2021
Autores
Abrantes, D; Maria Campos Ferreira, M; Costa, P; Felicio, S; Hora, J; Dangelo, C; Silva, J; Teresa Galvão Dias, M; Coimbra, M;
Publicação
Human Systems Engineering and Design (IHSED2021) Future Trends and Applications - AHFE International
Abstract
2022
Autores
Reza, S; Ferreira, MC; Machado, JJM; Tavares, JMRS;
Publicação
EXPERT SYSTEMS WITH APPLICATIONS
Abstract
Traffic flow forecasting is an essential component of an intelligent transportation system to mitigate congestion. Recurrent neural networks, particularly gated recurrent units and long short-term memory, have been the stateof-the-art traffic flow forecasting models for the last few years. However, a more sophisticated and resilient model is necessary to effectively acquire long-range correlations in the time-series data sequence under analysis. The dominant performance of transformers by overcoming the drawbacks of recurrent neural networks in natural language processing might tackle this need and lead to successful time-series forecasting. This article presents a multi-head attention based transformer model for traffic flow forecasting with a comparative analysis between a gated recurrent unit and a long-short term memory-based model on PeMS dataset in this context. The model uses 5 heads with 5 identical layers of encoder and decoder and relies on Square Subsequent Masking techniques. The results demonstrate the promising performance of the transform-based model in predicting long-term traffic flow patterns effectively after feeding it with substantial amount of data. It also demonstrates its worthiness by increasing the mean squared errors and mean absolute percentage errors by (1.25 - 47.8)% and (32.4 - 83.8)%, respectively, concerning the current baselines.
2022
Autores
Ferreira, MC; Costa, PD; Abrantes, D; Hora, J; Felicio, S; Coimbra, M; Dias, TG;
Publicação
TRANSPORTATION RESEARCH PART F-TRAFFIC PSYCHOLOGY AND BEHAVIOUR
Abstract
The continuous growth of the world population and its agglomeration in urban cities, demand an increasing need for mobility, which in turn contributes to the worsening of traffic congestion and pollution in cities. Therefore, it is necessary to promote active travel, such as walking and cycling. However, this is not an easy task, as pedestrians and cyclists are the most vulnerable link in the system, and low levels of safety, security and comfort can contribute to choosing private cars over active travel. Hence, it is essential to understand the determinants that affect the perceptions of pedestrians and cyclists, in order to support the definition of policies that promote the use of active modes of transport. Thus, this article fills an important gap in the literature by identifying and discussing the objective and subjective determinants that affect the perceptions of safety, security and comfort of pedestrians and cyclists, through a systematic review of the literature published in the last ten years. It followed the PRISMA statement guidelines and checklist, resulting in 68 relevant articles that were carefully analyzed. The results show that the perception of safety is negatively affected by fear of traffic-related injuries, fear of falling related to infra-structure and infrastructure maintenance, and negative behavior of drivers. Regarding security, crime was the major concern of pedestrians and cyclists, either with emphasis on the person or on personal property. With regard to comfort, high levels of air and noise pollution, lack of vege-tation, bad weather conditions, slopes and long commuting distances negatively affected the users' perception. The results also suggest that poor lighting affects all domains, providing a negative perception of safety, security and comfort. Similarly, the presence of people is seen as negatively influencing the perception of safety and comfort, while the absence of people nega-tively impacts the perception of security. Therefore, the findings achieved by this study are key to assist in the definition of transport policies and infrastructure creation in large smart cities. Additionally, new transport policies are proposed and discussed.
2022
Autores
Reza, S; Ferreira, MC; Machado, JJM; Tavares, JMRS;
Publicação
EXPERT SYSTEMS WITH APPLICATIONS
Abstract
Speech recognition aims to convert human speech into text and has applications in security, healthcare, commerce, automobiles, and technology, just to name a few. Inserting residual neural networks before recurrent neural network cells improves accuracy and cuts training time by a good margin. Furthermore, layer normalization instead of batch normalization is more effective in model training and performance enhancement. Also, the size of the datasets presents tremendous influences in achieving the best performance. Leveraging these tricks, this article proposes an automatic speech recognition model with a stacked five layers of customized Residual Convolution Neural Network and seven layers of Bi-Directional Gated Recurrent Units, including a logarithmic so f tmax for the model output. Each of them incorporates a learnable per-element affine parameter-based layer normalization technique. The training and testing of the new model were conducted on the LibriSpeech corpus and LJ Speech dataset. The experimental results demonstrate a character error rate (CER) of 4.7 and 3.61% on the two datasets, respectively, with only 33 million parameters without the requirement of any external language model.
2023
Autores
Abrantes, D; Ferreira, MC; Costa, PD; Hora, J; Felício, S; Dias, TG; Coimbra, M;
Publicação
International journal of environmental research and public health
Abstract
Due to an increase in population, urban centers are currently seeing an increase in traffic, resulting in negative consequences such as pollution and congestion. Efforts have been made to promote a modal shift towards the use of more sustainable means of transport, such as walking and cycling, but several deterrents influence the citizens' perceptions of safety, security and comfort, discouraging their choice of active modes of transport. This study focuses on the importance of providing meaningful information to vulnerable road users (VRUs) to support their perceptions and objectives while moving within urban spaces through a novel concept of route planning. A broad survey of the needs and concerns of VRUs through interviews, focus groups and questionnaires, applied to the Portuguese population of the Metropolitan Area of Porto, led to the development of a new concept of route planners that show personalized routes according to the individual perceptions of each user. This concept is materialized in a route planner prototype that has been extensively tested by potential users. Subjective evaluation and feedback showed the usefulness of the concept and added value to a familiar product, leading to a satisfying experience for participants. This study shows that there is an opportunity to improve these tools to provide a higher degree of power and customization to users on route planning, which includes addressing mobility restrictions and personal perceptions of safety, security and comfort. The ultimate goal of this new approach is to persuade citizens to switch to more sustainable means of transport.
2023
Autores
Barros, D; Ferreira, MC; Silva, AR;
Publicação
Advances in Transportation Studies
Abstract
Nowadays, cities face severe problems related to traffic management and mobility in general. Therefore, technologies have been developed that can handle these situations and somehow mitigate the caused impact, such as CCTV cameras. However, the techniques for analyzing the images collected by these cameras are increasingly complex and have numerous applications, being dispersed in the literature. Therefore, this article fills an important research gap by presenting a systematic review of the literature on the possible applications of data collected from CCTV cameras and the image analysis and processing techniques that have been developed and proposed in recent years. This systematic review followed the PRISMA statement guidelines and checklist, and three databases were searched, namely Scopus, Web of Science, and Inspec. From the analysis performed, the following applications were identified: Image/video analysis and traffic estimation, pedestrian detection, traffic data analysis, and forecasting, and traffic management. Regarding the image analysis and processing techniques YOLO (only look once), GMM (Gaussian mixture method), morphological methods, fuzzy logic, and other proprietary methods stand out. After a thorough analysis of traffic data, most works still implemented relatively trivial traffic management systems to generate a series of actions to be eventually applied to traffic controllers. Additionally, it was realized that these techniques could be implemented in industrial products from a future perspective. © 2023, Aracne Editrice. All rights reserved.
The access to the final selection minute is only available to applicants.
Please check the confirmation e-mail of your application to obtain the access code.