Article
Engineering, Civil
Yanchao Dong, Senbo Wang, Jiguang Yue, Ce Chen, Shibo He, Haotian Wang, Bin He
Summary: The paper introduces a novel Visual SLAM method that effectively utilizes texture-less object instances for mapping and localization. The method includes newly designed feature extraction, matching, localization, and mapping modules, which jointly use object features and point features to estimate camera 6-DOF poses and create richer maps. The advantages of the proposed Visual SLAM method are demonstrated through experiments on both synthetic and real-world datasets.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Transportation
SeJoon Park, Wooyeon Yu
Summary: This study examines the impact of different system parameters on the efficiency of carsharing operations by comparing round-trip carsharing systems with one-way carsharing systems. A regression model is developed to predict the lost sales rate of a one-way carsharing system based on various operational conditions. The findings provide insights into understanding the performance of carsharing systems and predicting their efficiency when modifications occur.
TRANSPORTATION LETTERS-THE INTERNATIONAL JOURNAL OF TRANSPORTATION RESEARCH
(2022)
Article
Green & Sustainable Science & Technology
Joanna Drobiazgiewicz, Agnieszka Pokorska
Summary: One of the main challenges in transportation policy is reducing the reliance on passenger cars in favor of sustainable and shared mobility options. The authors focused on analyzing the carsharing phenomenon and its relevance in Poland, examining the potential for carsharing to complement other transportation modes. The research involved literature review, analysis of carsharing companies' activities, public data analysis, and observations.
Article
Chemistry, Analytical
Dorin Palanciuc, Florin Pop
Summary: The DOORS system is designed for simplicity and efficiency in edge computing, expected to scale up to hundreds of nodes. It encapsulates application state and behavior into objects and allows them to exchange asynchronous messages. The system provides semi-synchronous replication and the capability to move objects between nodes, aiming for scalability and resilience. The paper outlines the system structure, describes object replication in DOORS, and presents a basic set of measurements for design improvements.
Article
Chemistry, Multidisciplinary
Boonyarit Changaival, Kittichai Lavangnananda, Gregoire Danoy, Dzmitry Kliazovich, Frederic Guinand, Matthias Brust, Jedrzej Musial, Pascal Bouvry
Summary: This article proposes a novel optimization model for the round-trip carsharing fleet placement problem, proves its NP-hard nature, investigates different optimization algorithms, and validates the results using real instances.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Geological
Mario Valiante, Domenico Guida, Marta Della Seta, Francesca Bozzano
Summary: LOOM (landslide object-oriented model) is proposed as a data structure for landslide inventories, aiming to effectively store and manipulate complex spatial and temporal relations between landslides. It uses hierarchical classification and aggregation levels to handle spatial relations, while temporal characterization is achieved by assigning exact dates or time ranges to each object. Tested in the Cilento UNESCO Global Geopark (Italy), the model proves powerful for inventorying landslides and reconstructing gravity-induced deformation history, leading to prediction of evolution.
Article
Computer Science, Artificial Intelligence
Yatie Xiao, Chi-Man Pun, Bo Liu
Summary: Deep learning excels at complex tasks, but Deep Neural Networks are vulnerable to carefully crafted adversarial perturbations. The AO(2)AM algorithm focuses on object-level adversarial perturbations to fool deep neural object detection networks effectively.
PATTERN RECOGNITION
(2021)
Article
Environmental Sciences
Yunfei Han, Ping Wang, Yongguo Zheng, Muhammad Yasir, Chunmei Xu, Shah Nazir, Md Sakaouth Hossain, Saleem Ullah, Sulaiman Khan
Summary: This study utilizes object-oriented classification to extract landslide data from high-resolution remote sensing data and explores the impact of geology, lithology, rainfall, and human activities on landslide occurrence. The study found a Kappa coefficient of 0.76, landslide extraction accuracy of 79.8%, and an overall classification accuracy of 87%. The causes of landslides are discussed and early warning information for unknown landslides can be obtained through feature analysis.
Article
Chemistry, Analytical
Nadia Nedjah, Alexandre V. Cardoso, Yuri M. Tavares, Luiza de Macedo Mourelle, Brij Booshan Gupta, Varsha Arya
Summary: The template matching technique is used in this study to find patterns in images. A hardware coprocessor is designed for the computationally demanding step of template matching. Six different swarm intelligence techniques are investigated to accelerate the target search process. The results show that the PSO-based search strategy performs the best in terms of processing time and accuracy.
