roof.</p> Full article ">
8100 KiB  
Review
Applications of 3D City Models: State of the Art Review
by Filip Biljecki, Jantien Stoter, Hugo Ledoux, Sisi Zlatanova and Arzu Çöltekin
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2842-2889; https://doi.org/10.3390/ijgi4042842 - 18 Dec 2015
Cited by 596 | Viewed by 54773
Abstract
In the last decades, 3D city models appear to have been predominantly used for visualisation; however, today they are being increasingly employed in a number of domains and for a large range of tasks beyond visualisation. In this paper, we seek to understand [...] Read more.
In the last decades, 3D city models appear to have been predominantly used for visualisation; however, today they are being increasingly employed in a number of domains and for a large range of tasks beyond visualisation. In this paper, we seek to understand and document the state of the art regarding the utilisation of 3D city models across multiple domains based on a comprehensive literature study including hundreds of research papers, technical reports and online resources. A challenge in a study such as ours is that the ways in which 3D city models are used cannot be readily listed due to fuzziness, terminological ambiguity, unclear added-value of 3D geoinformation in some instances, and absence of technical information. To address this challenge, we delineate a hierarchical terminology (spatial operations, use cases, applications), and develop a theoretical reasoning to segment and categorise the diverse uses of 3D city models. Following this framework, we provide a list of identified use cases of 3D city models (with a description of each), and their applications. Our study demonstrates that 3D city models are employed in at least 29 use cases that are a part of more than 100 applications. The classified inventory could be useful for scientists as well as stakeholders in the geospatial industry, such as companies and national mapping agencies, as it may serve as a reference document to better position their operations, design product portfolios, and to better understand the market. Full article
(This article belongs to the Special Issue Efficient Capturing of 3D Objects at a National Level: With a Focus on Buildings and Infrastructure)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>3D city models may be applied in a multitude of application domains for environmental simulations and decision support.</p> Full article ">Figure 2
<p>The tangled relations between spatial operations (dark blue), use cases (light blue) and applications (outlined with a dashed stroke), and their overlap which prevents a straightforward listing and classification. This example is simplified, as the application domains normally have more corresponding use cases, and use cases usually consist of more than a few spatial operations.</p> Full article ">Figure 3
<p>Estimation of the solar irradiation of buildings for a specific date and time. In this case the surrounding vegetation and the type of the receiving material is also taken into account in the estimations. (Image courtesy of Argedor).</p> Full article ">Figure 4
<p>Results of the estimation of the heat demand of buildings. (Image courtesy of Jean-Marie Bahu, EIFER [<a href="#B126-ijgi-04-02842" class="html-bibr">126</a>]).</p> Full article ">Figure 5
<p>Estimating the shadow cast by a building for a few positions of the sun. For instance, this use case is valuable in assessing the effect that a proposed building design has on its surrounding. (Image courtesy of CyberCity 3D).</p> Full article ">Figure 6
<p>A 3D noise simulation derived with a 3D city model. (Image courtesy of Kurakula [<a href="#B243-ijgi-04-02842" class="html-bibr">243</a>]).</p> Full article ">Figure 7
<p>Example of a complex property situation (in Rotterdam, The Netherlands) in which the limitations of 2D cadastre are exposed. The corresponding cadastral map (courtesy of the Dutch Kadaster) shows that multiple small parcels are necessary to register a single object.</p> Full article ">Figure 8
<p>3D city models may be used for simulation and analysis of the effects of explosions in urban areas. This example shows the blast pressure wave propagation in urban environments. Possible applications are the prediction of effects of structural integrity and soundness of the urban infrastructure, and aiding safety preparations for evacuation in the case of bomb discovery and defuse. (Image courtesy of virtualcitySYSTEMS).</p> Full article ">
35881 KiB  
Article
An Approach for Indoor Path Computation among Obstacles that Considers User Dimension
by Liu Liu and Sisi Zlatanova
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2821-2841; https://doi.org/10.3390/ijgi4042821 - 17 Dec 2015
Cited by 19 | Viewed by 7305
Abstract
People often transport objects within indoor environments, who need enough space for the motion. In such cases, the accessibility of indoor spaces relies on the dimensions, which includes a person and her/his operated objects. This paper proposes a new approach to avoid obstacles [...] Read more.
People often transport objects within indoor environments, who need enough space for the motion. In such cases, the accessibility of indoor spaces relies on the dimensions, which includes a person and her/his operated objects. This paper proposes a new approach to avoid obstacles and compute indoor paths with respect to the user dimension. The approach excludes inaccessible spaces for a user in five steps: (1) compute the minimum distance between obstacles and find the inaccessible gaps; (2) group obstacles according to the inaccessible gaps; (3) identify groups of obstacles that influence the path between two locations; (4) compute boundaries for the selected groups; and (5) build a network in the accessible area around the obstacles in the room. Compared to the Minkowski sum method for outlining inaccessible spaces, the proposed approach generates simpler polygons for groups of obstacles that do not contain inner rings. The creation of a navigation network becomes easier based on these simple polygons. By using this approach, we can create user- and task-specific networks in advance. Alternatively, the accessible path can be generated on the fly before the user enters a room. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Minkowski sums of obstacles to a user and the minimum distance between obstacles. (<b>a</b>) Minkowski sum of obstacles for a user approximated as a circle; (<b>b</b>) union of the Minkowski sum of obstacles and the minimum distance.</p> Full article ">Figure 2
<p>The union of Minkowski sums contains inner rings and self-intersections. (<b>a</b>) Union of Minkowski sums with one inner ring (the circle denotes a user); (<b>b</b>) self-intersection and inner rings of the Minkowski sums.</p> Full article ">Figure 3
<p>Creating polygonal boundaries of objects according to inaccessible gaps between the objects (the circle denotes a user with tools, and red lines represent inaccessible gaps). (<b>a</b>) The polygonal boundary for the six objects without inner rings; (<b>b</b>) the simple boundary for the nine objects.</p> Full article ">Figure 4
<p>Computing a path with simple boundaries of obstacle groups for a user. (<b>a</b>) Selecting groups of obstacles between two locations (the circle denotes the user; boundaries of obstacles are yellow; and the blue line is the direct path); (<b>b</b>) the navigation network considering the inaccessible gaps between obstacles and walls (blue lines denote the buffer of walls; red lines denote inaccessible gaps; and black lines form the network); (<b>c</b>) a schematic representation of the computed shortest path on the network (the path is black); (<b>d</b>) a realistic path by taking into account the size of the user (circles denote the user; and black lines are the path).</p> Full article ">Figure 5
<p>Overview of the proposed approach.</p> Full article ">Figure 6
<p>Computing bottlenecks where a user with a given dimension cannot pass.</p> Full article ">Figure 7
<p>Grouping 11 obstacles into three groups with respect to the bottlenecks.</p> Full article ">Figure 8
<p>Selecting obstacle groups with respect to the start and target location of a user. (<b>a</b>) The direct path intersects two obstacles; (<b>b</b>) selecting the groups of the intersected obstacles; (<b>c</b>) computing a convex hull (CH), and the CH intersects other obstacles; (<b>d</b>) selecting the group of new obstacles in and re-computing the CH; (<b>e</b>) no more obstacles intersect the CH, and then, three groups are selected.</p> Full article ">Figure 9
<p>Resulting non-overlapping boundaries of obstacle groups.</p> Full article ">Figure 10
<p>The boundary generation of three obstacle groups for a user with a size of 0.8 m.</p> Full article ">Figure 11
<p>Detecting bottlenecks between a wall and the boundary of an obstacle group and identifying inaccessible edges crossing the bottlenecks.</p> Full article ">Figure 12
<p>Creating a visibility graph (VG) and a room buffer, removing the inaccessible edges and computing a path for users with a size of 0.8 m. (<b>a</b>) Creating a VG; (<b>b</b>) computing a room buffer; (<b>c</b>) finding inaccessible edges in the bottlenecks; (<b>d</b>) removing the inaccessible edges; (<b>e</b>) the computed path.</p> Full article ">Figure 13
<p>The floor plans of two buildings. (<b>a</b>) The floor of the conventional neonatal intensive care unit (CNICU) at the hospital; (<b>b</b>) ground floor of the Architecture Faculty building.</p> Full article ">Figure 14
<p>Testing the complete approach between two locations for a user with a size of 0.6 m. (<b>a</b>) Computing the bottlenecks between obstacles; (<b>b</b>) grouping the obstacles; (<b>c</b>) selecting obstacle groups; (<b>d</b>) creating the boundaries of the selected groups; (<b>e</b>) creating a VG and removing the inaccessible edges.</p> Full article ">Figure 15
<p>Path-finding for the sizes of 0.5 m, 0.6 m and 0.8 m on the hospital floor plan. (<b>a</b>) The shortest path for the 0.5 m size; (<b>b</b>) the shortest path for the 0.6 m size; (<b>c</b>) no path for the 0.8 m size.</p> Full article ">Figure 16
<p>The shortest paths for the sizes 0.5 m, 0.8 m and 1.0 m on the floor plan of the campus building. (<b>a</b>) Path for the 0.5 m size; (<b>b</b>) path for the 0.8 m size; (<b>c</b>) path for the 1.0 m size.</p> Full article ">Figure 17
<p>(<b>a</b>) The VG regardless of a user’s dimension and (<b>b</b>–<b>d</b>) all of the shortest paths between all doors for the sizes of 0.5 m, 0.6 m and 0.8 m. (a) The complete VG of all obstacles without respect to a user’s dimension; (b) all of the shortest paths for the user with a size of 0.5 m; (c) all of the shortest paths for a user with a size of 0.6 m; (d) all of the shortest paths for the user with a size of 0.8 m.</p> Full article ">Figure 18
<p>The computed and genuine minimum distances (MDs) between a convex and a non-convex obstacle. (<b>a</b>) The computed MD in our approach; (<b>b</b>) the genuine MD.</p> Full article ">Figure 19
<p>The boundaries of three obstacles in our method and in a strict condition. (<b>a</b>) Three obstacles; (<b>b</b>) our boundary for the group of three obstacles; (<b>c</b>) a strict boundary for the group of three obstacles.</p> Full article ">Figure 20
<p>(<b>a</b>) A user can pass through a gap larger than the diameter of the user’s circle; (<b>b</b>) the user cannot pass through the gap when the diameter is larger; and (<b>c</b>) the user can pass through the gap after a 90-degree turn. (a) The user can pass through the gap when its length is longer than the diameter of the circle; (b) the user cannot pass the gap when its length is shorter than the diameter of the circle; (c) the user can pass through the gap because the user’s short side is smaller than the gap, though the diameter of the circle is larger than the gap.</p> Full article ">
1656 KiB  
Article
Estimating Plant Traits of Grasslands from UAV-Acquired Hyperspectral Images: A Comparison of Statistical Approaches
by Alessandra Capolupo, Lammert Kooistra, Clara Berendonk, Lorenzo Boccia and Juha Suomalainen
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2792-2820; https://doi.org/10.3390/ijgi4042792 - 10 Dec 2015
Cited by 114 | Viewed by 15998
Abstract
Grassland ecosystems cover around 40% of the entire Earth’s surface. Therefore, it is necessary to guarantee good grassland management at field scale in order to improve its conservation and to achieve optimal growth. This study identified the most appropriate statistical strategy, between partial [...] Read more.
Grassland ecosystems cover around 40% of the entire Earth’s surface. Therefore, it is necessary to guarantee good grassland management at field scale in order to improve its conservation and to achieve optimal growth. This study identified the most appropriate statistical strategy, between partial least squares regression (PLSR) and narrow vegetation indices, for estimating the structural and biochemical grassland traits from UAV-acquired hyperspectral images. Moreover, the influence of fertilizers on plant traits for grasslands was analyzed. Hyperspectral data were collected from an experimental field at the farm Haus Riswick, near Kleve in Germany, for two different flight campaigns in May and October. The collected image blocks were geometrically and radiometrically corrected for surface reflectance. Spectral signatures extracted for the plots were adopted to derive grassland traits by computing PLSR and the following narrow vegetation indices: the MERIS Terrestrial Chlorophyll Index (MTCI), the ratio of the Modified Chlorophyll Absorption in Reflectance and Optimized Soil-Adjusted Vegetation Index (MCARI/OSAVI) modified by Wu, the Red-edge Chlorophyll Index (CIred-edge), and the Normalized Difference Red Edge (NDRE). PLSR showed promising results for estimating grassland structural traits and gave less satisfying outcomes for the selected chemical traits (crude ash, crude fiber, crude protein, Na, K, metabolic energy). Established relations are not influenced by the type and the amount of fertilization, while they are affected by the grassland health status. PLSR is found to be the best strategy, among the approaches analyzed in this paper, for exploring structural and biochemical features of grasslands. Using UAV-based hyperspectral sensing allows for the highly detailed assessment of grassland experimental plots. Full article
(This article belongs to the Special Issue Unmanned Aerial Vehicles in Geomatics)
►▼ Show Figures

