Article
Mathematics
Kai Zhang, Zixuan Chu, Jiping Xing, Honggang Zhang, Qixiu Cheng
Summary: This study proposes a data-driven model for predicting traffic congestion by analyzing the spatio-temporal features of traffic flow. The model utilizes the traffic zone/grid method to store the average speed of vehicles on local area roads and employs a discrete snapshot set to characterize the spatial and temporal features of traffic flow. By transforming the global urban transportation network into traffic zones, the evolution of traffic congested flow in various time dimensions is examined. The model incorporates a convolutional LSTM network to learn the temporal and local spatial features of traffic flow, while a convolutional neural network effectively captures the global spatial features. Numerical experiments on two cities' transportation networks demonstrate that the proposed model outperforms traditional traffic flow prediction models.
Article
Environmental Sciences
Huifu Li, Nengsheng Luo
Summary: This study empirically analyzes the impact of transportation infrastructure on urban carbon emissions. The results show that improving transportation infrastructure has a significant negative effect on carbon emissions, and this effect is greater than the positive spillover from an increase in motor vehicles. Additionally, the study finds that improving transportation infrastructure does not significantly affect motor vehicle purchases. Therefore, attention should be given to improving transportation infrastructure in urban planning and construction to reduce urban carbon emissions.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2022)
Article
Computer Science, Hardware & Architecture
Yijing Huang, Wanyue Wei, Yang He, Qihong Wu, Kaiming Xu
Summary: Prior research on traffic event detection has encountered two problems: limited sample numbers and unbalanced datasets. To solve these issues, this study proposes a Hybrid Deep Learning-based Automated Incident Detection and Management (HDL-AIDM) system, which uses Temporal and Spatial Stacked Autoencoder (TSSAE) and generative adversarial network (GAN) to collect temporal and spatial associations of traffic conditions and improve sample numbers and dataset balance. The suggested method achieves higher incident detection rates with high accuracy.
COMPUTERS & ELECTRICAL ENGINEERING
(2023)
Article
Transportation Science & Technology
Lejun Jiang, Tamas G. Molnar, Gabor Orosz
Summary: This paper evaluates the capability of connected roadside infrastructure to provide traffic predictions on highways based on the motion of connected vehicles, establishing metrics to quantify the amount of traffic prediction available through V2I communication. It demonstrates that even with sparsely distributed roadside units and low connected vehicle penetration rates, considerable traffic predictions can be achieved, providing strategies for maximizing traffic prediction efficiency in deploying roadside units along highways. Ultimately, the results may serve as a guideline for the evaluation and deployment of connected roadside infrastructure.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Transportation Science & Technology
Caio Vitor Beojone, Nikolas Geroliminis
Summary: The emergence of shared-economy and smartphones has enabled on-demand transportation services, creating opportunities but also complexities in urban mobility. Increasing fleet sizes of Transportation Network Companies may reduce waiting time but could worsen congestion, leading to longer total travel time.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Computer Science, Artificial Intelligence
Yuanyuan Liu, Zhen Cai, Huili Dou
Summary: The rapid development of urbanization in China has led to traffic events such as accidents and delays. Traditional methods of detecting and resolving highway congestion are slow, labor-intensive, and require a lot of monitoring equipment. Therefore, the introduction of advanced technology, specifically deep learning, is necessary to address these challenges. This paper proposes a framework based on deep learning for next-generation highway traffic management, which accurately detects and evaluates traffic congestion, and predicts possible congestion. The framework was evaluated using data from highway monitoring scenes, and the results showed high accuracy in detecting and evaluating congestion. The study demonstrates that deep learning is an effective tool for detecting and evaluating highway congestion, providing accurate and timely information for traffic management.
