ISPRS International Journal of Geo-Information

Journal: ISPRS International Journal of Geo-Information

Abbreviation

ISPRS int. j. geo-inf.

Publisher

MDPI

ISSN

2220-9964

Show all metadata

Search Results

Publications1 - 10 of 31

Scale- and Resolution-Adapted Shaded Relief Generation Using U-Net
Farmakis-Serebryakova, Marianna; Heitzler, Magnus; Hurni, Lorenz (2024)
ISPRS International Journal of Geo-Information
On many maps, relief shading is one of the most significant graphical elements. Modern relief shading techniques include neural networks. To generate such shading automatically at an arbitrary scale, one needs to consider how the resolution of the input digital elevation model (DEM) relates to the neural network process and the maps used for training. Currently, there is no clear guidance on which DEM resolution to use to generate relief shading at specific scales. To address this gap, we trained the U-Net models on swisstopo manual relief shadings of Switzerland at four different scales and using four different resolutions of SwissALTI3D DEM. An interactive web application designed for this study allows users to outline a random area and compare histograms of varying brightness between predictions and manual relief shadings. The results showed that DEM resolution and output scale influence the appearance of the relief shading, with an overall scale/resolution ratio. We present guidelines for generating relief shading with neural networks for arbitrary areas and scales.
Trip purpose imputation using GPS trajectories with machine learning
Gao, Qinggang; Molloy, Joseph; Axhausen, Kay W. (2021)
ISPRS International Journal of Geo-Information
We studied trip purpose imputation using data mining and machine learning techniques based on a dataset of GPS-based trajectories gathered in Switzerland. With a large number of labeled activities in 8 categories, we explored location information using hierarchical clustering and achieved a classification accuracy of 86.7% using a random forest approach as a baseline. The contribution of this study is summarized below. Firstly, using information from GPS trajectories exclusively without personal information shows a negligible decrease in accuracy (0.9%), which indicates the good performance of our data mining steps and the wide applicability of our imputation scheme in case of limited information availability. Secondly, the dependence of model performance on the geographical location, the number of participants, and the duration of the survey is investigated to provide a reference when comparing classification accuracy. Furthermore, we show the ensemble filter to be an excellent tool in this research field not only because of the increased accuracy (93.6%) especially for minority classes, but also the reduced uncertainties in blindly trusting the labeling of activities by participants, which is vulnerable to class noise due to the large survey response burden. Finally, the trip purpose derivation accuracy across participants reaches 74.8%, which is significant and suggests the possibility of effectively applying a model trained on GPS trajectories of a small subset of citizens to a larger GPS trajectory sample.
Evaluation Methods for Citizen Design Science Studies: How Do Planners and Citizens Obtain Relevant Information from Map-Based E-Participation Tools?
Müller, Johannes (2021)
ISPRS International Journal of Geo-Information
A successful e-participation campaign in urban planning relies on good two-way communication between the expert and the citizen. While the presentation of information from planners to citizens is one concern of that topic, we address in this paper the question of how citizens’ inputs can be evaluated for map-based e-participation tools. The interest is, on the one side, in the usefulness of the input for the planner and, on the other side, in performing a quick assessment which can provide feedback to the participant via the tool’s interface. We use a test dataset that was acquired with an online city planning tool that uses 3D geometries and develop analysis methods from it that can also be generalized for other map-based e-participation tools. These analysis methods are meant to be applied to large datasets and to enhance e-participation methods in urban planning and design to citizen (design) science approaches. The methods range from the calculation of simple parameters and heatmaps over clustering to point pattern analysis. We evaluate the presented approaches by their computation time and their usefulness for the planner and non-expert citizen and investigate their potential to serve as a composite analysis. We found that functions of the point pattern analysis reveal relevant information of the users’ inputs but require a simplified presentation. We introduce a spatial dispersion index as an example to present the relations between objects in a clear way.
Multi-View Instance Matching with Learned Geometric Soft-Constraints
Nassar, Ahmed S.; Lefèvre, Sébastien; Wegner, Jan D. (2020)
ISPRS International Journal of Geo-Information
We present a new approach for matching urban object instances across multiple ground-level images for the ultimate goal of city-scale mapping of objects with high positioning accuracy. What makes this task challenging is the strong change in view-point, different lighting conditions, high similarity of neighboring objects, and variability in scale. We propose to turn object instance matching into a learning task, where image-appearance and geometric relationships between views fruitfully interact. Our approach constructs a Siamese convolutional neural network that learns to match two views of the same object given many candidate image cut-outs. In addition to image features, we propose utilizing location information about the camera and the object to support image evidence via soft geometric constraints. Our method is compared to existing patch matching methods to prove its edge over state-of-the-art. This takes us one step closer to the ultimate goal of city-wide object mapping from street-level imagery to benefit city administration.
Comparison of relief shading techniques applied to landforms
Farmakis-Serebryakova, Marianna; Hurni, Lorenz (2020)
ISPRS International Journal of Geo-Information