Article
Geochemistry & Geophysics
Minjian Zhang, Heqian Qiu, Hefei Mei, Lanxiao Wang, Fanman Meng, Linfeng Xu, Hongliang Li
Summary: In this paper, a novel anchor-free oriented object detection network called DRDet is proposed, which adopts dual-angle rotated lines (DRLs) as object representation. By introducing DRL, the network can adaptively rotate and extend to the boundary of the object, explicitly incorporating orientation information into the formulation of object representation. Furthermore, an orientation-guided feature encoder (OFE) and a dual-angle decoder (DD) are designed to enhance the flexibility and performance of DRLs. Experimental results demonstrate consistent improvement in oriented object detection using the proposed method.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Gong Cheng, Yanqing Yao, Shengyang Li, Ke Li, Xingxing Xie, Jiabao Wang, Xiwen Yao, Junwei Han
Summary: This article presents a two-stage oriented object detection method, called dual-aligned oriented detector (DODet), to address the spatial and feature misalignment problems. DODet uses an oriented proposal network to generate high-quality proposals and a localization-guided detection head to alleviate the feature misalignment between classification and localization. Extensive evaluations on multiple benchmarks show consistent and substantial improvements compared to baseline methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Joanna Zawadzka, Ian Truckell, Abdou Khouakhi, Monica Rivas Casado
Summary: The study utilized UAV-captured high resolution imagery for rapid assessment of flood damage, automatically detecting debris associated with residential housing damages through image segmentation and classifiers, achieving a good level of accuracy.
Article
Environmental Studies
Peraphan Jittrapirom, Saroch Boonsiripant, Monthira Phamornmongkhonchai
Summary: The study utilized a participatory group modeling approach to explore stakeholders' mental models regarding carsharing operation in Bangkok, Thailand, highlighting the differences in vision and understanding. By creating a causal loop diagram to represent a shared understanding, the research identified factors influencing the success of carsharing and potential policy interventions. The results demonstrated the effectiveness of the participatory approach in transport planning.
TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT
(2021)
Article
Geochemistry & Geophysics
Qiaolin Zeng, Xiang Ran, Hao Zhu, Yanghua Gao, Xinfa Qiu, Liangfu Chen
Summary: This study proposes a method to address the issue of complicated hand-designed components in object detection using the detection transformer (DETR) framework. By utilizing the D-angle module, adaptive proposal selection, and adaptive query selection, the proposed method effectively solves the problems of capturing directional objects in remote sensing images, slow convergence of DETR, and attention allocation in the decoder.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Engineering, Civil
Giulio Giorgione, Dzmitry Kliazovich, Luca Bolzani, Francesco Ciari, Francesco Viti
Summary: This study proposes an experimental method to maximize profit in carsharing services through dynamic pricing strategies. The results indicate that profit is heavily dependent on supply, with little impact from variability in potential demand.
JOURNAL OF ADVANCED TRANSPORTATION
(2022)
Article
Management
Minh Hoang Ha, Hoa Nguyen Phuong, Huyen Tran Ngoc Nhat, Andre Langevin, Martin Trepanier
Summary: This article studies the clustered traveling salesman problem with a prespecified order on the clusters. In this problem, delivery locations are divided into clusters with different urgency levels, and more urgent locations must be visited first. However, this may result in an inefficient route in terms of traveling cost. The article proposes two approaches to solve this problem and demonstrates their effectiveness through experiments.
INTERNATIONAL TRANSACTIONS IN OPERATIONAL RESEARCH
(2022)
Article
Engineering, Civil
Cen Zhang, Jan-Dirk Schmocker, Martin Trepanier
Summary: The study proposes a new model based on Markov Chains to predict the monthly usage frequency of members in a car-sharing scheme. By including five latent user 'life stages' and validating the model on panel data, the effectiveness of the model in predicting car-sharing usage frequency is demonstrated. This approach is effective for predicting usage in novel transport schemes.
Article
Computer Science, Artificial Intelligence
Kevin Pasini, Mostepha Khouadjia, Allou Same, Martin Trepanier, Latifa Oukhellou
Summary: The study aims to detect the impact of disturbances on a transportation network through smart card data analysis, focusing on contextual anomaly detection using machine learning models to build a robust anomaly score. The research highlights the importance of variance normalization on prediction residuals under a dynamic context, showing the relevance in building a reliable anomaly score.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Chemistry, Analytical
Aliasghar Mehdizadeh Dastjerdi, Catherine Morency
Summary: This study focuses on short-term demand prediction for bike-sharing services in Montreal using a deep learning approach. Results show that deep learning models outperform traditional ARIMA models, and the addition of extra input features improves prediction accuracy.
Article
Engineering, Civil
Jerome Laviolette, Catherine Morency, Owen D. Waygood, Konstadinos G. Goulias
Summary: This paper studies the spatial dependencies of household car ownership rates in the Island of Montreal and finds that sociodemographic and built environment variables have a significant impact on car ownership rates, with failure to control for spatial dependence resulting in an overestimation of the direct influence of built environment variables.