Article
Environmental Studies
Abdelfettah Laouzai, Rachid Ouafi
Summary: This paper proposes an enhanced approach to reduce atmospheric pollution in urban areas by considering additional constraints in traffic congestion analysis through a bi-level optimization program. The approach effectively respects eco-friendly threshold constraints, categorizes travel demand into two classes, and verifies its validity through optimality conditions. Two network examples are discussed to show that the proposed optimal solution outperforms common route choice policies in terms of reducing traffic congestion and minimizing air pollution.
ENVIRONMENT AND PLANNING B-URBAN ANALYTICS AND CITY SCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Ying Gao, Jinlong Li, Zhigang Xu, Zhangqi Liu, Xiangmo Zhao, Jianhua Chen
Summary: This study proposes a new image-based traffic congestion estimation method, which first defines the traffic congestion accurately and integrates a traffic parameter layer into a CNN model. By training and testing with a large dataset of traffic images, the proposed method shows better efficiency and stability in various traffic conditions and weather scenarios.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Green & Sustainable Science & Technology
Mingjun Ma, Meng Liu, Ziqiao Li
Summary: This study proposes a method to quantify the environmental impact of traffic diversions during road infrastructure construction processes. It reveals that different diversion plans can lead to different environmental impacts by evaluating the impact of pollutant emissions on the environment.
Article
Engineering, Civil
Mun Chon Ho, Joanne Mun-Yee Lim, Chun Yong Chong, Kah Keong Chua, Alvin Kuok Lim Siah
Summary: To alleviate traffic congestion in urban areas, a vehicle rerouting strategy specifically designed for vanet is proposed in this study. Unlike previous research, this strategy considers dynamic traffic information, such as travel time, and ensures collaborative rerouting among vehicles by updating the remaining capacity of roads. Through traffic simulations in simple grid and real-world KL networks, the proposed strategy outperforms other state-of-the-art strategies by at least 4.39% in terms of average travel time.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Jiahui Jin, Xiaoxuan Zhu, Biwei Wu, Jinghui Zhang, Yuxiang Wang
Summary: This paper proposes a method based on deep reinforcement learning to dynamically adjust online road toll based on traffic conditions and travelers' demands, aiming to resolve traffic congestion issues.
TSINGHUA SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Software Engineering
Ameni Chetouane, Sabra Mabrouk, Imen Jemili, Mohamed Mosbah
Summary: Due to the increasing number of vehicles, we investigated vehicle detection and traffic congestion classification methods and compared various methods to solve the traffic issues on the bridge of Bizerte.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2022)
Article
Chemistry, Multidisciplinary
Sachin Ranjan, Yeong-Chan Kim, Navin Ranjan, Sovit Bhandari, Hoon Kim
Summary: Traffic congestion is a significant problem worldwide, and a hybrid deep neural network algorithm based on HRNet and ConvLSTM has been proposed for traffic congestion prediction. The model outperforms four other state-of-the-art architectures in terms of accuracy, precision, and recall.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Civil
Kadiyala Ramana, Gautam Srivastava, Madapuri Rudra Kumar, Thippa Reddy Gadekallu, Jerry Chun-Wei Lin, Mamoun Alazab, Celestine Iwendi
Summary: Traffic problems are worsening due to population growth in urban areas, causing issues such as air pollution, fuel consumption, law violations, noise pollution, accidents, and time loss. Traffic prediction is crucial in smart cities to reduce congestion, but current methods are not suitable for real-world applications. In this study, Vision Transformers (VTs) were used with Convolutional Neural Networks (CNN) to accurately predict traffic flow, particularly during abnormal situations. The proposed technology outperformed traditional methods in terms of precision, accuracy, and recall, while also promoting energy conservation through rerouting.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Chemistry, Analytical
Maria V. Peppa, Tom Komar, Wen Xiao, Phil James, Craig Robson, Jin Xing, Stuart Barr
Summary: Near real-time urban traffic analysis and prediction are crucial for intelligent transport systems. This study presents an end-to-end framework using CCTV images to detect and predict urban traffic volume, incorporating novel techniques and machine learning models. The developed framework, tested at various locations under varying traffic conditions, shows that random forest and LSTM models provide accurate predictions and can potentially be implemented in real time.