Figure 1

Figure 1
<p>RGB ortho-mosaic of the experimental grassland plots on 15 May 2014, acquired using the Panasonic GX1 camera of the Wageningen UR Hyperspectral Mapping System (HYMSY).</p> Full article ">Figure 2
<p>Map of different types and levels of fertilization applied on the experimental grassland plots. The 60 plots were split into four groups of 15 plots and for every group the treatments were applied as indicated in the table. The black line is a measurement rail which is also shown in <a href="#ijgi-04-02792-f001" class="html-fig">Figure 1</a>.</p> Full article ">Figure 3
<p>The octocopter UAV Aerialtronics Altura AT8 v1A equipped with all necessary hardware and software tools for its control and programming.</p> Full article ">Figure 4
<p>Influence of the type and the amount of fertilizer treatment on grassland traits and their standard deviations.</p> Full article ">Figure 5
<p>Average and standard deviation of the spectra for the three different levels of organic fertilization (170, 230, and 340 kgN/ha) of the May and October harvest.</p> Full article ">Figure 6
<p>Correlogram between spectral variables of the hyperspectral dataset and height (H), fresh biomass (FB), crude protein (CP), and metabolic energy (ME) for the May harvest.</p> Full article ">Figure 7
<p>Scatterplot related to the influence of the type and the amount of fertilizer on selected narrow-band vegetation indices.</p> Full article ">Figure 8
<p>Scatterplot of the best relationships between selected grassland traits and vegetation indices. The best results of the linear regression model for height, fresh matter yield, and crude protein were found in the integrated dataset of May and October, including both organic and inorganic fertilizers; the best result for Metabolic Energy was found instead in the May dataset, combining both organic and inorganic fertilized plots.</p> Full article ">Figure 9
<p>Height variation within plots by applying the NDRE index on the acquired hyperspectral dataset of May. The plots with the lowest inorganic fertilization level (0 kgN/ha) are indicated with an arrow down and plots with the highest inorganic fertilization level (340 kgN/ha) are indicated with an arrow up.</p> Full article ">Figure 10
<p>Metabolic energy variation within plots by applying the CI<sub>red-edge</sub> index on the acquired hyperspectral dataset of May.</p> Full article ">Figure 11
<p>Scatterplot of the measured <span class="html-italic">vs.</span> predicted values for the best PLSR models presented for the integrated dataset of May and October and composed by both organic and inorganic fertilization plots (LOO = Leave-One-Out validation).</p> Full article ">Figure 12
<p>PLSR coefficients for height, fresh matter yield, crude protein, and metabolic energy for the PLSR model of the integrated dataset of May and October, including both organic and inorganic fertilization plots.</p> Full article ">
10361 KiB  
Article
Combining 2D Mapping and Low Density Elevation Data in a GIS for GNSS Shadow Prediction
by Conor Cahalane
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2769-2791; https://doi.org/10.3390/ijgi4042769 - 10 Dec 2015
Viewed by 5287
Abstract
The number of satellites visible to a Global Navigation Satellite System (GNSS) receiver is important for high accuracy surveys. To aid with this, there are software packages capable of predicting GNSS visibility at any location of the globe at any time of day. [...] Read more.
The number of satellites visible to a Global Navigation Satellite System (GNSS) receiver is important for high accuracy surveys. To aid with this, there are software packages capable of predicting GNSS visibility at any location of the globe at any time of day. These prediction packages operate by using regularly updated almanacs containing positional data for all navigation satellites; however, one issue that restricts their use is that most packages assume that there are no obstructions on the horizon. In an attempt to improve this, certain planning packages are now capable of modelling simple obstructions whereby portions of the horizon visible from one location can be blocked out, thereby simulating buildings or other vertical structures. While this is useful for static surveys, it is not applicable for dynamic surveys when the GNSS receiver is in motion. This problem has been tackled in the past by using detailed, high-accuracy building models and designing novel methods for modelling satellite positions using GNSS almanacs, which is a time-consuming and costly approach. The solution proposed in this paper is to use a GIS to combine existing, freely available GNSS prediction software to predict pseudo satellite locations, incorporate a 2.5D model of the buildings in an area created with national mapping agency 2D vector mapping and low density elevation data to minimise the need for a full survey, thereby providing savings in terms of cost and time. Following this, the ESRI ArcMap viewshed tool was used to ascertain what areas exhibit poor GNSS visibility due to obstructions over a wide area, and an accuracy assessment of the procedure was made. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>GNSS Survey Planning (<b>a</b>) GNSS signal during topographic surveys can be obstructed by surrounding objects (<b>b</b>) specifying portions of the horizon in Trimble Planning where obstructions are present to simulate buildings or other vertical structures.</p> Full article ">Figure 2
<p>Mobile Mapping Systems and GNSS Quality (<b>a</b>) the XP1 MMS, designed and developed at the NCG (<b>b</b>) a 2D plot of GNSS satellite signal exhibiting signal loss due to obstructions during an MMS survey.</p> Full article ">Figure 3
<p>Modelling obstructions (<b>a</b>) a 3D vector model of Maynooth University South Campus created during the 3D Campus project (<b>b</b>) a 2.5D raster DSM of buildings on Maynooth University North Campus created using photogrammetric methods from imagery captured by a Falcon 8 UAV.</p> Full article ">Figure 4
<p>Ordnance Survey Ireland 2D vector mapping (<b>a</b>) all layers active in the test area viewed in a CAD environment (<b>b</b>) building footprint layer isolated for obstruction modelling.</p> Full article ">Figure 5
<p>Calculating satellite positions (<b>a</b>) prediction results from Trimble Planning for 10 min intervals over the test site (<b>b</b>) a 2D plot of pseudo satellites positions at 10 min intervals over Maynooth University between 10:00 and 17:00.</p> Full article ">Figure 6
<p>Processing the 2D vector polygons (<b>a</b>) identifying the X ,Y coordinates of each building polygon vertex (<b>b</b>) applying a 1 m offset to the polygons to aid in triangulated irregular network creation.</p> Full article ">Figure 7
<p>Creating surfaces to model obstructions (<b>a</b>) TIN without 1 m offset exhibits poor definition of elevation changes (<b>b</b>) low quality raster DSM created using TIN without 1 m offset (<b>c</b>) 1 m offset results in an improved TIN (<b>d</b>) improved TIN results in an improved raster DSM.</p> Full article ">Figure 8
<p>Creating a viewshed observer point using the pseudo satellite position.</p> Full article ">Figure 9
<p>Results and accuracy Tests (<b>a</b>) visualising output from the proposed methodology—green areas: visible to four plus satellites, red: less than four satellites (<b>b</b>) a plot of satellite azimuths throughout the validations tests—red represents azimuth and number of satellites, grey numbers on Y axis represent total number of satellites visible from the observer location throughout the tests.</p> Full article ">Figure 10
<p>Validation tests (<b>a</b>) validation points selected as representative of the surrounding environment (<b>b</b>) azimuth of the two predominant shadowing objects throughout the validation tests at each of the ten test locations.</p> Full article ">Figure 11
<p>Possible error sources in GNSS predictions and validation (<b>a</b>) the goal is LOS to five satellites (<b>b</b>) multipath resulting in errors in validation results (<b>c</b>) a discrepancy between LOD2 models which better approximate real world objects and LOD1 models which ignore roof structure created with 2D vector mapping.</p> Full article ">
2586 KiB  
Article
Spatial Sampling Strategies for the Effect of Interpolation Accuracy
by Hairong Zhang, Lijiang Lu, Yanhua Liu and Wei Liu
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2742-2768; https://doi.org/10.3390/ijgi4042742 - 8 Dec 2015
Cited by 13 | Viewed by 6945
Abstract
Spatial interpolation methods are widely used in various fields and have been studied by several scholars with one or a few specific sampling datasets that do not reflect the complexity of the spatial characteristics and lead to conclusions that cannot be widely applied. [...] Read more.
Spatial interpolation methods are widely used in various fields and have been studied by several scholars with one or a few specific sampling datasets that do not reflect the complexity of the spatial characteristics and lead to conclusions that cannot be widely applied. In this paper, three factors that affect the accuracy of interpolation have been considered, i.e., sampling density, sampling mode, and sampling location. We studied the inverse distance weighted (IDW), regular spline (RS), and ordinary kriging (OK) interpolation methods using 162 DEM datasets considering six sampling densities, nine terrain complexities, and three sampling modes. The experimental results show that, in selective sampling and combined sampling, the maximum absolute errors of interpolation methods rapidly increase and the estimated values are overestimated. In regular-grid sampling, the RS method has the highest interpolation accuracy, and IDW has the lowest interpolation accuracy. However, in both selective and combined sampling, the accuracy of the IDW method is significantly improved and the RS method performs worse. The OK method does not significantly change between the three sampling modes. The following conclusion can be obtained from the above analysis: the combined sampling mode is recommended for sampling, and more sampling points should be added in the ridges, valleys, and other complex terrain. The IDW method should not be used in the regular-grid sampling mode, but it has good performance in the selective sampling mode and combined sampling mode. However, the RS method shows the opposite phenomenon. The sampling dataset should be analyzed before using the OK method, which can select suitable models based on the analysis results of the sampling dataset. Full article
(This article belongs to the Special Issue Bridging the Gap between Geospatial Theory and Technology)
►▼ Show Figures

Figure 1

Figure 1
<p>Terrain data with different terrain complexity. (<b>A</b>) <span class="html-italic">D</span> = 2.0; (<b>B</b>) <span class="html-italic">D</span> = 2.1; (<b>C</b>) <span class="html-italic">D</span> = 2.2; (<b>D</b>) <span class="html-italic">D</span> = 2.3; (<b>E</b>) <span class="html-italic">D</span> = 2.4; (<b>F</b>) <span class="html-italic">D</span> = 2.5; (<b>G</b>) <span class="html-italic">D</span> = 2.6; (<b>H</b>) <span class="html-italic">D</span> = 2.7; (<b>I</b>) <span class="html-italic">D</span> = 2.8.</p> Full article ">Figure 2
<p>Flow diagram of the experiment.</p> Full article ">Figure 3
<p>The distributions of errors in regular-grid sampling. (<b>A</b>) Error distributions of IDW; (<b>B</b>) Error distributions of IDW; (<b>C</b>) Error distributions of IDW.</p> Full article ">Figure 4
<p>The distributions of errors in selective sampling. (<b>A</b>) Error distributions of IDW; (<b>B</b>) Error distributions of IDW; (<b>C</b>) Error distributions of IDW.</p> Full article ">Figure 5
<p>The distributions of errors in combined sampling. (<b>A</b>) Error distributions of IDW; (<b>B</b>) Error distributions of IDW; (<b>C</b>) Error distributions of IDW.</p> Full article ">Figure 6
<p>The <span class="html-italic">RMSE</span>s of the 3 methods in regular-grid sampling. Sampling densities of (<b>A</b>) 18.1 points/km<sup>2</sup>; (<b>B</b>) 8.5 points/km<sup>2</sup>; (<b>C</b>) 4.7 points/km<sup>2</sup>; (<b>D</b>) 3.1 points/km<sup>2</sup>; (<b>E</b>) 2.3 points/km<sup>2</sup>; (<b>F</b>) 1.8 points/km<sup>2</sup>.</p> Full article ">Figure 7
<p>The <span class="html-italic">RMSE</span>s of the 3 methods in selective sampling. Sampling densities of (<b>A</b>) 18.1 points/km<sup>2</sup>; (<b>B</b>) 8.5 points/km<sup>2</sup>; (<b>C</b>) 4.7 points/km<sup>2</sup>; (<b>D</b>) 3.1 points/km<sup>2</sup>; (<b>E</b>) 2.3 points/km<sup>2</sup>; (<b>F</b>) 1.8 points/km<sup>2</sup>.</p> Full article ">Figure 8
<p>The <span class="html-italic">RMSE</span>s of the 3 methods in combined sampling. Sampling densities of (<b>A</b>) 18.1 points/km<sup>2</sup>; (<b>B</b>) 8.5 points/km<sup>2</sup>; (<b>C</b>) 4.7 points/km<sup>2</sup>; (<b>D</b>) 3.1 points/km<sup>2</sup>; (<b>E</b>) 2.3 points/km<sup>2</sup>; (<b>F</b>) 1.8 points/km<sup>2</sup>.</p> Full article ">Figure 9
<p>The <span class="html-italic">RMSE</span>s of the 3 methods in regular-grid sampling. (<b>A</b>) The trend surface of <span class="html-italic">RMSE</span>s for IDW; (<b>B</b>) The trend surface of <span class="html-italic">RMSE</span>s for RS; (<b>C</b>) The trend surface of <span class="html-italic">RMSE</span>s for OK.</p> Full article ">Figure 10
<p>The <span class="html-italic">RMSE</span>s of the 3 methods in selective sampling. (<b>A</b>) The trend surface of <span class="html-italic">RMSE</span>s for IDW; (<b>B</b>) The trend surface of <span class="html-italic">RMSE</span>s for RS; (<b>C</b>) The trend surface of <span class="html-italic">RMSE</span>s for OK.</p> Full article ">Figure 11
<p>The <span class="html-italic">RMSE</span>s of the 3 methods in combined sampling. (<b>A</b>) The trend surface of <span class="html-italic">RMSE</span>s for IDW; (<b>B</b>) The trend surface of <span class="html-italic">RMSE</span>s for RS; (<b>C</b>) The trend surface of <span class="html-italic">RMSE</span>s for OK.</p> Full article ">Figure 12
<p>The locations of the sampling points before modification. (<b>A</b>) Sampling points before modification of area A; (<b>B</b>) Sampling points before modification of area B.</p> Full article ">Figure 13
<p>The locations of the sampling points after modification. (<b>A</b>) Sampling points before modification of area A; (<b>B</b>) Sampling points before modification of area B.</p> Full article ">Figure 14
<p>The variations of the errors at the sampling points of area A. (<b>A</b>) Error variations at the sampling points of IDW in area A; (<b>B</b>) Error variations at the sampling points of OK in area A; (<b>C</b>) Error variations at the sampling points of RS in area A.</p> Full article ">Figure 15
<p>The variations of the errors at the sampling points of area A. (<b>A</b>) Error variations at the sampling points of IDW in area A; (<b>B</b>) Error variations at the sampling points of OK in area A; (<b>C</b>) Error variations at the sampling points of RS in area A.</p> Full article ">
2590 KiB  
Article
Fractal Characterization of Settlement Patterns and Their Spatial Determinants in Coastal Zones
by Zhonghao Zhang, Xiaoqin Yang and Rui Xiao
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2728-2741; https://doi.org/10.3390/ijgi4042728 - 3 Dec 2015
Cited by 4 | Viewed by 4963
Abstract
Using box-counting and spatial regression, this paper analyzes the morphological characteristics of coastal settlement patterns and their spatial determinants, with a case of the Wen-Tai region on the Chinese eastern coast. Coastal settlement patterns, which reflect the interactions between people and the surrounding [...] Read more.
Using box-counting and spatial regression, this paper analyzes the morphological characteristics of coastal settlement patterns and their spatial determinants, with a case of the Wen-Tai region on the Chinese eastern coast. Coastal settlement patterns, which reflect the interactions between people and the surrounding environment, can indicate the anthropogenic pressure sustained in the coastal zones. Characterization of settlement patterns in coastal zones is definitely needed for coastal management. Results indicate that coastal settlement patterns in the Wen-Tai region present significant fractal characteristics, and exhibit obvious spatial variations. The morphological characteristics of settlement patterns are significantly correlated with the standard deviation value of elevation and slope, as well as percentage of loam soils. In particular, cities with greater relief amplitude, higher slope variability, and higher percentage of loam soils would present more complexity in form. Proximity to roads and rivers are insignificant determinants. Our study contributes to the understanding of the spatial determinants of the morphological characteristics of settlement patterns in coastal zones. We argue that fractal dimension provides a useful tool to facilitate the identification of vulnerability hotspots for coastal studies. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Location of the Wen-Tai region, China.</p> Full article ">Figure 2
<p>Spatial patterns of settlement locations, roads, and rivers across the Wen-Tai region, China.</p> Full article ">Figure 3
<p>Graphs of number of boxes N(r) against box side length (in meters) on logarithmic scales for each city across the Wen-Tai region, China. (FD = Fractal Dimension value).</p> Full article ">
1433 KiB  
Article
Collaborative Strategies for Sustainable EU Flood Risk Management: FOSS and Geospatial Tools—Challenges and Opportunities for Operative Risk Analysis
by Raffaele Albano, Leonardo Mancusi, Aurelia Sole and Jan Adamowski
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2704-2727; https://doi.org/10.3390/ijgi4042704 - 2 Dec 2015
Cited by 36 | Viewed by 9796
Abstract
An analysis of global statistics shows a substantial increase in flood damage over the past few decades. Moreover, it is expected that flood risk will continue to rise due to the combined effect of increasing numbers of people and economic assets in risk-prone [...] Read more.
An analysis of global statistics shows a substantial increase in flood damage over the past few decades. Moreover, it is expected that flood risk will continue to rise due to the combined effect of increasing numbers of people and economic assets in risk-prone areas and the effects of climate change. In order to mitigate the impact of natural hazards on European economies and societies, improved risk assessment, and management needs to be pursued. With the recent transition to a more risk-based approach in European flood management policy, flood analysis models have become an important part of flood risk management (FRM). In this context, free and open-source (FOSS) geospatial models provide better and more complete information to stakeholders regarding their compliance with the Flood Directive (2007/60/EC) for effective and collaborative FRM. A geospatial model is an essential tool to address the European challenge for comprehensive and sustainable FRM because it allows for the use of integrated social and economic quantitative risk outcomes in a spatio-temporal domain. Moreover, a FOSS model can support governance processes using an interactive, transparent and collaborative approach, providing a meaningful experience that both promotes learning and generates knowledge through a process of guided discovery regarding flood risk management. This article aims to organize the available knowledge and characteristics of the methods available to give operational recommendations and principles that can support authorities, local entities, and the stakeholders involved in decision-making with regard to flood risk management in their compliance with the Floods Directive (2007/60/EC). Full article
(This article belongs to the Special Issue Bridging the Gap between Geospatial Theory and Technology)
►▼ Show Figures