As relief influences disposition of all the other objects displayed on maps, terrain representation plays one of the key roles in the map creation process. Originally a manual technique, relief shading creates the three-dimensional effect and allows the user to read the terrain in an intuitive way. With the advent of digital elevation models (DEMs) analytical relief shading came into a wider use, since it is faster, requires less effort, and delivers reproducible results. In contrast to manual relief shading, however, it often lacks clarity when representing heterogeneous landscapes with diverse landforms. The aim of this work is to evaluate analytical hillshading methods against a set of landforms within an online survey. The responses revealed that the clear sky model performs best applied to most of the landforms included in the survey, in particular all the mountain and valley types. Cluster shading proved to work well for the mountainous and hilly areas but less so in the depiction of valleys. Texture shading and the multidirectional, oblique-weighted (MDOW) method deliver too much detail for most of the landforms presented. Glaciers were depicted in the best way using the aspect tool. For alluvial fans, the standard relief shading with custom lighting direction proved to work best compared to the other methods.
Special Issue "Volunteered Geographic Information"
(2017)
ISPRS International Journal of Geo-Information
Geo-Tagged Social Media Data-Based Analytical Approach for Perceiving Impacts of Social Events
Zhu, Ruoxin; Lin, Diao; Jendryke, Michael; et al. (2019)
ISPRS International Journal of Geo-Information
Studying the impact of social events is important for the sustainable development of society. Given the growing popularity of social media applications, social sensing networks with users acting as smart social sensors provide a unique channel for understanding social events. Current research on social events through geo-tagged social media is mainly focused on the extraction of information about when, where, and what happened, i.e., event detection. There is a trend towards the machine learning of more complex events from even larger input data. This research work will undoubtedly lead to a better understanding of big geo-data. In this study, however, we start from known or detected events, raising further questions on how they happened, how they affect people’s lives, and for how long. By combining machine learning, natural language processing, and visualization methods in a generic analytical framework, we attempt to interpret the impact of known social events from the dimensions of time, space, and semantics based on geo-tagged social media data. The whole analysis process consists of four parts: (1) preprocessing; (2) extraction of event-related information; (3) analysis of event impact; and (4) visualization. We conducted a case study on the “2014 Shanghai Stampede” event on the basis of Chinese Sina Weibo data. The results are visualized in various ways, thus ensuring the feasibility and effectiveness of our proposed framework. Both the methods and the case study can serve as decision references for situational awareness and city management.
Improving destination choice modeling using location-based big data
Molloy, Joseph; Moeckel, Rolf (2017)
ISPRS International Journal of Geo-Information
Citizens are increasingly sharing their location and movements through “check-ins” on location based social networks (LBSNs). These services are collecting unprecedented amounts of big data that can be used to study how we travel and interact with our environment. This paper presents the development of a long distance destination choice model for Ontario, Canada, using data from Foursquare to model destination attractiveness. A methodology to collect and process historical check-in counts has been developed, allowing the utility of each destination to be calculated based on the intensity of different activities performed at the destination. Destinations such as national parks and ski areas are very strong attractors of leisure trips, yet do not employ many people and have few residents. Trip counts to such destinations are therefore poorly predicted by models based on population and employment. Traditionally, this has been remedied by extensive manual data collection. The integration of Foursquare data offers an alternative approach to this problem. The Foursquare based destination choice model was evaluated against a traditional model estimated only with population and employment. The results demonstrate that data from LBSNs can be used to improve destination choice models, particularly for leisure travel.
Getting Real: The Challenge of Building and Validating a Large-Scale Digital Twin of Barcelona’s Traffic with Empirical Data
Argota Sánchez-Vaquerizo, Javier (2022)
ISPRS International Journal of Geo-Information
Large-scale microsimulations are increasingly resourceful tools for analysing in detail citywide effects and alternative scenarios of our policy decisions, approximating the ideal of ‘urban digital twins’. Yet, these models are costly and impractical, and there are surprisingly few published examples robustly validated with empirical data. This paper, therefore, presents a new large-scale agent-based traffic microsimulation for the Barcelona urban area using SUMO to show the possibilities and challenges of building these scenarios based on novel fine-grained empirical big data. It combines novel mobility data from real cell phone records with conventional surveys to calibrate the model comparing two different dynamic assignment methods for getting an operationally realistic and efficient simulation. Including through traffic and the use of a stochastic adaptive routing approach results in a larger 24-hour model closer to reality. Based on an extensive multi-scalar evaluation including traffic counts, hourly distribution of trips, and macroscopic metrics, this model expands and outperforms previous large-scale scenarios, which provides new operational opportunities in city co-creation and policy. The novelty of this work relies on the effective modelling approach using newly available data and the realistic robust evaluation. This allows the identification of the fundamental challenges of simulation to accurately capture real-world dynamical systems and to their predictive power at a large scale, even when fed by big data, as envisioned by the digital twin concept applied to smart cities.
GIS for Renewable Energy
Raubal, Martin (2014)
ISPRS International Journal of Geo-Information

Publications1 - 10 of 31

Journal: ISPRS International Journal of Geo-Information

Abbreviation

Publisher

Journal Volumes

ISSN

Description

Filters

Search Results