Article
Engineering, Mechanical
Mauricio Arredondo-Soto, Enrique Cuan-Urquizo, Alfonso Gomez-Espinosa
Summary: This paper presents the mathematical formulation of the Compliance Matrix Method (CMM) for the kinetostatic analysis of Flexure-Based Compliant Mechanisms (FBCM) and integrates it with inverse kinematics. The effectiveness of this method is validated using two FBCPM examples and shows excellent agreement when compared with Finite Element Analysis results.
MECHANISM AND MACHINE THEORY
(2022)
Article
Computer Science, Artificial Intelligence
J. F. Ciprian-Sanchez, G. Ochoa-Ruiz, M. Gonzalez-Mendoza, L. Rossi
Summary: The researchers selected three state-of-the-art deep learning-based image fusion techniques and evaluated their performance on fire image fusion task. They also proposed an improved method for generating artificial infrared and fused images.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Fernando Camarena, Leonardo Chang, Miguel Gonzalez-Mendoza, Ricardo J. Cuevas-Ascencio
Summary: Human action recognition is an important research field that utilizes deep learning and hand-crafted features to improve classification accuracy. This study proposes a key trajectories approach that utilizes pose estimation to find meaningful key points, reducing computational time and noise. The proposed method is tested on multiple datasets.
PATTERN ANALYSIS AND APPLICATIONS
(2022)
Article
Chemistry, Analytical
Andres J. Barreto-Cubero, Alfonso Gomez-Espinosa, Jesus Arturo Escobedo Cabello, Enrique Cuan-Urquizo, Sergio R. Cruz-Ramirez
Summary: Mobile robots require an accurate map of their surroundings to navigate. To detect materials that may not be detected by a single sensor, a fusion scheme using at least two sensors is necessary. By using an artificial neural network to fuse data from three sensors, a 2D occupancy map can be generated to identify glass obstacles.
Article
Chemistry, Analytical
Josue Gonzalez-Garcia, Alfonso Gomez-Espinosa, Luis Govinda Garcia-Valdovinos, Tomas Salgado-Jimenez, Enrique Cuan-Urquizo, Jesus Arturo Escobedo Cabello
Summary: This paper presents a model-free high-order sliding mode controller with finite-time convergence. The experimental results show that the proposed controller can drive the robot to the desired trajectories in a predefined time with reduced error and energy consumption compared to traditional PID controller and the same sliding mode controller with asymptotic convergence.
Article
Computer Science, Information Systems
Yoanna Martinez-Diaz, Heydi Mendez-Vazquez, Luis S. Luevano, Miguel Nicolas-Diaz, Leonardo Chang, Miguel Gonzalez-Mendoza
Summary: Given the impact of the COVID-19 pandemic, the accuracy of current face recognition methods for masked faces has decreased, necessitating the improvement of existing technologies. This paper studies the effectiveness of three lightweight face recognition models for accurate and efficient masked face recognition, finding that fine-tuning existing models on synthetic masked images achieves better performance. Furthermore, the effectiveness of the masked-based models is evaluated on established unmasked benchmarks and compared with state-of-the-art face models to assess the efficiency of the lightweight architectures used.
Correction
Computer Science, Artificial Intelligence
Fernando Camarena, Leonardo Chang, Miguel Gonzalez-Mendoza, Ricardo J. Cuevas-Ascencio
PATTERN ANALYSIS AND APPLICATIONS
(2022)
Article
Chemistry, Analytical
Josue Gonzalez-Garcia, Alfonso Gomez-Espinosa, Luis Govinda Garcia-Valdovinos, Tomas Salgado-Jimenez, Enrique Cuan-Urquizo, Jesus Arturo Escobedo Cabello
Summary: This paper validates the feasibility of a model-free high-order sliding mode controller for the station-keeping problem in AUVs. The proposed controller exhibits robust performance against unknown disturbances and maintains a small root mean square error (RMSE) even in the presence of strong ocean currents. The evaluation was done through numerical simulations and experiments, demonstrating the effectiveness of the controller.