Figure 1

Figure 1
<p>Flood Risk Concept Chain.</p> Full article ">Figure 2
<p>Overview of general flood assessment.</p> Full article ">Figure 3
<p>Example of change in susceptibility on the basis of the economic sector embodied by depth-damage curves [<a href="#B34-ijgi-04-02704" class="html-bibr">34</a>].</p> Full article ">Figure 4
<p>Comparison of total damage for the urban sector obtained with different damage models.</p> Full article ">Figure 5
<p>Example of the spatial distribution of the damage in a hypothetical study case.</p> Full article ">Figure 6
<p>Example of risk information based on comparison of F–D curves for an idealized case study.</p> Full article ">Figure 7
<p>Example of risk information based on comparison of F-N curves for an idealized case study.</p> Full article ">Figure 8
<p>The final framework of recommendations for supporting a pan-European flood risk management approach.</p> Full article ">
1742 KiB  
Article
Weather Conditions, Weather Information and Car Crashes
by Adriaan Perrels, Athanasios Votsis, Väinö Nurmi and Karoliina Pilli-Sihvola
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2681-2703; https://doi.org/10.3390/ijgi4042681 - 27 Nov 2015
Cited by 21 | Viewed by 8131
Abstract
Road traffic safety is the result of a complex interaction of factors, and causes behind road vehicle crashes require different measures to reduce their impacts. This study assesses how strongly the variation in daily winter crash rates associates with weather conditions in Finland. [...] Read more.
Road traffic safety is the result of a complex interaction of factors, and causes behind road vehicle crashes require different measures to reduce their impacts. This study assesses how strongly the variation in daily winter crash rates associates with weather conditions in Finland. This is done by illustrating trends and spatiotemporal variation in the crash rates, by showing how a GIS application can evidence the association between temporary rises in regional crash rates and the occurrence of bad weather, and with a regression model on crash rate sensitivity to adverse weather conditions. The analysis indicates that a base rate of crashes depending on non-weather factors exists, and some combinations of extreme weather conditions are able to substantially push up crash rates on days with bad weather. Some spatial causation factors, such as variation of geophysical characteristics causing systematic differences in the distributions of weather variables, exist. Yet, even in winter, non-spatial factors are normally more significant. GIS data can support optimal deployment of rescue services and enhance in-depth quantitative analysis by helping to identify the most appropriate spatial and temporal resolutions. However, the supportive role of GIS should not be inferred as existence of highly significant spatial causation. Full article
(This article belongs to the Special Issue Geoinformation for Disaster Risk Management)
►▼ Show Figures

Figure 1

Figure 1
<p>Number of fatalities per million inhabitants. <b>Source:</b> EC DG Mobility and Transport—Road Safety Statistics website [<a href="#B30-ijgi-04-02681" class="html-bibr">30</a>].</p> Full article ">Figure 2
<p>Monthly crash rates (crashes per million vehicle km) and average number of casualties per accident for the period 2000–2010. (crash rates: left-hand scale; casualties: right-hand scale). <b>Source</b>: study dataset (see <a href="#sec3-ijgi-04-02681" class="html-sec">Section 3</a>).</p> Full article ">Figure 3
<p>Daily number of crashes by day of the week by region in November and December 2000–2010. <b>Source</b>: study dataset (see <a href="#sec3-ijgi-04-02681" class="html-sec">Section 3</a>).</p> Full article ">Figure 4
<p>An example of spatiotemporal association between crash rates and severe weather.</p> Full article ">Figure 4 Cont.
<p>An example of spatiotemporal association between crash rates and severe weather.</p> Full article ">Figure 5
<p><b>(a)</b> Scatterplot of daily averages augmented by 2x standard deviation for precipitation and wind speed by region in winter periods based on observations from 2000 to 2010; and <b>(b)</b> Regional clustering (along the coast) of the highest observed combined wind speed and precipitation maximums (product of both values).</p> Full article ">Figure 6
<p>(<b>a</b>) Road vehicle crash rates (per million vehicle kilometers) in normal weather conditions in 2000 per Finnish province; (<b>b</b>) Crash rates in normal weather conditions in 2010, and growth from 2000; (<b>c</b>) Crash rates in very bad weather conditions in 2010. The colour scale divisions are common for all maps.</p> Full article ">
1480 KiB  
Article
Lane-Level Road Information Mining from Vehicle GPS Trajectories Based on Naïve Bayesian Classification
by Luliang Tang, Xue Yang, Zihan Kan and Qingquan Li
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2660-2680; https://doi.org/10.3390/ijgi4042660 - 26 Nov 2015
Cited by 60 | Viewed by 7587
Abstract
In this paper, we propose a novel approach for mining lane-level road network information from low-precision vehicle GPS trajectories (MLIT), which includes the number and turn rules of traffic lanes based on naïve Bayesian classification. First, the proposed method (MLIT) uses an adaptive [...] Read more.
In this paper, we propose a novel approach for mining lane-level road network information from low-precision vehicle GPS trajectories (MLIT), which includes the number and turn rules of traffic lanes based on naïve Bayesian classification. First, the proposed method (MLIT) uses an adaptive density optimization method to remove outliers from the raw GPS trajectories based on their space-time distribution and density clustering. Second, MLIT acquires the number of lanes in two steps. The first step establishes a naïve Bayesian classifier according to the trace features of the road plane and road profiles and the real number of lanes, as found in the training samples. The second step confirms the number of lanes using test samples in reference to the naïve Bayesian classifier using the known trace features of test sample. Third, MLIT infers the turn rules of each lane through tracking GPS trajectories. Experiments were conducted using the GPS trajectories of taxis in Wuhan, China. Compared with human-interpreted results, the automatically generated lane-level road network information was demonstrated to be of higher quality in terms of displaying detailed road networks with the number of lanes and turn rules of each lane. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Lane information extraction architecture.</p> Full article ">Figure 2
<p>Trajectories optimization. (<b>a</b>) shows the distribution of tracking points, <span class="html-italic">p</span> is the central point and the circle in (a) is the neighborhood of <span class="html-italic">p</span>; (<b>b</b>)indicates a subset of tracking points and be denoted as <span class="html-italic">A</span>; (<b>c</b>) is the Delaunay triangulation of <span class="html-italic">A</span>.</p> Full article ">Figure 3
<p>TS and TSS. (<b>a</b>) is the description of TS and (<b>b</b>) is the depiction of TSS.</p> Full article ">Figure 4
<p>Trajectories optimization way. (<b>a</b>) is the tracking points of TS<sub>i</sub>; (<b>b</b>) shows the Delaunay triangulation network of TSS<sub>i</sub>.</p> Full article ">Figure 5
<p>The detection of trajectories strip width.</p> Full article ">Figure 6
<p>Trajectories strip width analysis.</p> Full article ">Figure 7
<p>Width detection results proprecessing.</p> Full article ">Figure 8
<p>Trajectories and trajectory vector description. (<b>a</b>) shows the trajectory vector constructed according to traditional style; (<b>b</b>) is the trajectory vector proposed in this paper.</p> Full article ">Figure 9
<p>Trajectory tracking. (<b>a</b>) indicates the trajectories with 40 s sampling interval; (<b>b</b>) shows the trajectories with 20 s sampling interval; (<b>c</b>) descripts the different driving directions of vehicles.</p> Full article ">Figure 10
<p>Intersection turns: left turn, right turn and U-turn detection.</p> Full article ">Figure 11
<p>Experimental data. (<b>a</b>) is the road network of the experimental area; (<b>b</b>) shows the raw trajectories collected by taxis.</p> Full article ">Figure 12
<p>Optimization results. The result of all experimental data (<b>left</b>); the magnification of one segment (<b>right</b>).</p> Full article ">Figure 13
<p>Optimization results evaluation: (<b>a</b>) shows the evaluation results of optimized trajectories; (<b>b</b>) indicates the evaluation results of non-optimized trajectories.</p> Full article ">Figure 14
<p>The overlay of image and trajectories. (<b>a</b>) shows the tracking results of one intersection; (<b>b</b>) indicates the tracking results of another intersection.</p> Full article ">
1325 KiB  
Article
An Improved PDR/Magnetometer/Floor Map Integration Algorithm for Ubiquitous Positioning Using the Adaptive Unscented Kalman Filter
by Jian Wang, Andong Hu, Xin Li and Yan Wang
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2638-2659; https://doi.org/10.3390/ijgi4042638 - 25 Nov 2015
Cited by 30 | Viewed by 6467
Abstract
In this paper, a scheme is presented for fusing a foot-mounted Inertial Measurement Unit (IMU) and a floor map to provide ubiquitous positioning in a number of settings, such as in a supermarket as a shopping guide, in a fire emergency service for [...] Read more.
In this paper, a scheme is presented for fusing a foot-mounted Inertial Measurement Unit (IMU) and a floor map to provide ubiquitous positioning in a number of settings, such as in a supermarket as a shopping guide, in a fire emergency service for navigation, or with a hospital patient to be tracked. First, several Zero-Velocity Detection (ZDET) algorithms are compared and discussed when used in the static detection of a pedestrian. By introducing information on the Zero Velocity of the pedestrian, fused with a magnetometer measurement, an improved Pedestrian Dead Reckoning (PDR) model is developed to constrain the accumulating errors associated with the PDR positioning. Second, a Correlation Matching Algorithm based on map projection (CMAP) is presented, and a zone division of a floor map is demonstrated for fusion of the PDR algorithm. Finally, in order to use the dynamic characteristics of a pedestrian’s trajectory, the Adaptive Unscented Kalman Filter (A-UKF) is applied to tightly integrate the IMU, magnetometers and floor map for ubiquitous positioning. The results of a field experiment performed on the fourth floor of the School of Environmental Science and Spatial Informatics (SESSI) building on the China University of Mining and Technology (CUMT) campus confirm that the proposed scheme can reliably achieve meter-level positioning. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Illustration of the two coordinate systems.</p> Full article ">Figure 2
<p>A test of heading fusion.</p> Full article ">Figure 3
<p>Flow chart of the improved PDR.</p> Full article ">Figure 4
<p>Demonstration of Correlation Matching Algorithm based on Projection (CMAP).</p> Full article ">Figure 5
<p>Zone division.</p> Full article ">Figure 6
<p>Accumulated gyroscope data.</p> Full article ">Figure 7
<p>The flow of map matching.</p> Full article ">Figure 8
<p>Flowchart of the filter process.</p> Full article ">Figure 9
<p>Test Route in the School of Environmental Science and Spatial Informatics (SESSI) Building.</p> Full article ">Figure 10
<p>The results of the test: (<b>a</b>) estimated position based on Unscented Kalman Filter (UKF) and A-UKF and (<b>b</b>) variance of the adaptive parameter.</p> Full article ">Figure 11
<p>The general flowchart.</p> Full article ">Figure 12
<p>Photos of the equipment and their usage: (<b>a</b>) the x-IMU and (<b>b</b>) foot-mount installation.</p> Full article ">Figure 13
<p>Experimental site: (<b>a</b>) floor map of the fourth floor and (<b>b</b>) 3D model of the experimental site.</p> Full article ">Figure 14
<p>Estimated position of the three schemes.</p> Full article ">Figure 15
<p>The accumulated error of the PDR algorithm.</p> Full article ">Figure 16
<p>The error step series of the three schemes.</p> Full article ">Figure 17
<p>The variance of the adaptive parameter.</p> Full article ">Figure 18
<p>Estimated position based on A-UKF and UKF.</p> Full article ">Figure 19
<p>Estimated position based on UKF and A-UKF.</p> Full article ">
627 KiB  
Article
Fast Inversion of Air-Coupled Spectral Analysis of Surface Wave (SASW) Using in situ Particle Displacement
by Yifeng Lu, Yinghong Cao, J. Gregory McDaniel and Ming L. Wang
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2619-2637; https://doi.org/10.3390/ijgi4042619 - 24 Nov 2015
Cited by 5 | Viewed by 6177
Abstract
Spectral Analysis of Surface Wave (SASW) is widely used in nondestructive subsurface profiling for geological sites. The air-coupled SASW is an extension from conventional SASW methods by replacing ground-mounted accelerometers with non-contact microphones, which acquire a leaky surface wave instead of ground vibration. [...] Read more.
Spectral Analysis of Surface Wave (SASW) is widely used in nondestructive subsurface profiling for geological sites. The air-coupled SASW is an extension from conventional SASW methods by replacing ground-mounted accelerometers with non-contact microphones, which acquire a leaky surface wave instead of ground vibration. The air-coupled SASW is a good candidate for fast inspection in shallow geological studies. Especially for pavement maintenance, minimum traffic interference might be induced. One issue that restrains SASW from fast inspection is the traditional slow inversion which relies on guess-and-check iteration techniques including a forward analysis. In this article, a fast inversion analysis algorithm is proposed to estimate the shear velocity profile without performing conventional forward simulation. By investigating the attenuation of particle displacement along penetrating depths, a weighted combination relationship is derived to connect the dispersion curve with the shear velocity profile directly. Using this relationship, the shear velocity profile could be estimated from a given/measured dispersion curve. The proposed procedure allows the surface wave-based method to be fully automatic and even operated in real-time for geological site and pavement assessment. The method is verified by the forward analysis with stiffness matrix method. It is also proved by comparing with other published results using various inversion methods. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Propagation of surface wave.</p> Full article ">Figure 2
<p>Typical air-coupled SASW/MASW configuration.</p> Full article ">Figure 3
<p>(<b>a</b>) Sample data collected by accelerometer data; and (<b>b</b>) Sample data collected by microphone.</p> Full article ">Figure 4
<p>Distribution of stress waves from a vertical load on a homogeneous elastic half-space (after Richart <span class="html-italic">et al.</span> [<a href="#B24-ijgi-04-02619" class="html-bibr">24</a>]).</p> Full article ">Figure 5
<p>Distribution of particle displacement along with penetrating depth of surface wave (after Richart <span class="html-italic">et al.</span> [<a href="#B24-ijgi-04-02619" class="html-bibr">24</a>]).</p> Full article ">Figure 6
<p>Relationship between shear velocity profile and dispersion curve for layered system; (<b>a</b>) Shear velocity profile; (<b>b</b>) Particle displacement; and (<b>c</b>) Dispersion curve.</p> Full article ">Figure 7
<p>Flowchart of fast inversion analysis; (<b>a</b>) Initializing stage; and (<b>b</b>) Adjusting stage.</p> Full article ">Figure 8
<p>Analytical dispersion curve (dominant mode).</p> Full article ">Figure 9
<p>Inverted shear velocity profile <span class="html-italic">vs.</span> original profile for ascending stiffness profile; (<b>a</b>) Inverted shear profile; and (<b>b</b>) Averaged shear profile.</p> Full article ">Figure 10
<p>Inverted shear velocity profile <span class="html-italic">vs.</span> original profile for descending stiffness profile; (<b>a</b>) Analytical dispersion curve; and (<b>b</b>) Inverted averaged shear profile.</p> Full article ">Figure 11
<p>Estimated shear velocity profiles using the proposed inversion algorithm; (<b>a</b>) Automatically generated shear velocity profile; and (<b>b</b>) Averaged shear velocity profile. (The vertical axis scale is the wavelength for the dispersion curve and depth for profile).</p> Full article ">Figure 12
<p>Estimated shear velocity profiles in comparison with the reported result by Joh [<a href="#B8-ijgi-04-02619" class="html-bibr">8</a>]. (The vertical axis scale is the wavelength for the dispersion curve and depth for profile).</p> Full article ">Figure 13
<p>Mobile testing at Northeastern University Campus.</p> Full article ">Figure 14
<p>Microphone data of parking lot test: (<b>a</b>) Raw acoustic data; and (<b>b</b>) Filtered radiating surface wave.</p> Full article ">Figure 15
<p>(<b>a</b>) Coherence function; and (<b>b</b>) Dispersion curve.</p> Full article ">Figure 16
<p>Estimated elastic modulus profile at parking lot test.</p> Full article ">
652 KiB  
Article
Finding Causes of Irregular Headways Integrating Data Mining and AHP
by Shi An, Xinming Zhang and Jian Wang
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2604-2618; https://doi.org/10.3390/ijgi4042604 - 24 Nov 2015
Cited by 7 | Viewed by 4237
Abstract
Irregular headways could reduce the public transit service level heavily. Finding out the exact causes of irregular headways will greatly help to develop efficient strategies aiming to improve transit service quality. This paper utilizes bus GPS data of Harbin to evaluate the headway [...] Read more.
Irregular headways could reduce the public transit service level heavily. Finding out the exact causes of irregular headways will greatly help to develop efficient strategies aiming to improve transit service quality. This paper utilizes bus GPS data of Harbin to evaluate the headway performance and proposes a statistical method to identify the abnormal headways. Association mining is used to dig deeper and recognize six causes of bus bunching. The AHP, embedded data analysis, is applied to determine the weight of each cause in the case of that these causes are combined with each other constantly. Results show that the front bus has a greater effect on bus bunching than the following bus, and the traffic condition is the most critical factor affecting bus headway. Full article
(This article belongs to the Special Issue Advances in Spatio-Temporal Data Analysis and Mining)
►▼ Show Figures

