Article
Geosciences, Multidisciplinary
Robert Szczepanek, Cezary Tos, Marek Bodziony
Summary: Flood marks have significant historical and cultural importance, but many of them have not been properly categorized and surveyed. This study used photogrammetric methods to recreate the precise location of a lost flood mark and proposed an innovative method using smartphones to register temporary flood marks. Test results demonstrated that the proposed guidelines allow for inexpensive and quick creation of accurate flood marks with an accuracy of 10 mm.
INTERNATIONAL JOURNAL OF DISASTER RISK REDUCTION
(2022)
Review
Urban Studies
Jiaxuan Wang, Yoong Shin Chow, Filip Biljecki
Summary: This study examines urban characteristics through homestay guest experiences and analyzes dominant topics about urban characteristics in Airbnb reviews. The research finds that these reviews reflect transportation convenience, amenities, sense of safety, and noise pollution. The spatial distribution of perceptions may be affected by the morphology and development of a city, as well as the diversity of guests.
Article
Geosciences, Multidisciplinary
Gabriel Araujo de Oliveira, Ayane Aparecida da Silva Ribeiro, Jose Almir Cirilo
Summary: This study explores the use of collaborative spatial data as an alternative for hydrodynamic model calibration in flood risk mapping. The study addresses the uncertainties in the modeling process, especially in calibration and validation stages. It develops a collaborative platform, identifies strategies to engage potential collaborators, analyzes data consistency, and assesses the capability and restrictions of using collaborative information. Results confirm that collaborative spatial data can be a valid alternative for hydrodynamic model calibration, with simulations using such data showing errors below 10 cm. Public motivation and engagement, particularly through face-to-face workshops, are found to be crucial for successful data collection and calibration. The study aims to reduce uncertainty in hydrological risk mapping by addressing observed data scarcity and its impact on model calibration and validation.
Article
Chemistry, Analytical
Juan Salazar-Carrillo, Miguel Torres-Ruiz, Clodoveu A. Davis, Rolando Quintero, Marco Moreno-Ibarra, Giovanni Guzman
Summary: This paper introduces how to geocode and predict traffic congestion using social media data, build a prediction model, and display spatial distribution using heat maps, demonstrating that social media is a good alternative for gathering dynamic city information.
Article
Geography
Rotem Elinson, Noam Levin
Summary: The study found that Volunteered Geographic Information (VGI) can be used to study nature-based recreational activities, beyond traditional monitoring methods and databases. By analyzing the database from an Israeli VGI app and website, the study revealed significant correlations between the density of points of interest (POIs) and accessibility and attraction variables.
Review
Geography, Physical
Mikko Roenneberg, Pyry Kettunen
Summary: Due to various factors, Finland's cadastral index map contains numerous low-quality border markers. In order to address this issue, researchers have developed a gamified map-based artefact that allows citizens to improve the quality of these markers. The study demonstrates that gamified crowdsourcing is an effective way to motivate citizens to perform complex tasks. Progression, points, and leaderboards were found to be the most motivating gamification features. Involving stakeholders early on and focusing on usability resulted in a positive user experience.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Environmental Sciences
Yanzhe Yin, Andrew Grundstein, Deepak R. Mishra, Lakshmish Ramaswamy, Navid Hashemi Tonekaboni, John Dowd
Summary: The newly proposed Dynamic urban Thermal Exposure index (DTEx) combines human movement patterns and heat hazard patterns, successfully identifying areas with high heat exposure, providing crucial information for heat mitigation measures and risk communication in urban environments.
ENVIRONMENT INTERNATIONAL
(2021)
Article
Geography
Christina Ludwig, Sascha Fendrich, Alexander Zipf
Summary: This study investigates the variability of association rules extracted from OpenStreetMap across different geographic regions and depending on different context variables. It was found that limiting the association rule analysis to specific context variables in parks increased the number of rules applicable across multiple cities.
TRANSACTIONS IN GIS
(2021)
Article
Computer Science, Information Systems
Stathis G. Arapostathis
Summary: The study analyzes social network data as a VGI source for the September 2016 flood event in Messinia, South Greece, utilizing machine learning to classify Twitter content into ten categories. The methodology involves geo-referencing tweets, validating data, and creating maps and graphs for disaster management purposes. The automated process aims to deliver real-time advanced information to all disaster management stakeholders for effective response and mitigation efforts.