Editorial Material
Chemistry, Multidisciplinary
Hiram Ponce, Lourdes Martinez-Villasenor, Miguel Gonzalez-Mendoza, Pablo A. Fonseca
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Raymundo Romero-Arenas, Alfonso Gomez-Espinosa, Benjamin Valdes-Aguirre
Summary: This study focuses on the application of singing voice detection in different music genres. It adapts a Long-Term Recurrent Convolutional Network (LRCN) to detect vocals in a dataset of electronic music and compares its performance with other state-of-the-art experiments in pop music. The results provide a benchmark for evaluating electronic music and its intricacies.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Analytical
Josue Gonzalez-Garcia, Nestor Alejandro Narcizo-Nuci, Alfonso Gomez-Espinosa, Luis Govinda Garcia-Valdovinos, Tomas Salgado-Jimenez
Summary: This paper proposes a synchronous navigation scheme for two BlueROV2 underwater vehicles to perform a coordinated multi-vehicle task without vehicle-to-vehicle communication. The scheme utilizes a model-free second-order sliding mode controller with finite-time convergence, which is user-defined and independent of the vehicle's physical or hydrodynamic parameters. Simulation experiments demonstrate the controller's robust performance, even in the presence of high ocean currents, without the need for parameter readjustment.
Article
Chemistry, Multidisciplinary
Jesus Dassaef Lopez-Barrios, Jesus Arturo Escobedo Cabello, Alfonso Gomez-Espinosa, Luis-Enrique Montoya-Cavero
Summary: In this paper, a Mask R-CNN is used to improve the performance of machine vision in detecting peduncles and fruits of green sweet peppers in greenhouses. The ResNet-101 + FPN network is adopted for feature extraction and object representation enhancement. The proposed implementation manages to segment the peduncle and fruit of the green sweet pepper in real-time.
APPLIED SCIENCES-BASEL
(2023)
Review
Mathematics, Interdisciplinary Applications
Fernando Camarena, Miguel Gonzalez-Mendoza, Leonardo Chang, Ricardo Cuevas-Ascencio
Summary: The rapid advancement of artificial intelligence has enabled various applications such as intelligent video surveillance systems, assisted living, and human-computer interaction. One of the core tasks for these applications is video-based human action recognition. However, the extensive and ongoing research in this field makes it challenging to assess the full scope of available methods and current trends. This survey aims to concisely explore the field of vision-based human action recognition, defining core concepts, explaining common challenges, and providing an overview of commonly used datasets. Additionally, it presents a comprehensive review of literature approaches and their evolution over time, emphasizing intuitive notions.
MATHEMATICAL AND COMPUTATIONAL APPLICATIONS
(2023)
Article
Automation & Control Systems
Carmina Perez-Guerrero, Jorge Francisco Ciprian-Sanchez, Adriana Palacios, Gilberto Ochoa-Ruiz, Miguel Gonzalez-Mendoza, Vahid Foroughi, Elsa Pastor, Gerardo Rodriguez-Hernandez
Summary: This paper proposes a method that uses Generative Adversarial Networks to generate infrared images from visible ones, reducing the cost of jet fire experiments and enabling other potential applications. The feasibility of the approach is validated by comparing the measurements obtained from real and generated infrared images.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2024)
Proceedings Paper
Computer Science, Cybernetics
Carlos Vilchis, Miguel Gonzalez-Mendoza, Leonardo Chang, Sergio A. Navarro-Tuch, Gilberto Ochoa Ruiz, Isaac Rudomin
Summary: This article discusses techniques and methods to improve our perception of digital humans and analyzes available frameworks for replication and analysis. It also provides an overview analysis of present-day environments for digital humans, aiming to determine future research objectives in this field.
PROCEEDINGS OF THE 17TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS (HUCAPP), VOL 2
(2022)