Figure 1

Figure 1
<p>The route map of Line 104 in Harbin.</p> Full article ">Figure 2
<p>Temporal and spatial distributions of the headway service level of Route 104.</p> Full article ">Figure 3
<p>Abnormal headway identification on stop No.15 of Route 104.</p> Full article ">Figure 4
<p>(<b>a</b>) Distributions of too short headways of Route 104; and (<b>b</b>) distributions of too long headways of Route 104.</p> Full article ">
1234 KiB  
Article
Pitch and Flat Roof Factors’ Association with Spatiotemporal Patterns of Dengue Disease Analysed Using Pan-Sharpened Worldview 2 Imagery
by Fedri Ruluwedrata Rinawan, Ryutaro Tateishi, Ardini Saptaningsih Raksanagara, Dwi Agustian, Bayan Alsaaideh, Yessika Adelwin Natalia and Ahyani Raksanagara
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2586-2603; https://doi.org/10.3390/ijgi4042586 - 23 Nov 2015
Cited by 3 | Viewed by 6752
Abstract
Dengue disease incidence is related with the construction of a house roof, which is an Aedes mosquito habitat. This study was conducted to classify pitch roof (PR) and flat roof (FR) surfaces using pan-sharpened Worldview 2 to identify dengue disease patterns (DDPs) and [...] Read more.
Dengue disease incidence is related with the construction of a house roof, which is an Aedes mosquito habitat. This study was conducted to classify pitch roof (PR) and flat roof (FR) surfaces using pan-sharpened Worldview 2 to identify dengue disease patterns (DDPs) and their association with DDP. A Supervised Minimum Distance classifier was applied to 653 training data from image object segmentations: PR (81 polygons), FR (50), and non-roof (NR) class (522). Ground validation of 272 pixels (52 for PR, 51 for FR, and 169 for NR) was done using a global positioning system (GPS) tool. Getis-Ord score pattern analysis was applied to 1154 dengue disease incidence with address-approach-based data with weighted temporal value of 28 days within a 1194 m spatial radius. We used ordinary least squares (OLS) and geographically weighted regression (GWR) to assess spatial association. Our findings showed 70.59% overall accuracy with a 0.51 Kappa coefficient of the roof classification images. Results show that DDPs were found in hotspot, random, and dispersed patterns. Smaller PR size and larger FR size showed some association with increasing DDP into more clusters (OLS: PR value = −0.27; FR = 0.04; R2 = 0.076; GWR: R2 = 0.76). The associations in hotspot patterns are stronger than in other patterns (GWR: R2 in hotspot = 0.39, random = 0.37, dispersed = 0.23). Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>(<b>a</b>) Map of Indonesia showing the approximate location of northern Bandung city; (<b>b</b>) Bandung city and the white border of study area; (<b>c</b>) Pan-sharpened Worldview 2 satellite image, red, green, blue (RGB) color composite showing the area depicted in high resolution satellite imagery.</p> Full article ">Figure 2
<p>Data preparation and analysis procedures.</p> Full article ">Figure 3
<p>(<b>a</b>) RGB color composite of pan-sharpened Worldview 2 and (<b>b</b>) segmentation.</p> Full article ">Figure 4
<p>(<b>a</b>) RGB color composite of pan-sharpened 2 and (<b>b</b>) PR, FR, and NR class image.</p> Full article ">Figure 5
<p>Dengue disease patterns.</p> Full article ">Figure 6
<p>(<b>a</b>) Scatter plot of PR with GiZ score and (<b>b</b>) scatter plot of FR with GiZ score.</p> Full article ">Figure 7
<p>(<b>a</b>) Standardized residuals in DDP; (<b>b</b>) standardized residuals are shown within hotspot, random, and dispersed patterns; and (<b>c</b>) closer examination of the residuals on roofs.</p> Full article ">
96 KiB  
Editorial
GIS for Sustainable Urban Transport
by Mark H.P. Zuidgeest, Mark J.G. Brussel and Martin F.A.M. Van Maarseveen
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2583-2585; https://doi.org/10.3390/ijgi4042583 - 23 Nov 2015
Cited by 7 | Viewed by 5007
Abstract
The world is urbanizing at a very fast pace. Modern geography, particularly geo-information systems (GIS) and global positioning systems (GPS) are reshaping the way urban and transport planners are collecting, exploring, synthesizing, analyzing, evaluating and presenting their data. [...] Full article
(This article belongs to the Special Issue GIS for Sustainable Urban Transport)
1599 KiB  
Article
Data Integration for Climate Vulnerability Mapping in West Africa
by Alex De Sherbinin, Tricia Chai-Onn, Malanding Jaiteh, Valentina Mara, Linda Pistolesi, Emilie Schnarr and Sylwia Trzaska
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2561-2582; https://doi.org/10.3390/ijgi4042561 - 19 Nov 2015
Cited by 19 | Viewed by 9678
Abstract
Vulnerability mapping reveals areas that are likely to be at greater risk of climate-related disasters in the future. Through integration of climate, biophysical, and socioeconomic data in an overall vulnerability framework, so-called “hotspots” of vulnerability can be identified. These maps can be used [...] Read more.
Vulnerability mapping reveals areas that are likely to be at greater risk of climate-related disasters in the future. Through integration of climate, biophysical, and socioeconomic data in an overall vulnerability framework, so-called “hotspots” of vulnerability can be identified. These maps can be used as an aid to targeting adaptation and disaster risk management interventions. This paper reviews vulnerability mapping efforts in West Africa conducted under the USAID-funded African and Latin American Resilience to Climate Change (ARCC) project. The focus is on the integration of remotely sensed and socioeconomic data. Data inputs included a range of sensor data (e.g., MODIS NDVI, Landsat, SRTM elevation, DMSP-OLS night-time lights) as well as high-resolution poverty, conflict, and infrastructure data. Two basic methods were used, one in which each layer was transformed into standardized indicators in an additive approach, and another in which remote sensing data were used to contextualize the results of composite indicators. We assess the benefits and challenges of data integration, and the lessons learned from these mapping exercises. Full article
(This article belongs to the Special Issue Geoinformation for Disaster Risk Management)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Vulnerability mapping processing flow chart.</p> Full article ">Figure 2
<p>Interannual Coefficient of Variation in Greenness (NDVI)—Derived from GIMMS, raw data (<b>left</b>) and transformed (<b>right</b>).</p> Full article ">Figure 3
<p>Flood Extent (events per 100 years)—Derived in part from MODIS flood extent data at the Dartmouth Flood Observatory, raw data (<b>left</b>) and transformed (<b>right</b>).</p> Full article ">Figure 4
<p>Soil Carbon Partly Derived from MODIS data raw data (<b>left</b>) and transformed (<b>right</b>).</p> Full article ">Figure 5
<p>Anthropogenic Biomes raw data (<b>left</b>) and transformed (<b>right</b>).</p> Full article ">Figure 6
<p>Comparison of SRTM with ACE2 in Coastal Benin.</p> Full article ">Figure 7
<p>Urban areas of Cotonou, Benin and Lagos, Nigeria in comparison to the LECZ.</p> Full article ">Figure 8
<p>Mangroves and the LECZ.</p> Full article ">Figure 9
<p>Landsat-scale deforestation data aggregated to one square kilometer pixels.</p> Full article ">Figure 10
<p>Mali vulnerability mapping: Components of vulnerability rolled up into an overall vulnerability index. Note: Northern portions of Mali were excluded owing to low population densities.</p> Full article ">Figure 11
<p>SVI in relation to the West Africa LECZ.</p> Full article ">
2463 KiB  
Article
Assessing the Effect of Temporal Interval Length on the Blending of Landsat-MODIS Surface Reflectance for Different Land Cover Types in Southwestern Continental United States
by Dongjie Fu, Lifu Zhang, Hao Chen, Juan Wang, Xuejian Sun and Taixia Wu
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2542-2560; https://doi.org/10.3390/ijgi4042542 - 17 Nov 2015
Cited by 7 | Viewed by 4903
Abstract
Capturing spatial and temporal dynamics is a key issue for many remote-sensing based applications. Consequently, several image-blending algorithms that can simulate the surface reflectance with high spatial-temporal resolution have been developed recently. However, the performance of the algorithm against the effect of temporal [...] Read more.
Capturing spatial and temporal dynamics is a key issue for many remote-sensing based applications. Consequently, several image-blending algorithms that can simulate the surface reflectance with high spatial-temporal resolution have been developed recently. However, the performance of the algorithm against the effect of temporal interval length between the base and simulation dates has not been reported. In this study, our aim was to evaluate the effect of different temporal interval lengths on the accuracy using the widely used blending algorithm, Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM), based on Landsat, Moderate-resolution Imaging Spectroradiometer (MODIS) images and National Land Cover Database (NLCD). Taking the southwestern continental United States as the study area, a series of experiments was conducted using two schemes, which were the assessment of STARFM with (i) a fixed base date and varied simulation date and (ii) varied base date and specific simulation date, respectively. The result showed that the coefficient of determination (R2), Root Mean Squared Error (RMSE) varied, and overall trend of R2 decreased along with the increasing temporal interval between the base and simulation dates for six land cover types. The mean R2 value of cropland was lowest, whereas shrub had the highest value for two schemes. The result may facilitate selection of an appropriate temporal interval when using STARFM. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Land cover map (National Land Cover Database (NLCD), year: 2001) of the study area.</p> Full article ">Figure 2
<p>Distribution of the Landsat data used in the study. (<b>a</b>) day of year (DOY), (<b>b</b>) frequency from the period 2001 to 2012. L5 denotes Landsat TM data, L7 denotes Landsat ETM+ data.</p> Full article ">Figure 3
<p>Distribution of the day of year (DOY) for maximum R<sup>2</sup> (not filled marker) and minimum RMSE (filled marker) value at each year from 2001 to 2012. The corresponding base date (dash line) is (<b>a</b>) 9 February (DOY: 40), (<b>b</b>) 16 May (DOY: 136), (<b>c</b>) 28 August (DOY: 240) and (<b>d</b>) 8 November (DOY: 312) for year 2001, respectively. The markers for six Landsat-like bands are circle, triangle, diamond, square, hexagram and pentagram, respectively. The color for each land cover type is black (WAT: water), red (URB: urban), green (SHR: shrub), blue (GRA, grassland), violet (CRO: cropland) and brown (WET: wetland). The meaning of marker and color for symbols is the same for <a href="#ijgi-04-02542-f004" class="html-fig">Figure 4</a>.</p> Full article ">Figure 4
<p>Distribution of the day of year (DOY) for minimum R<sup>2</sup> (not filled marker) and maximum RMSE (filled marker) value at each year from 2001 to 2012. The corresponding base date (dash line) is (<b>a</b>) 9 February (DOY: 40), (<b>b</b>) 16 May (DOY: 136), (<b>c</b>) 28 August (DOY: 240) and (<b>d</b>) 8 November (DOY: 312) for year 2001, respectively.</p> Full article ">Figure 5
<p>Distribution of the day of year (DOY) for maximum R<sup>2</sup> (not filled marker) and minimum RMSE (filled marker) value for year 2001, 2003 and 2008. The corresponding simulation date (dash line) is (<b>a</b>) 14 January (DOY: 14), (<b>b</b>) 12 April (DOY: 102), (<b>c</b>) 9 July (DOY: 190) and (<b>d</b>) 21 October (DOY: 294) for year 2009, respectively. The size of marker from small to big denotes years 2001, 2003 and 2008, respectively. The markers for six Landsat-like bands are circle, triangle, diamond, square, hexagram and pentagram, respectively. The color for each land cover type is black (WAT: water), red (URB: urban), green (SHR: shrub), blue (GRA, grassland), violet (CRO: cropland) and brown (WET: wetland). The meaning of marker and color for symbols is the same for <a href="#ijgi-04-02542-f006" class="html-fig">Figure 6</a>.</p> Full article ">Figure 6
<p>Distribution of the day of year (DOY) for minimum R<sup>2</sup> (not filled marker) and maximum RMSE (filled marker) value for year 2001, 2003 and 2008. The corresponding simulation date (dash line) is (<b>a</b>) 14 January (DOY: 14), (<b>b</b>) 12 April (DOY: 102), (<b>c</b>) 9 July (DOY: 190) and (<b>d</b>) 21 October (DOY: 294) for year 2009, respectively. The size of marker from small to big denotes year 2001, 2003 and 2008 respectively.</p> Full article ">Figure 7
<p>Visual comparison between the observed and simulated Landsat reflectance (Near-infrared (NIR)-red-green composite) using the base Landsat-MODIS pair data at 9 February 2001. The upper/lower row is the observed/simulated Landsat reflectance.</p> Full article ">Figure 8
<p>Spatial distribution of land cover changes for the study area for (<b>a</b>) 2001 to 2006, (<b>b</b>) 2006 to 2011, and (<b>c</b>) 2001 to 2011. The base map is band 4 (near-infrared band) of Landsat data on 8 May 2001. The land cover shown on the map is the latest land cover type.</p> Full article ">
1054 KiB  
Article
Evaluation of the Consistency of MODIS Land Cover Product (MCD12Q1) Based on Chinese 30 m GlobeLand30 Datasets: A Case Study in Anhui Province, China
by Dong Liang, Yan Zuo, Linsheng Huang, Jinling Zhao, Ling Teng and Fan Yang
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2519-2541; https://doi.org/10.3390/ijgi4042519 - 16 Nov 2015
Cited by 88 | Viewed by 9861
Abstract
Land cover plays an important role in the climate and biogeochemistry of the Earth system. It is of great significance to produce and evaluate the global land cover (GLC) data when applying the data to the practice at a specific spatial scale. The [...] Read more.
Land cover plays an important role in the climate and biogeochemistry of the Earth system. It is of great significance to produce and evaluate the global land cover (GLC) data when applying the data to the practice at a specific spatial scale. The objective of this study is to evaluate and validate the consistency of the Moderate Resolution Imaging Spectroradiometer (MODIS) land cover product (MCD12Q1) at a provincial scale (Anhui Province, China) based on the Chinese 30 m GLC product (GlobeLand30). A harmonization method is firstly used to reclassify the land cover types between five classification schemes (International Geosphere Biosphere Programme (IGBP) global vegetation classification, University of Maryland (UMD), MODIS-derived Leaf Area Index and Fractional Photosynthetically Active Radiation (LAI/FPAR), MODIS-derived Net Primary Production (NPP), and Plant Functional Type (PFT)) of MCD12Q1 and ten classes of GlobeLand30, based on the knowledge rule (KR) and C4.5 decision tree (DT) classification algorithm. A total of five harmonized land cover types are derived including woodland, grassland, cropland, wetland and artificial surfaces, and four evaluation indicators are selected including the area consistency, spatial consistency, classification accuracy and landscape diversity in the three sub-regions of Wanbei, Wanzhong and Wannan. The results indicate that the consistency of IGBP is the best among the five schemes of MCD12Q1 according to the correlation coefficient (R). The “woodland” LAI/FPAR is the worst, with a spatial similarity (O) of 58.17% due to the misclassification between “woodland” and “others”. The consistency of NPP is the worst among the five schemes as the agreement varied from 1.61% to 56.23% in the three sub-regions. Furthermore, with the biggest difference of diversity indices between LAI/FPAR and GlobeLand30, the consistency of LAI/FPAR is the weakest. This study provides a methodological reference for evaluating the consistency of different GLC products derived from multi-source and multi-resolution remote sensing datasets on various spatial scales. Full article
(This article belongs to the Special Issue Advances in Spatio-Temporal Data Analysis and Mining)
►▼ Show Figures