INFORMATION SYSTEMS FRONTIERS
(2021)
Article
Computer Science, Artificial Intelligence
Zahra Pourbahman, Saeedeh Momtazi, Alireza Bagheri
Summary: Information Retrieval (IR) provides access to unstructured data to satisfy user needs. This study proposes a distributed transformer-based deep neural ranking model called DSDNN, which captures deep semantic understanding using a context-aware language representation model. Experimental results on three datasets and three IR tasks show the effectiveness of DSDNN compared to existing neural ranking models.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Wenying Du, Chang Ge, Shuang Yao, Nengcheng Chen, Lei Xu
Summary: This study conducted applicability analysis on BERT and four traditional methods for social media textual data and found their performance varies in different situations. The study also discovered the topic evolution pattern and hotspot changes related to a rainstorm in Henan. By integrating multiple methods, the overall topic classification accuracy of Sina microblogs was improved, facilitating spatiotemporal analysis of flooding.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2023)
Editorial Material
Computer Science, Information Systems
Sultan Kocaman, Sameer Saran, Murat Durmaz, Senthil Kumar
Summary: This article introduces the importance of citizen science and geospatial capacity building, emphasizing their roles and advantages in geospatial data-related work, and points out future research directions, including data quality, ontology studies, development of tools based on machine learning and artificial intelligence, and open science and open data practices.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2021)
Article
Plant Sciences
Elizabeth A. Schrammeijer, Boris T. van Zanten, Jac Davis, Peter H. Verburg
Summary: Urban green spaces are crucial for sustainable cities, and this study used a crowdsourcing method to explore the use, perception, and preferences of different green configurations in a large urban park in Amsterdam. The findings revealed differences in respondent samples and stated preferences between the mobile app and traditional survey methods.
URBAN FORESTRY & URBAN GREENING
(2022)
Article
Computer Science, Information Systems
Clessio Moura de Souza, Dominik Kremer, Blake Byron Walker
Summary: This study examines how individuals in Recife, Brazil contribute to the reproduction of violent spaces through knowledge and experiences of violence. The results indicate significant differences in placial anchorings between police officers and civilians, and highlight dynamic flows of VGI among residents, local news organizations, and social media.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Article
Environmental Studies
Colin Ferster, Trisalyn Nelson, Kevin Manaugh, Jeneva Beairsto, Karen Laberee, Meghan Winters
Summary: High-quality and consistent cycling infrastructure data are needed for research and planning, but there is no national dataset in Canada. The goal is to develop a dataset using the Can-BICS system by classifying OSM, and an accuracy assessment showed estimated accuracies for presence/absence, comfort class, and infrastructure type.
ENVIRONMENT AND PLANNING B-URBAN ANALYTICS AND CITY SCIENCE
(2023)
Review
Computer Science, Information Systems
Yu Feng, Xiao Huang, Monika Sester
Summary: With the rapid development of deep learning techniques, the concept of 'citizen as sensors' has gradually become a reality, involving citizens in acquiring information on natural disasters via social media. Integrating information extracted from various sources to provide comprehensive details on disasters is of great significance.
INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE
(2022)
Article
Ophthalmology
Zhe Zeng, Sai Liu, Hao Cheng, Hailong Liu, Yang Li, Yu Feng, Felix Wilhelm Siebert
Summary: This paper introduces gaze input as a promising method for contact-free interaction in human-machine systems. The authors present the GazeVending interface (GaVe) that allows users to control actions on a display using their eyes. GaVe works on a regular webcam and only requires a short calibration before use. The interface is designed in a hierarchical structure, presenting broad item clusters to users and guiding them through another selection round based on dwell time. User study results show that participants quickly understood how to interact with GaVe and performed well in selecting target items.
JOURNAL OF EYE MOVEMENT RESEARCH
(2023)
Article
Computer Science, Information Systems
Stefania Zourlidou, Monika Sester, Shaohan Hu
Summary: In this paper, a new method is proposed to detect traffic regulations at intersections using GPS traces. The lack of mapping traffic regulations on maps and the cost and time constraints of surveying equipment for mapping motivate this study. A new method based on summarizing the collective behavior of vehicle crossing intersections is proposed, and a clustering algorithm is modified for detecting stopping and deceleration episodes.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2023)
Article
Computer Science, Information Systems
Stefan Fuest, Olga Shkedova, Monika Sester
Summary: This article presents an approach to promote routes that reduce road users' exposure to areas that should be temporarily avoided due to traffic or environmental considerations. At the same time, routes that are already heavily polluted should be avoided to reduce pollution and distribute it more equitably in the environment. In this research, a design study is presented for the creation of route maps that aim to visually communicate the preference for route options in the case of high air pollution.