TRANSPORTATION RESEARCH RECORD
(2022)
Article
Operations Research & Management Science
Zhanhong Cheng, Martin Trepanier, Lijun Sun
Summary: This paper proposes a method for short-term OD matrix forecasting based on DMD and low-rank high-order VAR model. By reconstructing the problem as a data-driven regression model and using a tailored online update algorithm for updating model coefficients, the method shows robustness in handling noisy and sparse data and outperforms baseline models in predicting OD matrices and boarding flow.
TRANSPORTATION SCIENCE
(2022)
Article
Green & Sustainable Science & Technology
Simona Zapolskyte, Martin Trepanier, Marija Burinskiene, Oksana Survile
Summary: Currently, there is no established method to assess the level of urban smartness. This article develops a hierarchical evaluation model to assess smart city transportation systems using relevant and influential indicators. The evaluation and comparison are conducted using a hybrid multi-criteria decision-making method. The study identifies and ranks a leading city based on its level of smartness.
Article
Economics
Elodie Deschaintres, Catherine Morency, Martin Trepanier
Summary: This paper evaluates the value of traditional surveys and emerging data in cross-analyzing the temporal variability of travel behaviors. The results show that day-to-day variability in travel behavior can be accurately inferred from a single-day survey, and demonstrate the complementarity and potential combination of the two data sources.
TRANSPORTATION RESEARCH PART A-POLICY AND PRACTICE
(2022)
Article
Transportation Science & Technology
Remi Decouvelaere, Martin Trepanier, Bruno Agard
Summary: This study presents a spatiotemporal clustering tool that allows for adjusting the importance of space and time. By testing different parameter values, it is found that the influence of space and time can be controlled and the obtained clusters vary depending on whether one or both dimensions are considered.
Article
Public, Environmental & Occupational Health
Camille Garnier, Martin Trepanier, Catherine Morency
Summary: The research aims to estimate the number of people in Quebec eligible for paratransit services and to estimate the latent demand for the STM paratransit service in Montreal. A filter algorithm is used to determine eligibility and compare it with current service usage, revealing a significant latent eligible population.
JOURNAL OF TRANSPORT & HEALTH
(2022)
Article
Chemistry, Analytical
Miratul Khusna Mufida, Abdessamad Ait El Cadi, Thierry Delot, Martin Trepanier, Dorsaf Zekri
Summary: This study addresses the challenge of developing accurate and efficient parking occupancy forecasting models for autonomous vehicles at the city level. A novel two-step clustering technique is proposed to group parking lots based on their spatiotemporal patterns, allowing for the development of accurate occupancy forecasting models for a set of parking lots. Real-time parking data was used to build and evaluate the models, demonstrating the effectiveness of the proposed strategy in reducing model deployment costs and improving model applicability and transfer learning across parking lots.
Article
Engineering, Civil
Nazmul Arefin Khan, Catherine Morency
Summary: The COVID-19 pandemic in 2020 has significantly changed daily mobility patterns worldwide. A web-based survey was conducted in Montreal, Canada in April-May 2020 to capture the impacts of the pandemic on travel behaviors. Using data from this survey, this paper proposes insights into how people are planning to travel in a post-COVID-19 world. The study identifies two latent segments, suburbanite and urbanite people, and finds considerable heterogeneity across sample individuals.
TRANSPORTATION RESEARCH RECORD
(2023)
Article
Engineering, Civil
Hubert Verreault, Catherine Morency
Summary: Most transportation organizations have collected a large amount of data through travel surveys, which is used for transportation planning and modeling. However, relying on outdated data is becoming problematic, especially for less populous areas. This paper proposes a methodology to combine different survey samples to better utilize historical travel survey data.
TRANSPORTATION RESEARCH RECORD
(2023)
Article
Operations Research & Management Science
Xiaoxu Chen, Zhanhong Cheng, Jian Gang Jin, Martin Trepanier, Lijun Sun
Summary: This paper proposes a Bayesian probabilistic model for forecasting bus travel time and estimated time of arrival (ETA). The model can capture interactions between buses, handle missing values, and depict the multimodality in bus travel time distributions. Using time-varying mixing coefficients, it can infer systematic temporal variations in bus operation.
TRANSPORTATION SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Zhanhong Cheng, Xudong Wang, Xinyuan Chen, Martin Trepanier, Lijun Sun
Summary: This study explores the relationship between vehicle speed and density on the road, highlighting biases in calibrating single-regime speed-density models using least-squares method. By modeling the covariance of residuals with zero-mean Gaussian Process, a new calibration method is proposed that significantly reduces biases, achieves similar effects as the weighted least-squares method, functions as a non-parametric speed-density model, and provides a Bayesian solution for estimating posterior distributions.
IEEE OPEN JOURNAL OF INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