Figure 1

Figure 1
<p>Location of Anhui Province and the Wanbei, Wanzhong and Wannan regions.</p> Full article ">Figure 2
<p>Conceptual diagram and the general workflow.</p> Full article ">Figure 3
<p>The flow chart of harmonizing land-cover classification based on KR.</p> Full article ">Figure 4
<p>Comparison of percentage disagreement of the five schemes in (<b>a</b>) Wanbei, (<b>b</b>) Wanzhong, and (<b>c</b>) Wannan.</p> Full article ">Figure 5
<p>Analysis of spatial consistency of “woodland” under the (<b>a</b>) IGBP, (<b>b</b>) UMD, (<b>c</b>) LAI/FPAR, (<b>d</b>) NPP and (<b>e</b>) PFT schemes using the MCD12Q1 and GlobeLand30 data.</p> Full article ">Figure 6
<p>Comparison of landscape diversity indices of <span class="html-italic">MSIDI</span> and <span class="html-italic">MSIEI</span> in (<b>a</b>) Wanbei, (<b>b</b>) Wanzhong and (<b>c</b>) Wannan.</p> Full article ">
2213 KiB  
Article
Impacts of Species Misidentification on Species Distribution Modeling with Presence-Only Data
by Hugo Costa, Giles M. Foody, Sílvia Jiménez and Luís Silva
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2496-2518; https://doi.org/10.3390/ijgi4042496 - 16 Nov 2015
Cited by 40 | Viewed by 7482
Abstract
Spatial records of species are commonly misidentified, which can change the predicted distribution of a species obtained from a species distribution model (SDM). Experiments were undertaken to predict the distribution of real and simulated species using MaxEnt and presence-only data “contaminated” with varying [...] Read more.
Spatial records of species are commonly misidentified, which can change the predicted distribution of a species obtained from a species distribution model (SDM). Experiments were undertaken to predict the distribution of real and simulated species using MaxEnt and presence-only data “contaminated” with varying rates of misidentification error. Additionally, the difference between the niche of the target and contaminating species was varied. The results show that species misidentification errors may act to contract or expand the predicted distribution of a species while shifting the predicted distribution towards that of the contaminating species. Furthermore the magnitude of the effects was positively related to the ecological distance between the species’ niches and the size of the error rates. Critically, the magnitude of the effects was substantial even when using small error rates, smaller than common average rates reported in the literature, which may go unnoticed while using a standard evaluation method, such as the area under the receiver operating characteristic curve. Finally, the effects outlined were shown to impact negatively on practical applications that use SDMs to identify priority areas, commonly selected for various purposes such as management. The results highlight that species misidentification should not be neglected in species distribution modeling. Full article
(This article belongs to the Special Issue Spatial Ecology)
►▼ Show Figures

