Article
Chemistry, Analytical
Gahyun Kim, Ho Gi Jung, Jae Kyu Suhr
Summary: This paper conducted a comparative study on different methods for detecting the vehicle BFQ in surveillance camera environments. Three approaches (corner-based, position/size/angle-based, and line-based) were compared, and a suggested implementation using YOLO as the base detector was proposed. Experimental results showed that the suggested implementation adequately detected the vehicle BFQ, and the three approaches were quantitatively evaluated, compared, and analyzed.
Article
Computer Science, Hardware & Architecture
Linrun Qiu, Dongbo Zhang, Yuan Tian, Najla Al-Nabhan
Summary: This study develops a target detection algorithm based on deep learning technologies, which improves recognition precision and model's convergence speed through fused edge features. The experimental results demonstrate efficient recognition rate and real-time performance of the algorithm.
JOURNAL OF SUPERCOMPUTING
(2021)
Article
Engineering, Civil
Saeed Rahmani, Asiye Baghbani, Nizar Bouguila, Zachary Patterson
Summary: Graph neural networks (GNNs) have gained popularity in the field of intelligent transportation systems (ITS) due to their ability to analyze graph-structured data. However, there is currently no comprehensive review of recent advancements and future research directions in all transportation areas. This survey provides an overview of GNN studies in ITS, exploring various applications such as traffic forecasting, demand prediction, autonomous vehicles, intersection management, parking management, urban planning, and transportation safety. It also identifies domain-specific research directions, opportunities, challenges, and previously overlooked research opportunities in edge and graph learning, multi-modal models, and unsupervised and reinforcement learning methods for developing more powerful GNNs. The survey also highlights popular baseline models and datasets for each transportation domain.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Environmental Sciences
Chenwei Deng, Donglin Jing, Zhihan Ding, Yuqi Han
Summary: This paper proposes an iterative pruning framework based on assistant distillation for object detection on remote sensing images. The framework prunes the network channels using a structured sparse pruning strategy, reducing the size of the detector, and recovers the network performance through a teacher assistant distillation model. Extensive experiments show that the method achieves an effective balance between speed and accuracy.
Article
Engineering, Civil
Youcef Djenouri, Asma Belhadi, Djamel Djenouri, Gautam Srivastava, Jerry Chun-Wei Lin
Summary: This paper introduces a novel deep learning architecture for identifying outliers in intelligent transportation systems. It explores the use of a convolutional neural network with decomposition to detect abnormal behavior in maritime data and proposes a simple and efficient fusion strategy for combining the results of different models. Experimental results demonstrate the superiority of the proposed framework in terms of accuracy metrics compared to baseline solutions.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Civil
Othman O. Khalifa, Muhammad H. Wajdi, Rashid A. Saeed, Aisha H. A. Hashim, Muhammed Z. Ahmed, Elmustafa Sayed Ali
Summary: Vehicle detection is crucial in Intelligent Transportation Systems (ITS) for ensuring road safety. However, existing solutions have limitations in terms of cost, information provided, and computational speed. This paper proposes a new method using Convolutional Neural Network and video processing techniques, and achieves promising results.
JOURNAL OF ADVANCED TRANSPORTATION
(2022)
Article
Computer Science, Artificial Intelligence
Mengran Liu, Weiwei Fang, Xiaodong Ma, Wenyuan Xu, Naixue Xiong, Yi Ding
Summary: The paper introduces a new channel pruning method, SCA and CPSCA, that achieves optimal resource consumption while maintaining accuracy.
APPLIED SOFT COMPUTING
(2021)
Article
Engineering, Civil
Zhihan Lv, Yuxi Li, Hailin Feng, Haibin Lv
Summary: The study aims to enhance the security performance of digital twins in the Cooperative Intelligent Transportation System in a deep learning environment. By combining Convolutional Neural Network with Support Vector Regression, a model is constructed and analyzed through simulation experiments. Results show that the proposed algorithm has significant advantages in security performance and data transmission speed.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Ghulam Muhammad, M. Shamim Hossain
Summary: This paper proposes light convolutional neural network (CNN) models for cognitive networking in an intelligent transportation system (ITS). The models include a 1D CNN for processing 1D temporal data and a deep tree CNN for processing image data from car camera sensors. By processing data independently on edge devices, the load and time of model execution are reduced. The proposed method achieves an accuracy of approximately 94-96% and an information density of 4.4 when tested on a publicly available facial emotion database.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Lili Geng, Baoning Niu
Summary: Deep learning has shown excellent performance in various fields, especially in image recognition and retrieval. However, the computational complexity of convolutional neural networks (CNNs) hinders their deployment on resource-limited devices. Network compression methods, such as pruning, can significantly reduce the complexity of CNNs. Nonetheless, existing pruning methods often overlook the actual output of filters and struggle to control the loss of accuracy. This paper proposes a novel pruning method called filter similarity analysis with backward pruning (FSABP), which calculates the similarity coefficients of filters and performs layer-by-layer pruning in the backward direction to effectively control accuracy loss.
Article
Computer Science, Artificial Intelligence
Deepak Ghimire, Kilho Lee, Seong-heum Kim
Summary: This study presents an efficient loss-aware automatic selection of structured pruning (LAASP) criteria for slimming and accelerating deep neural networks. The proposed technique adopts a pruning-while-training approach and automatically determines the optimal pruning rates for each layer in the network. Experimental results demonstrate the effectiveness of the proposed method.
IMAGE AND VISION COMPUTING
(2023)
Article
Engineering, Civil
Kai Wang, Aiheng Zhang, Haoran Sun, Bailing Wang
Summary: This paper investigates the application of deep-learning methods in intrusion detection in in-vehicle networks. Ten representative advanced deep-learning methods are analyzed, and their performance is compared. The study provides directions and suggestions for future research.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Civil
Tiago do Vale Saraiva, Carlos Alberto Vieira Campos, Ramon dos Reis Fontes, Christian Esteve Rothenberg, Sameh Sorour, Shahrokh Valaee
Summary: Vehicular networks, as critical components of advanced intelligent transportation systems, face challenges in meeting diverse communication requirements in complex dynamic environments. This paper proposes a new application-driven vehicular network framework using 5G network slicing and designs algorithms for heterogeneous traffic in dynamic vehicular environments. Simulation results show significant improvements in network performance compared to state-of-the-art approaches.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Computer Science, Hardware & Architecture
Kutub Thakur, Hamed Alqahtani, Gulshan Kumar
Summary: The intelligent system IDGADS is capable of quickly detecting algorithmically generated domains with 99% accuracy based on easy-to-compute features of real domain name system (DNS) traffic. It can serve as the first line of defense in a security stack for validating DNS queries.
COMPUTERS & ELECTRICAL ENGINEERING
(2021)
Article
Engineering, Electrical & Electronic
Anusha Vangala, Basudeb Bera, Sourav Saha, Ashok Kumar Das, Neeraj Kumar, Youngho Park
Summary: The study introduces a novel blockchain-enabled certificate-based authentication scheme, BCAS-VADN, for vehicle accident detection and notification in an Internet of Vehicles (IoV) environment, ensuring secure storage, transmission, and verification of accident-related data.
IEEE SENSORS JOURNAL
(2021)