INTERNATIONAL JOURNAL OF CARTOGRAPHY
(2023)
Article
Remote Sensing
Qianbao Hou, Rui Xia, Jiahuan Zhang, Yu Feng, Zongqian Zhan, Xin Wang
Summary: This paper addresses the challenge of efficiently and accurately identifying visual overlapping image pairs in large-scale SfM. It fine-tunes several popular CNN models and utilizes local overlapping information to improve retrieval performance.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Article
Geography
Cheng Fu, Zhiyong Zhou, Yu Feng, Robert Weibel
Summary: This study proposes a layered data representation model to address the "wobbly-wall" effect in raster-based building generalization. The Residual U-Net model can already produce straight walls and maintain rectangularity and parallelism of buildings at different scales. Increasing the input tensor size is an effective solution to improve prediction. Adjustments in sample balancing and data augmentation can enhance network learning, but tradeoffs exist.
CARTOGRAPHY AND GEOGRAPHIC INFORMATION SCIENCE
(2023)
Proceedings Paper
Computer Science, Cybernetics
Vinu Kamalasanan, Awad Mukbil, Monika Sester, Jorg P. Muller
Summary: Virtual Reality (VR) has been widely used in behavioral studies, including urban traffic research. This paper presents a framework called Pedestrian-in-the-Loop (PIL) Mixed Reality (MR), where virtual cyclist avatars coexist with humans in a real-world outdoor space. The novelty of this approach lies in modeling virtual avatars as cognitive agents, allowing them to sense and interact with the environment. The framework shows great potential in effectively studying pedestrian interactions.
2022 IEEE INTERNATIONAL SYMPOSIUM ON MIXED AND AUGMENTED REALITY ADJUNCT (ISMAR-ADJUNCT 2022)
(2022)
Proceedings Paper
Geography, Physical
Y. Yuan, M. Sester
Summary: In this paper, a RANSAC-based method is proposed to correct relative localization errors between Connected Autonomous Vehicles (CAVs). This method leverages real-time data sharing and uses dynamic objects, in addition to static environmental information, to improve the accuracy of localization. Experimental results demonstrate that the proposed method greatly reduces relative localization errors, making it suitable for real-time scenarios in autonomous driving.
XXIV ISPRS CONGRESS: IMAGING TODAY, FORESEEING TOMORROW, COMMISSION I
(2022)
Proceedings Paper
Geography, Physical
F. Politz, M. Sester
Summary: National Mapping Agencies acquire point cloud data from ALS sensors and DIM on aerial images to detect building changes using a residual neural network. The experiments show that this approach can automatically and reliably detect building changes.
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION II
(2022)
Proceedings Paper
Geography, Physical
Qianqian Zou, Monika Sester
Summary: This study proposes a method for modeling 3D building facades with uncertainty quantification in complex urban scenes. The uncertainties of windows and facades are indicated in a probabilistic fashion, and the potential locations of missing objects are inferred using the Monte Carlo sampling approach. The proposed 3D building model and uncertainty measures are evaluated using real-world LiDAR point clouds.
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION II
(2022)
Proceedings Paper
Geography, Physical
Vinu Kamalasanan, Yu Feng, Monika Sester
Summary: Collisions and safety are crucial in urban design like shared spaces. Research directions include developing intelligent mobility aids to prevent collisions, using AR devices for virtual infrastructure and traffic systems, and improving 3D object detection frameworks for better pedestrian detection.
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION IV
(2022)
Proceedings Paper
Geography, Physical
O. Wage, L. Bienzeisler, M. Sester
Summary: This article introduces a method for identifying accident-prone locations using bicycle accident data and bicycle volumes. By using an agent-based traffic simulation model to generate bicycle volumes and normalizing accidents, a risk score can be estimated to help city planners and cyclists avoid high-risk areas.
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION IV
(2022)
Proceedings Paper
Geography, Physical
Hao Cheng, Haoran Lei, Stefania Zourlidou, Monika Sester
Summary: Traffic regulators play a crucial role in influencing traffic flow and commuters' route choices. This paper proposes an automatic method based on CVAE to recognize traffic regulators at intersections using GPS data and satellite imagery from digital maps. Experimental results show that the proposed method achieves better accuracy and F1-score in detecting traffic arm rules compared to other models.
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION IV
(2022)
Proceedings Paper
Geography, Physical
J. Golze, U. Feuerhake, M. Sester
Summary: This article investigates traffic accidents in Germany by analyzing various environmental impacting factors, including road characteristics, weather information, population density, and points of interest. By utilizing open data sources such as OpenStreetMap and an open weather API, different categories of features describing the on-site situation of accidents are generated. Through k-means clustering, six clusters are identified and further analyzed in terms of their feature space and spatial distribution. The findings provide valuable insight for regional comparison across the state borders of Germany.
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION IV
(2022)