Figure 1

Figure 1
<p>Study area and species data used: (<b>a</b>) island of São Miguel in the Azores, Portugal; and (<b>b</b>) location of the <span class="html-italic">Cyathea cooperi</span> (orange squares) and <span class="html-italic">C. medullaris</span> (blue dots) presence recorded in São Miguel between September 2011 and May 2012. Note: grey background represents the island relief as bright tones for high altitude and insolation and dark tones for opposite conditions.</p> Full article ">Figure 2
<p>The ecological niche of the simulated target and contaminating species in five scenarios (ellipses show the 95% probability regions). The niche of the target species was defined as the multiplicative interaction of precipitation (P) and temperature (T). The response of the target species to P and T was defined using normal curves where the mean μ and standard deviation σ of P (μ<sub>P</sub>, σ<sub>P</sub>) and T (μ<sub>T</sub>, σ<sub>T</sub>) are 314.77 and 162.90, and 21.83 and 1.79, respectively. For the simulated contaminating species, μ and σ of P and T varied in standard deviation units in relation to the simulated target species as shown in each scenario (for example, the normal curve of P in scenario III used a mean value of μ<sub>P</sub> + σ<sub>P</sub> and a standard deviation of 2σ<sub>P</sub>).</p> Full article ">Figure 3
<p>Gaussian kernel density estimates of the MaxEnt predictions: (<b>a</b>) real data; (<b>b</b>) simulated data in scenario I; (<b>c</b>) simulated data in scenario II; (<b>d</b>) simulated data in scenario III; (<b>e</b>) simulated data in scenario IV; and (<b>f</b>) simulated data in scenario V.</p> Full article ">Figure 4
<p>Similarity (Schoener’s <span class="html-italic">D</span>) between the MaxEnt outputs produced with and without contaminated data: (<b>a</b>) comparison between the contaminated outputs (1% ≤ <span class="html-italic">r</span> ≤ 32%) and those produced using the gold standard data set (<span class="html-italic">r</span> = 0%); and (<b>b</b>) comparison between the contaminated outputs (1% ≤ <span class="html-italic">r</span> ≤ 32%) and those produced using the contaminating species (<span class="html-italic">r</span> = 100%).</p> Full article ">Figure 5
<p>Some examples of predictions of presence of the real and simulated species produced by MaxEnt: (<b>a</b>) predictions of presence of <span class="html-italic">C. cooperi</span> produced using the gold standard data set (<span class="html-italic">r</span> = 0%); (<b>b</b>) predictions of presence of <span class="html-italic">C. cooperi</span> produced using contaminated data (<span class="html-italic">r</span> = 16%); (<b>c</b>) predictions of presence of the contaminating species <span class="html-italic">C. medullaris</span> (<span class="html-italic">r</span> = 100%); (<b>d</b>) predictions of presence of the simulated target species using the gold standard data set (<span class="html-italic">r</span> = 0%); (<b>e</b>) predictions of presence of the simulated target species using contaminated data in scenario III (<span class="html-italic">r</span> = 16%); and (<b>f</b>) predictions of presence of the simulated contaminating species in scenario III (<span class="html-italic">r</span> = 100%); Notes: greyscale from black (high prediction values) to white (low prediction values). Black arrows in parts (b) and (c) highlight an area where the influence of the distribution of the contaminating species over that of the target species is particularly noted when using contaminated data.</p> Full article ">Figure 6
<p>Magnitude of the effects caused by species mis-identification on the MaxEnt predictions: (<b>a</b>) proportion of the predictions that differed significantly at the various misidentification rates for the real and simulated species; (<b>b</b>) location of the predictions that changed significantly (in black) for <span class="html-italic">C. cooperi</span> using the misidentification rate of 1%; (<b>c</b>) Location of the predictions that changed significantly (in black) for <span class="html-italic">C. cooperi</span> using the misidentification rate of 4%; and (<b>d</b>) location of the predictions that changed significantly (in black) for <span class="html-italic">C. cooperi</span> using the misidentification rate 16%.</p> Full article ">Figure 7
<p>Omission and commission errors committed by MaxEnt while defining priority areas using contaminated data: (<b>a</b>) omission or commission error committed at the various misidentification rates for the real and simulated species; (<b>b</b>) location of the omission and commission errors committed for <span class="html-italic">C. cooperi</span> using the misidentification rate of 1%; (<b>c</b>) location of the omission and commission errors committed for <span class="html-italic">C. cooperi</span> using the mis-identification rate of 4%; and (<b>d</b>) location of the omission and commission errors committed for <span class="html-italic">C. cooperi</span> using the misidentification rate of 16%. Note: The priority areas that were correctly defined using contaminated data are shown in black in parts (b), (c) and (d).</p> Full article ">Figure 8
<p>AUC values of the MaxEnt outputs produced using the simulated data as a function of the misidentification error rate.</p> Full article ">
1566 KiB  
Article
Towards a Standard Plant Species Spectral Library Protocol for Vegetation Mapping: A Case Study in the Shrubland of Doñana National Park
by Marcos Jiménez and Ricardo Díaz-Delgado
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2472-2495; https://doi.org/10.3390/ijgi4042472 - 16 Nov 2015
Cited by 32 | Viewed by 7389
Abstract
One of the main applications of field spectroscopy is the generation of spectral libraries of Earth’s surfaces or materials to support mapping activities using imaging spectroscopy. To enhance the reliability of these libraries, spectral signature acquisition should be carried out following standard procedures [...] Read more.
One of the main applications of field spectroscopy is the generation of spectral libraries of Earth’s surfaces or materials to support mapping activities using imaging spectroscopy. To enhance the reliability of these libraries, spectral signature acquisition should be carried out following standard procedures and controlled experimental approaches. This paper presents a standard protocol for the creation of a spectral library for plant species. The protocol is based on characterizing the reflectance spectral response of different species in the spatiotemporal domain, by accounting for intra-species variation and inter-species similarity. A practical case study was conducted on the shrubland located in Doñana National Park (SW Spain). Spectral libraries of the five dominant shrub species were built (Erica scoparia, Halimium halimifolium, Ulex australis, Rosmarinus officinalis, and Stauracanthus genistoides). An estimation was made of the separability between species: on one hand, the Student’s t-test evaluates significant intra-species variability (p < 0.05) and on the other hand, spectral similarity value (SSV) and spectral angle mapper (SAM) algorithms obtain significant separability values for dominant species, although it was not possible to discriminate the legume species Ulex australis and Stauracanthus genistoides. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Field spectroscopy protocol for plant spectral library collection. The complete procedure and corresponding metadata are indicated for the sampling protocol and data processing.</p> Full article ">Figure 2
<p>Location of Doñana National Park (SW, Spain). Square indicates the stabilized sand dunes ecosystem of Doñana Biological Reserve. The sub-image shows a digital elevation model of the study area.</p> Full article ">Figure 3
<p>Site location for plant spectral library measurements. Yellow tacks identify spectral signatures acquisition sites and red tacks identify spectral signatures and LAI measurements sites.</p> Full article ">Figure 4
<p>Spectral library of the five dominant shrub species in Doñana National Park. The graphics show the average and standard deviation spectral reflectance for dry and wet season. PAI and CV values are shown for every species spectral response in dry season.</p> Full article ">Figure 5
<p>Intra-species variability for the DNP shrubland dominant species. t-test for species-to-species comparison for dry and wet season separately. The red line in the graphics identifies t-critical values according to the corresponding degrees of freedom.</p> Full article ">Figure 6
<p>Matrix of spectral similarity indexes in grayscale, where black indicates total similarity and white the highest dissimilarity for the dominant and the less abundant shrub species in Doñana National Park. Foliage type is also shown by coloring the species name. Species names are also underlined according to the shrubland community they belong to.</p> Full article ">
4509 KiB  
Article
Inferring Directed Road Networks from GPS Traces by Track Alignment
by Xingzhe Xie, Kevin Bing-YungWong, Hamid Aghajan, Peter Veelaert and Wilfried Philips
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2446-2471; https://doi.org/10.3390/ijgi4042446 - 11 Nov 2015
Cited by 32 | Viewed by 5769
Abstract
This paper proposes a method to infer road networks from GPS traces. These networks include intersections between roads, the connectivity between the intersections and the possible traffic directions between directly-connected intersections. These intersections are localized by detecting and clustering turning points, which are [...] Read more.
This paper proposes a method to infer road networks from GPS traces. These networks include intersections between roads, the connectivity between the intersections and the possible traffic directions between directly-connected intersections. These intersections are localized by detecting and clustering turning points, which are locations where the moving direction changes on GPS traces. We infer the structure of road networks by segmenting all of the GPS traces to identify these intersections. We can then form both a connectivity matrix of the intersections and a small representative GPS track for each road segment. The road segment between each pair of directly-connected intersections is represented using a series of geographical locations, which are averaged from all of the tracks on this road segment by aligning them using the dynamic time warping (DTW) algorithm. Our contribution is two-fold. First, we detect potential intersections by clustering the turning points on the GPS traces. Second, we infer the geometry of the road segments between intersections by aligning GPS tracks point by point using a “stretch and then compress” strategy based on the DTW algorithm. This approach not only allows road estimation by averaging the aligned tracks, but also a deeper statistical analysis based on the individual track’s time alignment, for example the variance of speed along a road segment. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of the proposed method.</p> Full article ">Figure 2
<p>An example of a turn. A part of a GPS trace around one intersection is represented by a black line with circles indicating the position of data recordings and arrows indicating the moving direction at each point.</p> Full article ">Figure 3
<p>Bend detection. Bends are detected by checking the entering and exiting directions.</p> Full article ">Figure 4
<p>Example of two-track alignment. (<b>a</b>) The elements in two tracks are paired using DTW with the distance between these elements as features. (<b>b</b>) the warp path with the horizontal and the vertical axis indicates the time index of Track <math display="inline"> <msup> <mrow> <mtext mathvariant="bold">r</mtext> </mrow> <mrow> <mo>(</mo> <mn>0</mn> <mo>)</mo> </mrow> </msup> </math> and Track <math display="inline"> <msup> <mrow> <mtext mathvariant="bold">r</mtext> </mrow> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msup> </math> separately.</p> Full article ">Figure 5
<p>Multiple-track alignment. In Block 1, the tracks are aligned one by one in random order using a “stretch and then compress”, producing intermediate paths; Block 2 depicts how to find the normalized paths, which are used to align all of the tracks with one-to-one correspondence among the points using intermediate paths produced in Block 1.</p> Full article ">Figure 6
<p>Example of four track alignment. (<b>a</b>) shows the data points of the tracks along time. In Sub-figure 2, Tracks <math display="inline"> <msup> <mrow> <mtext mathvariant="bold">r</mtext> </mrow> <mrow> <mo>(</mo> <mn>0</mn> <mo>)</mo> </mrow> </msup> </math> and <math display="inline"> <msup> <mrow> <mtext mathvariant="bold">r</mtext> </mrow> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msup> </math> are stretched and compressed, resulting in <math display="inline"> <msubsup> <mrow> <mtext mathvariant="bold">r</mtext> </mrow> <mi>c</mi> <mrow> <mo>(</mo> <mn>0</mn> <mo>)</mo> </mrow> </msubsup> </math> and <math display="inline"> <msubsup> <mrow> <mtext mathvariant="bold">r</mtext> </mrow> <mi>c</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> </math>, with black lines indicating their association. A once-compressed track and a new track are input to the “stretch and then compress” module, so as to find the one-to-one correspondence among the tracks in Sub-figures 3 and 4. Sub-figure 5 depicts how to find the normalized path by the connecting lines. (<b>b</b>) shows how the data points of normalized tracks are selected from the original tracks along the normalized path and connected using black lines.</p> Full article ">Figure 7
<p>Example of a single road. (<b>a</b>) shows a generated road segment and its ground truth as an example. (<b>b</b>) shows them after interpolation, so as to reduce the error of their distance caused by the uneven distribution of the data points on the road.</p> Full article ">Figure 8
<p>Intersection extraction. Intersections in red circles are detected from the turning points in green dots, where the shuttles change their moving directions.</p> Full article ">Figure 9
<p>Ground truth map. The directed ground truth map is manually made through reducing the OpenStreetMap of this area to the streets traversed by the shuttles.</p> Full article ">Figure 10
<p>Directed roads. (<b>a</b>) shows part of the tracks, which are averaged to produce the directed roads in (<b>b</b>) ; (<b>c</b>) shows the other part of the tracks, which are averaged to produce the directed roads in (<b>d</b>).</p> Full article ">Figure 11
<p>Statistic analysis. The speed and the speed variation along the road segment from Intersection <math display="inline"> <msub> <mtext mathvariant="bold">q</mtext> <mn>3</mn> </msub> </math> to <math display="inline"> <msub> <mtext mathvariant="bold">q</mtext> <mn>1</mn> </msub> </math> are depicted.</p> Full article ">Figure 12
<p>Topological accuracy evaluation. (<b>a</b>) shows the accuracy of the generated intersections; (<b>b</b>) shows the accuracy of the connectivity between the intersections. With different matching distances between the generated intersections and their ground truth. The bigger the F-score is, the more accurate its is.</p> Full article ">Figure 13
<p>Influence of the aligning order. No matter in which order the tracks are aligned, the average distance between all of the generated roads and their ground truth is around <math display="inline"> <mrow> <mn>7</mn> <mo>.</mo> <mn>37</mn> </mrow> </math> m stably, which means the aligning order does not affect the track alignment significantly.</p> Full article ">Figure 14
<p>Comparison with other methods. (<b>a</b>) shows the some high-error GPS traces, which are used to extract the directed roads in (<b>b</b>); the traces in (<b>c</b>) have relatively low error, resulting in the directed roads in (<b>d</b>) .</p> Full article ">
1256 KiB  
Article
Optimized Route Selection Method based on the Turns of Road Intersections: A Case Study on Oversized Cargo Transportation
by Lingkui Meng, Zhenghua Hu, Changqing Huang, Wen Zhang and Tao Jia
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2428-2445; https://doi.org/10.3390/ijgi4042428 - 6 Nov 2015
Cited by 7 | Viewed by 5197
Abstract
For oversized cargo transportation, traditional transportation schemes only consider road length, road width, the transportation cost as weight values in analysis and calculation of route selection. However, for oversized trucks, turning direction at road intersections is also a factor worth considering. By introducing [...] Read more.
For oversized cargo transportation, traditional transportation schemes only consider road length, road width, the transportation cost as weight values in analysis and calculation of route selection. However, for oversized trucks, turning direction at road intersections is also a factor worth considering. By introducing the classical algorithm of Dijkstra into the model of road network, this research considers the size of turning angle at intersections as the weight value of the edge in the auxiliary network based on the weight values of road corners, upon which the shortest path analysis is performed. Then, an optimal path with minimum time cost was eventually obtained. The proposed algorithm was analyzed and compared with the traditional shortest path algorithm and it reported that our method could reduce the time for oversized trucks to pass through intersections. In addition, the proposed algorithm could be adapted to the complex and diverse road networks and provide a reliable scheme for route selection of oversized trucks. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Model of the dual graph.</p> Full article ">Figure 2
<p>Construction of the auxiliary network based on road corners.</p> Full article ">Figure 3
<p>The turning model at road intersections.</p> Full article ">Figure 4
<p>The calculation of the value of turn angles: (<b>a</b>) case a, (<b>b</b>) case b (<b>c</b>) case c and (<b>d</b>) case d.</p> Full article ">Figure 5
<p>Calculation of the time consumed by making turns: (<b>a</b>) test area and (<b>b</b>) the definition of the time consumed by oversized trucks to make turns.</p> Full article ">Figure 6
<p>Fitting result of turning delay at road intersections: (<b>a</b>) delay of oversized trucks when making left turns at intersections; and (<b>b</b>) delay of oversized trucks when making right turns at intersections.</p> Full article ">Figure 7
<p>Basic flow of algorithm</p> Full article ">Figure 8
<p>Comparison of the results of the traditional algorithm and the one based on the weighted model of corners: (<b>a</b>) road map of a city in Qingdao; (<b>b</b>) network based on the weight values of the corners at road intersections; (<b>c</b>) superimposed renderings; (<b>d</b>) result of the shortest path analysis with the weighted model of corners; and (<b>e</b>) result of the existing traditional algorithm.</p> Full article ">Figure 9
<p>Total time spent by oversized trucks to pass through different routes with the traditional shortest path algorithm and the one based on the size of the corners.</p> Full article ">Figure 10
<p>Statistics of the number of turns made at road intersections using the shortest path algorithm based on the size of the corner.</p> Full article ">Figure 11
<p>Statistics of the number of turns made at road intersections based on the traditional shortest path algorithm.</p> Full article ">
2195 KiB  
Article
Improving Post-Earthquake Insurance Claim Management: A Novel Approach to Prioritize Geospatial Data Collection
by Massimiliano Pittore, Marc Wieland, Mustafa Errize, Cagatay Kariptas and Ismet Güngör
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2401-2427; https://doi.org/10.3390/ijgi4042401 - 30 Oct 2015
Cited by 6 | Viewed by 6028
Abstract
With a population exceeding 14 million and a GDP of more than 300 billion USD, Istanbul dominates the Turkish economy. Unfortunately, this concentration of social and economic assets is permanently threatened by potentially devastating earthquakes, given the city’s close proximity to several well-known [...] Read more.
With a population exceeding 14 million and a GDP of more than 300 billion USD, Istanbul dominates the Turkish economy. Unfortunately, this concentration of social and economic assets is permanently threatened by potentially devastating earthquakes, given the city’s close proximity to several well-known fault systems. As a measure to mitigate the consequences of such events, and to increase the resilience of the exposed communities, the Turkish Catastrophe Insurance Pool (TCIP) has been set up to provide affordable and reliable earthquake insurance to households all over the country. In the aftermath of a damaging event, especially in Istanbul, the operational capacity of TCIP will be seriously challenged by the high number of claims whose settlement would have to be swift and fair in order to kick-start the recovery process. In this paper we explore an integrated approach based on mobile mapping and ad hoc prioritization techniques to streamline the data collection and analysis process, with application to both the pre-event and post-event phases. Preliminary results obtained in Besiktas, a populous district of Istanbul, are presented and discussed. Full article
(This article belongs to the Special Issue Geoinformation for Disaster Risk Management)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Overview of the study area Besiktas, Istanbul, Turkey (<b>left</b>); zooming in (yellow rectangle in the left map) on the distribution of buildings and related earthquake insurance policies in the Besiktas district (<b>right</b>).</p> Full article ">Figure 2
<p>Distribution of policies with respect to buildings’ usage (<b>left</b>), and relative distribution of apartments and offices in the test areas (<b>right</b>).</p> Full article ">Figure 3
<p>Diagram showing the prioritization work-flow with iterative multi-stage sampling.</p> Full article ">Figure 4
<p>(<b>A</b>) Example of a focus map obtained by additive pooling of two normalized layers describing the spatial density of policies with weight w1:0.7 and the estimated seismic hazard in the district, with weight w2:0.3. (<b>B</b>) Same focus map computed with weights (0.5, 0.5). (<b>C</b>) Same focus map computed with weights (0.1, 0.9). Lower Part: post-event focus maps with increasing weights in order to increase selectivity of the sampling approach. From left to right, weights are, respectively, (<b>D</b>) (1, 1,1), (<b>E</b>) (2, 2, 2), (<b>F</b>) (3, 3, 3) where layer 1: building/policy distribution, layer 2: seismic hazard, layer 3: road blockages.</p> Full article ">Figure 5
<p>Composition of the training dataset of 1000 buildings randomly selected among the 19,158 buildings of the original dataset (“<span class="html-italic">count</span>” refers to the number of TCIP policies in the building).</p> Full article ">Figure 6
<p>GFZ-MOMA omnidirectional mobile mapping system with data capturing and storage unit, and battery pack.</p> Full article ">Figure 7
<p>Computation of pre-event focus map. (<b>A</b>) spatial distribution of buildings weighted with related number of earthquake policies. (<b>B</b>) expected seismic hazard (PGA) with exceedance probability of 10% in 50 years. (<b>C</b>) resulting focus map obtained by linear pooling of the normalized layers with weights (0.7,0.3).</p> Full article ">Figure 8
<p>Computation of post-event focus map. (<b>A</b>) spatial distribution of buildings weighted with related number of earthquake policies. (<b>B</b>) expected seismic hazard (PGA) with exceedance probability of 10% in 50 years. (<b>C</b>) expected spatial distribution of road blockages. (<b>D</b>) resulting focus map obtained by log-linear pooling of the normalized layers.</p> Full article ">Figure 9
<p>Iterative, multi-stage sampling and routing for a two-day survey. The upper row shows the computed first-level focus map (<b>left</b>) and the second-level focus map (<b>right</b>) obtained by considering the first-level focus map as additional (negatively weighted) layer. The lower row shows the optimized routing for the first day and second day, under consideration of the first day’s route.</p> Full article ">Figure 10
<p>The color of the trace is proportional to the absolute speed of the mobile mapping system (<b>left)</b>. Distribution of the speed during the survey (logarithmic scale) in Km/h (<b>right)</b>.</p> Full article ">Figure 11
<p>Visibility analysis based on viewshed estimation for a subset of the buildings theoretically imaged by the mobile mapping system.</p> Full article ">Figure 12
<p>Example of an omnidirectional image showing a (non-structural) damage of a building. The distance between building and camera is of about 20 m. The damage is visible in the <b>lower</b> part of the image, circled in yellow. The <b>upper right</b> inset shows the geographical location of the captured omnidirectional image.</p> Full article ">Figure 13
<p>Upper section: example of omnidirectional image. Lower section: (<b>A</b>) Partial rectilinear re-projection of the omnidirectional image. Only a portion of the original image is visible, but the operator can dynamically change the angle of view, direction, and apparent zoom in the re-projected image. (<b>B</b>) Rectilinear projection of an omnidirectional image taken in the sample place at a different time, and with rainy weather.</p> Full article ">Figure 14
<p>Comparison among different single stage surveys based on different sample sets. From the left, results are shown for 50-, 200-, and 400-point sample sets.</p> Full article ">Figure 15
<p>Comparison between the length of the resulting route in the optimized case (blue plot) and in the random case (black plot with red points) (<b>top)</b>. As the plot shows, the route length is proportional to the square root of the dimension of the sample set, and the random route is systematically longer. Comparison of the expected number of covered policies with respect to route length (in km) for random sampling (black line, red dots) and focused sampling (blue line, blue dots) <b>(bottom)</b>. The relationship is apparently linear for the focused sampling, while it appears to not be linear for the random sampling, which is much less efficient in any case.</p> Full article ">
1778 KiB  
Article
Operational Monitoring of the Desert Locust Habitat with Earth Observation: An Assessment
by François Waldner, Mohamed Abdallahi Babah Ebbe, Keith Cressman and Pierre Defourny
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2379-2400; https://doi.org/10.3390/ijgi4042379 - 30 Oct 2015
Cited by 46 | Viewed by 8343
Abstract
Desert locust swarms intermittently damage crops and pastures in sixty countries from Africa to western Asia, threatening the food security of 10% of the world’s population. During the 20th century, desert locust control operations began organizing, and nowadays, they are coordinated by the [...] Read more.
Desert locust swarms intermittently damage crops and pastures in sixty countries from Africa to western Asia, threatening the food security of 10% of the world’s population. During the 20th century, desert locust control operations began organizing, and nowadays, they are coordinated by the Food and Agriculture Organization (FAO), which promotes a preventative strategy based on early warning and rapid response. This strategy implies a constant monitoring of the populations and of the ecological conditions favorable to their development. Satellite remote sensing can provide a near real-time monitoring of these conditions at the continental scale. Thus, the desert locust control community needs a reliable detection of green vegetation in arid and semi-arid areas as an indicator of potential desert locust habitat. To meet this need, a colorimetric transformation has been developed on both SPOT-VEGETATION and MODIS data to produce dynamic greenness maps. After their integration in the daily locust control activities, this research aimed at assessing those dynamic greenness maps from the producers’ and the users’ points of view. Eight confusion matrices and Pareto boundaries were derived from high resolution reference maps representative of the temporal and spatial diversity of Mauritanian habitats. The dynamic greenness maps were found to be accurate in summer breeding areas (F-score = 0.64–0.87), but accuracy dropped in winter breeding areas (F-score = 0.28–0.40). Accuracy is related to landscape fragmentation (R2 = 0.9): the current spatial resolution remains too coarse to resolve complex fragmented patterns and accounts for a substantial (60%) part of the error. The exploitation of PROBA-V 100-m images at the finest resolution (100-m) would enhance by 20% the vegetation detection in fragmented habitat. A survey revealed that end-users are satisfied with the product and find it fit for monitoring, thanks to an intuitive interpretation, leading to more efficiency. Full article
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Dynamic greenness map products for Mauritania on two contrasted dates: (<b>a</b>) second decade of February 2011; and (<b>b</b>) first decade of September 2011; (<b>c</b>) the color code of the time meter. This illustrates the spatial-temporal variability of vegetation and the vegetation response to seasonal rainfall. The product with its time meter flags priority areas to be surveyed (warm colors) because of a recent greening of vegetation becoming suitable for locusts, as they prefer fresh vegetation. On the contrary, dark-colored areas present a lower interest for locusts.</p> Full article ">Figure 2
<p>Mauritania and its six ecological domains. Two observations on the four study sites ensure a robust spatial-temporal sampling. Summer breeding areas cover the south domain below the 18th parallel. In the spring and the winter, locusts breed in favorable areas of the north and the center-west domains.</p> Full article ">Figure 3
<p>Percentage of vegetation in the reference maps aggregated at 250 m: (<b>a</b>) center-west 20 February 2010; (<b>b</b>) center-west 19 November 2010; (<b>c</b>) south Tidjikja 21 January 2010; (<b>d</b>) south Tidjikja 29 July 2009; (<b>e</b>) Chemama 20 February 2010; (<b>f</b>) Chemama 19 November 2010; (<b>g</b>) south Nema 17 February 2010; (<b>h</b>) south Nema 12 October 2009.</p> Full article ">Figure 4
<p>Flowchart of the accuracy assessment. The analysis of the ROC curve built with the greenness maps and <span class="html-italic">in situ</span> observation allowed defining an optimal vegetation threshold for detection. This threshold was then applied to continuous vegetation reference maps degraded at 250 m to derive confusion matrices for the greenness maps. The Pareto boundary was also computed thanks to the continuous vegetation reference maps, which permitted identifying the part of the error solely due to the resolution.</p> Full article ">Figure 5
<p>ROC curve for the field vegetation density (<span class="html-italic">n</span> = 113). The best cut-off value (in red) corresponds to the optimal percentage of vegetation coverage for detection. The Youden index J identifies the threshold to be further used in the computation of the error matrices. The AUC of 84.9% indicates good performance compared to a random classifier.</p> Full article ">Figure 6
<p>Pareto boundaries for the eight reference maps and the simulated boundaries for the SPOT-VEGETATION (1000 m), MODIS (250 m) and PROBA-V(100 m) sensors. In the current 250-m product, the unreachable region accounts for up to 60% of the errors. (<b>a</b>) Atar, dry season; (<b>b</b>) Atar, rainy season; (<b>c</b>) Tidjikja, dry season; (<b>d</b>) Tidjikja, rainy season; (<b>e</b>) Aleg, dry season; (<b>f</b>) Aleg, rainy season; (<b>g</b>) Nema, dry season; (<b>h</b>) Nema, rainy season.</p> Full article ">Figure 7
<p>(<b>a</b>) Habitat fragmentation is an indicator of the achieved accuracy. The higher the fragmentation, the lower the accuracy rate. (<b>b</b>) Simulated reduction of the unreachable region for a spatial resolution of 100 m corresponding to that of PROBA-V. In a fragmented habitat, a product at 100 m would reduce the low -resolution bias by 20%.</p> Full article ">Figure 8
<p>Results of the user survey. National information officers are generally satisfied about the operational provision of remote sensing products for desert locust monitoring in their countries (online questionnaire, April 2012).</p> Full article ">
771 KiB  
Project Report
Innovation in OGC: The Interoperability Program
by George Percivall, Terry Idol, Nadine Alameh and Jeff Harrison
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2362-2378; https://doi.org/10.3390/ijgi4042362 - 30 Oct 2015
Cited by 1 | Viewed by 6979
Abstract
The OGC Interoperability Program is a source of innovation in the development of open standards. The approach to innovation is based on hands-on; collaborative engineering leading to more mature standards and implementations. The process of the Interoperability Program engages a community of sponsors [...] Read more.
The OGC Interoperability Program is a source of innovation in the development of open standards. The approach to innovation is based on hands-on; collaborative engineering leading to more mature standards and implementations. The process of the Interoperability Program engages a community of sponsors and participants based on an economic model that benefits all involved. Each initiative begins with an innovative approach to identify interoperability needs followed by agile software development to advance the state of technology to the benefit of society. Over eighty initiatives have been conducted in the Interoperability Program since the breakthrough Web Mapping Testbed began the program in 1999. OGC standards that were initiated in Interoperability Program are the basis of two thirds of the certified compliant products. Full article
(This article belongs to the Special Issue 20 Years of OGC: Open Geo-Data, Software, and Standards)
►▼ Show Figures

