arXiv subjects

André Luiz Cunha

Publications and source records attributed to André Luiz Cunha.

7 recordsLinked to original sources

arXiv2025

ORDENA: ORigin-DEstiNAtion data exploration

Analyzing origin-destination flows is an important problem that has been extensively investigated in several scientific fields, particularly by the visualization community. The problem becomes especially challenging when involving massive data, demanding mechanisms such as data aggregation and interactive filtering to make the exploratory process doable. However, data aggregation tends to smooth out certain patterns, and deciding which data should be filtered is not straightforward. In this work, we propose ORDENA, a visual analytic tool to explore origin and destination data. ORDENA is built upon a simple and intuitive scatter plot where the horizontal and vertical axes correspond to origins and destinations. Therefore, each origin-destination flow is represented as a point in the scatter plot. How the points are organized in the plot layout reveals important spatial phenomena present in the data. Moreover, ORDENA provides explainability resources that allow users to better understand the relation between origin-destination flows and associated attributes. We illustrate ORDENA's effectiveness in a set of case studies, which have also been elaborated in collaboration with domain experts. The proposed tool has also been evaluated by domain experts not involved in its development, which provided quite positive feedback about ORDENA.

cs.GR↗

arXiv2025

Evoluindo resiliência em rotas de ônibus: Proposta de um método para a maximização de acessibilidade em cenários de incerteza por meio de algoritmo genético

Resilience has raised interest in transport planning as rare phenomena, such as fuel supply crises, have recently shown their potential to destabilize transport systems. However, the proposed methods for planning resilience in transit systems fail to consider the impact that bus frequency has on user accessibility. To address this gap, this paper proposes a bus allocation method aimed at maximizing accessibility in impact scenarios - where some bus routes have their frequency reduced - making use of a genetic algorithm. The method is applied in the city of São Paulo and the results show that evolving the system foreseeing moderate impacts not only contributes to reducing the negative effects of lower route frequency, but also improves its efficiency in normal conditions, showing the importance of the contribution of this research to the planning of efficient systems.

physics.soc-ph↗

arXiv2023

Resilience in Highways: Proposal of Roadway Redundancy Indicators and Application in Segments of the Brazilian Network

With the growing realization that transport systems must operate satisfactorily not only in typical situations, but also in adverse circumstances, ensuring redundancies in road systems has gained crucial importance. In this context, several methods have been proposed for measuring the vulnerabilities and resilience of transport systems. However, a simple metric to understand and quantify the degree of redundancy of a given road segment is still necessary, mainly to guide the responsible bodies regarding the need for intervention or special care with certain sections of the system. Thus, this paper proposes a redundancy indicator based on network analyses in the vicinity of an element. The proposed indicator was first calculated on nine application examples and then on a substantial sample of the Brazilian road network (~10% of segments). The results demonstrate that the indicator can satisfactorily describe the variety of cases in the Brazilian network, capturing cases where there is significant redundancy in the elements, as in some regions of the Southeast and South; or cases of very low redundancy, such as the sparse grid in the north of the country. It was also verified that the indicator has a particular sensitivity to parameters of the defined function, requiring further research for an acceptable calibration.

physics.soc-ph↗

arXiv2023

Um banco de dados de empregos formais georreferenciados em cidades brasileiras

Currently, transport planning has changed its paradigm from projects oriented to guarantee service levels to projects oriented to guarantee accessibility to opportunities. In this context, a number of studies and tools aimed at calculating accessibility are being made available, however these tools depend on job location data that are not always easily accessible. Thus, this work proposes the creation of a database with the locations of formal jobs in Brazilian cities. The method uses the RAIS jobs database and the CNEFE street faces database to infer the location of jobs in urban regions from the zip code and the number of non-residential addresses on street faces. As a result, jobs can be located more accurately in large and medium-sized cities and approximately in single zip code cities. Finally, the databases are made available openly so that researchers and planning professionals can easily apply accessibility analyzes throughout the national territory.

cs.CY↗

arXiv2023

Truck Axle Detection with Convolutional Neural Networks

Axle count in trucks is important to the classification of vehicles and to the operation of road systems. It is used in the determination of service fees and in the impact on the pavement. Although axle count can be achieved with traditional methods, such as manual labor, it is increasingly possible to count axles using deep learning and computer vision methods. This paper aims to compare three deep-learning object detection algorithms, YOLO, Faster R-CNN, and SSD, for the detection of truck axles. A dataset was built to provide training and testing examples for the neural networks. The training was done on different base models, to increase training time efficiency and to compare results. We evaluated results based on five metrics: precision, recall, mAP, F1-score, and FPS count. Results indicate that YOLO and SSD have similar accuracy and performance, with more than 96\% mAP for both models. Datasets and codes are publicly available for download.

cs.CV↗

arXiv2020

Measuring urban road network resilience to extreme events: an application for urban floods

Designing and maintaining resilient transportation systems rely on identifying potential vulnerabilities and inefficiencies before crises occur. However, given the complexity of transportation networks, as well as the diversity of ways in which systems can fail, the problem of assessing the impacts of exceptional phenomena still lacks tools and metrics for analysis. Therefore, this paper aims to present a method for assessing the behavior of road networks during crises in which segments are disabled or unusable for a time. The method proposed was structured around two metrics calculated from the road network and trip distribution: network continuity and efficiency of alternative. With these metrics, we assess, as an example, the local and global impacts of various flooding scenarios in a case study at São Carlos, a medium-sized Brazilian city in the state of São Paulo. Our findings indicate that pedestrian movements are significantly less impacted than motorized vehicles by floods in the city, which may be due to the car-oriented nature of the avenues closest to rivers, and the shortest average length of pedestrian trips. These findings and metrics set a framework for further research involving other cities and natural phenomena to enrich the understanding of resilience in urban transportation networks.

physics.soc-ph↗

arXiv2019

Classification of Motorcycles using Extracted Images of Traffic Monitoring Videos

Due to the great growth of motorcycles in the urban fleet and the growth of the study on its behavior and of how this vehicle affects the flow of traffic becomes necessary the development of tools and techniques different from the conventional ones to identify its presence in the traffic flow and be able to extract your information. The article in question attempts to contribute to the study on this type of vehicle by generating a motorcycle image bank and developing and calibrating a motorcycle classifier by combining the LBP techniques to create the characteristic vectors and the classification technique LinearSVC to perform the predictions. In this way the classifier of vehicles of the type motorcycle developed in this research can classify the images of vehicles extracted of videos of monitoring between two classes motorcycles and non-motorcycles with a precision and an accuracy superior to 0,9.

cs.CV↗