Figure 1

Figure 1
<p>Iterative software development processes driving standards.</p> Full article ">Figure 2
<p>Developments within an OGC-IP Initiative (Source [<a href="#B9-ijgi-04-02362" class="html-bibr">9</a>]).</p> Full article ">Figure 3
<p>Increasing Technology Readiness in OGC-IP Initiatives.</p> Full article ">Figure 4
<p>OWS-9 Testbed Kickoff.</p> Full article ">Figure 5
<p>Development of Sensor Web Enablement in OGC Testbeds.</p> Full article ">Figure 6
<p>OGC-IP influence on SWE Deployments.</p> Full article ">Figure 7
<p>Aviation in OGC-IP.</p> Full article ">Figure 8
<p>Roles within an OGC-IP Initiative.</p> Full article ">Figure 9
<p>Interactions with OGC Programs.</p> Full article ">Figure 10
<p>OGC-IP Advantages.</p> Full article ">Figure 11
<p>Ingredients for Success in OGC-IP.</p> Full article ">
2003 KiB  
Article
Bridge Performance Assessment Based on an Adaptive Neuro-Fuzzy Inference System with Wavelet Filter for the GPS Measurements
by Mosbeh R. Kaloop, Jong Wan Hu and Mohamed A. Sayed
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2339-2361; https://doi.org/10.3390/ijgi4042339 - 28 Oct 2015
Cited by 9 | Viewed by 5978
Abstract
This study describes the performance assessment of the Huangpu Bridge in Guangzhou, China based on long-term monitoring in real-time by the kinematic global positioning system (RTK-GPS) technique. Wavelet transformde-noising is applied to filter the GPS measurements, while the adaptive neuro-fuzzy inference system (ANFIS) [...] Read more.
This study describes the performance assessment of the Huangpu Bridge in Guangzhou, China based on long-term monitoring in real-time by the kinematic global positioning system (RTK-GPS) technique. Wavelet transformde-noising is applied to filter the GPS measurements, while the adaptive neuro-fuzzy inference system (ANFIS) time series output-only model is used to predict the deformations of GPS-bridge monitoring points. In addition, GPS and accelerometer monitoring systems are used to evaluate the bridge oscillation performance. The conclusions drawn from investigating the numerical results show that: (1)the wavelet de-noising of the GPS measurements of the different recording points on the bridge is a suitable tool to efficiently eliminate the signal noise and extract the different deformation components such as: semi-static and dynamic displacements; (2) the ANFIS method with two multi-input single output model is revealed to powerfully predict GPS movement measurements and assess the bridge deformations; and (3) The installed structural health monitoring system and the applied ANFIS movement prediction performance model are solely sufficient to assure bridge safety based on the analyses of the different filtered movement components. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Huangpu bridge view.</p> Full article ">Figure 2
<p>Bridge elevation and health monitoring system.</p> Full article ">Figure 3
<p>WT de-noising of GPS measurements of (<b>a</b>) Points 1 (Deck Point) and (<b>b</b>) T1 (Tower Point) in Y direction.</p> Full article ">Figure 4
<p>GPS standard deviation of three direction measurements before (B) and after (A) WT de-noising.</p> Full article ">Figure 5
<p>ANFIS architecture model.</p> Full article ">Figure 6
<p>Hybrid ANFIS learning technique algorithm.</p> Full article ">Figure 7
<p>Training and testing RMSE with epoch number.</p> Full article ">Figure 8
<p>The (<b>a</b>) design model, (<b>b</b>) MF adjustment and (<b>c</b>) model application of two MF and two inputs time series ANFIS prediction model.</p> Full article ">Figure 9
<p>The ANFIS prediction movement for the GPS measurements (<b>a</b>) before and (<b>c</b>) after wavelet filter applied and ACF for the test monitoring data (<b>b</b>) before and (<b>d</b>) after filter applied.</p> Full article ">Figure 10
<p>Displacement time series of bridge deck on 10 Sept. for (<b>a</b>) point 3; (<b>b</b>) point 7 and (<b>c</b>) Displacement range of GPS measurements.</p> Full article ">Figure 11
<p>(<b>a</b>) The time series of Displacement in Y-direction and (<b>b</b>) scattered of GPS measurements for the three towers of bridge(the arrows refers to the movement direction).</p> Full article ">Figure 12
<p>GPS prediction model performance analysis.</p> Full article ">Figure 13
<p>TheANFISPrediction movement performance on (<b>a</b>) 15 Sept.; (<b>b</b>) 25 Sept. and ACF for the check time on (<b>c</b>) 15 Sept.; (<b>d</b>) 25 Sept. of ANFIS model design.</p> Full article ">
322 KiB  
Article
Spatiotemporal Data Mining: A Computational Perspective
by Shashi Shekhar, Zhe Jiang, Reem Y. Ali, Emre Eftelioglu, Xun Tang, Venkata M. V. Gunturi and Xun Zhou
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2306-2338; https://doi.org/10.3390/ijgi4042306 - 28 Oct 2015
Cited by 152 | Viewed by 19274
Abstract
Explosive growth in geospatial and temporal data as well as the emergence of new technologies emphasize the need for automated discovery of spatiotemporal knowledge. Spatiotemporal data mining studies the process of discovering interesting and previously unknown, but potentially useful patterns from large spatiotemporal [...] Read more.
Explosive growth in geospatial and temporal data as well as the emergence of new technologies emphasize the need for automated discovery of spatiotemporal knowledge. Spatiotemporal data mining studies the process of discovering interesting and previously unknown, but potentially useful patterns from large spatiotemporal databases. It has broad application domains including ecology and environmental management, public safety, transportation, earth science, epidemiology, and climatology. The complexity of spatiotemporal data and intrinsic relationships limits the usefulness of conventional data science techniques for extracting spatiotemporal patterns. In this survey, we review recent computational techniques and tools in spatiotemporal data mining, focusing on several major pattern families: spatiotemporal outlier, spatiotemporal coupling and tele-coupling, spatiotemporal prediction, spatiotemporal partitioning and summarization, spatiotemporal hotspots, and change detection. Compared with other surveys in the literature, this paper emphasizes the statistical foundations of spatiotemporal data mining and provides comprehensive coverage of computational approaches for various pattern families. ISPRS Int. J. Geo-Inf. 2015, 4 2307 We also list popular software tools for spatiotemporal data analysis. The survey concludes with a look at future research needs. Full article
(This article belongs to the Special Issue Advances in Spatio-Temporal Data Analysis and Mining)
►▼ Show Figures

Figure 1

Figure 1
<p>The process of spatiotemporal data mining.</p> Full article ">Figure 2
<p>Categorization of spatial and spatiotemporal data mining surveys.</p> Full article ">Figure 3
<p>Flow Anomaly Example. (<b>a</b>) Example input. (<b>b</b>) Example output.</p> Full article ">
1151 KiB  
Article
Image Segmentation Parameter Optimization Considering Within- and Between-Segment Heterogeneity at Multiple Scale Levels: Test Case for Mapping Residential Areas Using Landsat Imagery
by Brian A. Johnson, Milben Bragais, Isao Endo, Damasa B. Magcale-Macandog and Paula Beatrice M. Macandog
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2292-2305; https://doi.org/10.3390/ijgi4042292 - 27 Oct 2015
Cited by 81 | Viewed by 8053
Abstract
Multi-scale/multi-level geographic object-based image analysis (MS-GEOBIA) methods are becoming widely-used in remote sensing because single-scale/single-level (SS-GEOBIA) methods are often unable to obtain an accurate segmentation and classification of all land use/land cover (LULC) types in an image. However, there have been few comparisons [...] Read more.
Multi-scale/multi-level geographic object-based image analysis (MS-GEOBIA) methods are becoming widely-used in remote sensing because single-scale/single-level (SS-GEOBIA) methods are often unable to obtain an accurate segmentation and classification of all land use/land cover (LULC) types in an image. However, there have been few comparisons between SS-GEOBIA and MS-GEOBIA approaches for the purpose of mapping a specific LULC type, so it is not well understood which is more appropriate for this task. In addition, there are few methods for automating the selection of segmentation parameters for MS-GEOBIA, while manual selection (i.e., trial-and-error approach) of parameters can be quite challenging and time-consuming. In this study, we examined SS-GEOBIA and MS-GEOBIA approaches for extracting residential areas in Landsat 8 imagery, and compared naïve and parameter-optimized segmentation approaches to assess whether unsupervised segmentation parameter optimization (USPO) could improve the extraction of residential areas. Our main findings were: (i) the MS-GEOBIA approaches achieved higher classification accuracies than the SS-GEOBIA approach, and (ii) USPO resulted in more accurate MS-GEOBIA classification results while reducing the number of segmentation levels and classification variables considerably. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of study area and Landsat 8 satellite imagery (natural color composite).</p> Full article ">Figure 2
<p>(<b>a</b>) Subset of the study area image. (<b>b</b>) segmentation with excessive oversegmentation (gray lines) and excessive undersegmentation (black lines) of residential areas. The yellow boxes show some examples of residential areas with different spectral properties. The red box (industrial area) and blue boxes (bare soil areas) show land use/land cover types with high spectral similarity to residential areas.</p> Full article ">Figure 3
<p>(<b>a</b>) Subset of the study area image. Optimal segmentations for the three-level GEOBIA approach: (<b>b</b>) SP120 segmentation; (<b>c</b>) SP80 segmentation; (<b>d</b>) SP40 segmentation.</p> Full article ">Figure 4
<p>Study area image and the three-level MS-GEOBIA classification result.</p> Full article ">
12642 KiB  
Article
Cloud-Based Geospatial 3D Image Spaces—A Powerful Urban Model for the Smart City
by Stephan Nebiker, Stefan Cavegn and Benjamin Loesch
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2267-2291; https://doi.org/10.3390/ijgi4042267 - 26 Oct 2015
Cited by 27 | Viewed by 8316
Abstract
In this paper, we introduce the concept and an implementation of geospatial 3D image spaces as new type of native urban models. 3D image spaces are based on collections of georeferenced RGB-D imagery. This imagery is typically acquired using multi-view stereo mobile [...] Read more.
In this paper, we introduce the concept and an implementation of geospatial 3D image spaces as new type of native urban models. 3D image spaces are based on collections of georeferenced RGB-D imagery. This imagery is typically acquired using multi-view stereo mobile mapping systems capturing dense sequences of street level imagery. Ideally, image depth information is derived using dense image matching. This delivers a very dense depth representation and ensures the spatial and temporal coherence of radiometric and depth data. This results in a high-definition WYSIWYG (“what you see is what you get”) urban model, which is intuitive to interpret and easy to interact with, and which provides powerful augmentation and 3D measuring capabilities. Furthermore, we present a scalable cloud-based framework for generating 3D image spaces of entire cities or states and a client architecture for their web-based exploitation. The model and the framework strongly support the smart city notion of efficiently connecting the urban environment and its processes with experts and citizens alike. In the paper we particularly investigate quality aspects of the urban model, namely the obtainable georeferencing accuracy and the quality of the depth map extraction. We show that our image-based georeferencing approach is capable of improving the original direct georeferencing accuracy by an order of magnitude and that the presented new multi-image matching approach is capable of providing high accuracies along with a significantly improved completeness of the depth maps. Full article
(This article belongs to the Special Issue Geo-Information Fostering Innovative Solutions for Smart Cities)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>(<b>a</b>) Conceptual illustration of 3D image spaces consisting of collections of georeferenced multi-view RGB-D imagery; and (<b>b</b>) georeferenced RGB image with its associated depth map (D) containing a depth value for each pixel of the RGB image.</p> Full article ">Figure 2
<p>System architecture and workflow of the infraVIS project.</p> Full article ">Figure 3
<p>(<b>a</b>) Multiview multi-sensor stereovision IVGI mobile mapping system; and (<b>b</b>) detail view showing the three stereo camera systems and the GNSS/IMU positioning system.</p> Full article ">Figure 4
<p>(<b>a</b>) Base map of the test area with overlaid projection centers of the selected image sequences, ground control points (GCPs), terrestrial laser scanning (TLS) stations, and locations of figures of this paper (Source: Geodaten Kanton Basel-Stadt); forward-looking mobile mapping imagery illustrating typical challenges; and (<b>b</b>) GNSS shadowing and numerous pedestrians; (<b>c</b>) heavy traffic with multiple trams, cars and cyclists.</p> Full article ">Figure 5
<p>Perspective view of image sequence 1 with stereo frames (blue rectangles) and GCPs (yellow flags) following a bundle adjustment in PhotoScan. The location of frame 80 with a trajectory jump is marked with a dashed ellipse.</p> Full article ">Figure 6
<p>Differences in [m] between directly georeferenced sensor orientations (projection center coordinates of right and left stereo cameras) and image-based georeferencing from bundle adjustment for image sequence one. Trajectory discontinuity between image frames #80 and #82 are marked with a dashed line and the corresponding images are shown to the right.</p> Full article ">Figure 7
<p>Selected image matching configurations, red: base image, green: match images.</p> Full article ">Figure 8
<p>Base image (L<sub>t0</sub>) and its neighboring images rectified by SURE using polar rectification.</p> Full article ">Figure 9
<p>Deviations of depth maps generated by the SURE triangulation module.</p> Full article ">Figure 10
<p>Depth deviations between point clouds from SURE triangulation and TLS.</p> Full article ">Figure 11
<p>Overview of selected tools and features for interacting with 3D image spaces. (<b>a</b>) Measuring a polygonal area by 3D monoplotting; (<b>b</b>) superimposing existing infrastructure data e.g. water (blue) and waste water (red) pipes in an urban street scene; (<b>c</b>) single-click height measurement from a traffic light to the road surface; (<b>d</b>) measuring of a perpendicular distance (red line) from an orthogonal reference line (green) extracted from pavement border; (<b>e</b>) resulting images of a multi-view query looking for a 3D world coordinate (red square) under different viewing angles; and (<b>f</b>) mobile web application.</p> Full article ">
1331 KiB  
Article
Early Flood Detection for Rapid Humanitarian Response: Harnessing Near Real-Time Satellite and Twitter Signals
by Brenden Jongman, Jurjen Wagemaker, Beatriz Revilla Romero and Erin Coughlan De Perez
ISPRS Int. J. Geo-Inf. 2015, 4(4), 2246-2266; https://doi.org/10.3390/ijgi4042246 - 23 Oct 2015
Cited by 111 | Viewed by 15122
Abstract
Humanitarian organizations have a crucial role in response and relief efforts after floods. The effectiveness of disaster response is contingent on accurate and timely information regarding the location, timing and impacts of the event. Here we show how two near-real-time data sources, satellite [...] Read more.
Humanitarian organizations have a crucial role in response and relief efforts after floods. The effectiveness of disaster response is contingent on accurate and timely information regarding the location, timing and impacts of the event. Here we show how two near-real-time data sources, satellite observations of water coverage and flood-related social media activity from Twitter, can be used to support rapid disaster response, using case-studies in the Philippines and Pakistan. For these countries we analyze information from disaster response organizations, the Global Flood Detection System (GFDS) satellite flood signal, and flood-related Twitter activity analysis. The results demonstrate that these sources of near-real-time information can be used to gain a quicker understanding of the location, the timing, as well as the causes and impacts of floods. In terms of location, we produce daily impact maps based on both satellite information and social media, which can dynamically and rapidly outline the affected area during a disaster. In terms of timing, the results show that GFDS and/or Twitter signals flagging ongoing or upcoming flooding are regularly available one to several days before the event was reported to humanitarian organizations. In terms of event understanding, we show that both GFDS and social media can be used to detect and understand unexpected or controversial flood events, for example due to the sudden opening of hydropower dams or the breaching of flood protection. The performance of the GFDS and Twitter data for early detection and location mapping is mixed, depending on specific hydrological circumstances (GFDS) and social media penetration (Twitter). Further research is needed to improve the interpretation of the GFDS signal in different situations, and to improve the pre-processing of social media data for operational use. Full article
(This article belongs to the Special Issue Geoinformation for Disaster Risk Management)
►▼ Show Figures

Figure 1

Figure 1
<p>Schematic display of a typical Twitter count pattern leading up to a flood event (graph is illustrative; tweets are actual messages derived from flood events in the Philippines in 2014).</p> Full article ">Figure 2
<p>Affected flood areas as derived from different sources: (<b>A</b>) flood signal from GFDS for 6, 9 and 12 September. (<b>B</b>) heat map based on flood related Twitter activity for 6, 9 and 12 September; and (<b>C</b>) inundation map published by UN-OCHA in October 2014, outlining all areas that were inundated at some point in September 2014.</p> Full article ">Figure 3
<p>Flood signal from GFDS (blue line) and Twitter analysis (black lines for tweets in English language (fine dash); Filipino language (coarse dash); and both languages (solid line). The red lines indicate when the event occurred (solid) and was reported (dashed). If only one red line is shown, the dates of occurrence and reporting are the same. Flood types refer to dam break (“Dam”) and typhoon (“TY”). The individual graphs are for different events that occurred throughout 2014, including (<b>A</b>) one dam overtopping and (<b>B</b>,<b>C</b>).two typhoon-triggered river floods.</p> Full article ">Figure 4
<p>Box plots showing the distribution of signal deviation from the mean, for (<b>A</b>) all 80 flood events (<span class="html-italic">i.e.</span>, 80 locations) reported throughout 2014 in the Philippines; and (<b>B</b>) the September 2014 floods in Pakistan at 17 locations. For the event date (“day 0” on x-axis), we used the specific reported event date for each individual event for the Philippines; and 4 September 2014 for Pakistan.</p> Full article ">Figure 5
<p>Twitter activity in Pakistan surrounding the blowing of flood defenses around Athara Hazari for the protection of the Sugar Mills at Trimmu and the city Jhang.</p> Full article ">Figure 6
<p>GFDS signal for locations directly (<b>A</b>) downstream and (<b>B</b>) upstream of the Tarbella Dam, in North-Eastern Pakistan.</p> Full article ">Figure 7
<p>Tweet count related to dams and barrages, along the Jhelum, Chenab, and Ravi rivers. The locations are roughly ordered from upstream to downstream (numbered 1 to 6).</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-53
Previous Issue
Next Issue
Back to TopTop