Designing an Efficient Object Detection Model for Autonomous Driving Applications

Big data algorithms and applications in intelligent transportation system: A review and bibliometric analysis

Aim. Threats of a non-physical nature have a significant effect on the security of intelligent transportation systems (ITS). They may have the form of unlawful interference in the development and implementation of ITS. The creation of a trusted framework for the development and implementation of ITS is examined as in the case of intelligent water transportation systems (IWTS). Problems. By decree of the President of the Russian Federation, ITS, telecommunications and security of information processing are among the priority areas of scientific and technological development. ITS technologies, as well as those involved in the creation of trusted, secure system and application software are among the most important critical high technologies. The operation of ITS involves wide use of computerised systems that implement the latest information and telecommunication technologies, automated and automatic control technologies, artificial intelligence that can pose security threats. ITS are to be developed and operated in a trusted environment. Methods. The paper used the methodology for ensuring the security of IWTS, development of secure hardware and software platforms for secure automated systems, methods of system analysis, dependability theory, information protection, and law. Results. The paper defines the problem of creating a trusted framework for the development and implementation of IWTS, the applicable terminology is developed. The author examined the effect of IWTS on the security of critical information infrastructure (CII) and national security, developed a model of relationships between the IWTS security domains taking into account threats of physical and non-physical origin. Examples of computer incidents within IWTS that caused consequences at the national and international levels are given. The composition of the IWTS facilities attributed to CII is defined, critical processes implemented by standard CII facilities as part of the IWTS are set forth. The author lists conceptual problems of IWTS security, defines the principles of creating a trusted framework for the development and implementation of IWTS. Conclusion. Ensuring the security of IWTS against modern threats requires solving a number of problems associated with the creation of a trusted framework for the development and implementation of IWTS. For the purpose of improving the timeliness and quality of their solution, the paper proposes an intuitive terminology that reflects the subject area and helps finding a common understanding of the security domain by experts from various industries. IWTS facilities have an effect on CII security and national security in general. That is taken into account in the model of relationships between IWTS security domains, demonstrated using cases of computer incidents within IWTS that caused consequences at the national and international level. Given the above, the paper lists IWTS facilities attributed to CII and sets forth examples of critical processes implemented by standard CII facilities as part of the IWTS. The defined list of conceptual IWTS security problems take into account the growing landscape of IWTS security threats that includes insecure software, hardware and software platforms, software and hardware systems and emerging technologies. When developing the principles for the creation of a trusted IWTS development and implementation framework, the author took into account the best practice of implementing the methodology for creating national secure hardware and software platforms of CII facilities that enables the creation of secure automated systems for various applications that are based on domestically-developed solutions. The examined matters are of a systemic nature, which allows using the findings in the development and implementation of ITS in other modes of transportation.

The development of autonomous vehicle or self-driving car integrates with the wireless communication technology which would be a forward step for road transportation in the near future. The autonomous crossing of an intersection with an autonomous vehicle will play a crucial role in the future of intelligent transportation system (ITS). The fundamental objectives of this work are to manage autonomous vehicles crossing an intersection with no collisions, maintaining that a vehicle drives continuously, and to decrease the waiting time at an intersection. In this paper, a discrete model of the one-way single intersection is designed. The vehicle-to-infrastructure (V2I) communication is implemented to exchange information between a vehicle and an intersection manager which is the roadside infrastructure. The safe trajectory of autonomous vehicles for the autonomous intersection management is determined and presented by using discrete mathematics.

The automotive industry is more concerned about vehicle and road safety. Nowadays the Govt. has introduced new safety regulations to improve the vehicle and road safety. Most of the accidents are happening because of human errors, violation of traffic rules, etc. To avoid or reduce these accidents and to meet the Govt. safety regulations the automotive industries started implementing new technologies like Advanced Driver Assistance System (ADAS), Passive and Active safety system in vehicles. ADAS is the most important safety feature in modern automobiles. The key functionality of ADAS is to provide more comfort to the driver and to improve road and vehicle safety by minimizing human errors. To achieve these functionalities the system use sensors like Radar (Radio Detection and Ranging), Camera, Light Detection and Ranging (LIDAR), and Ultrasonic sensors. Based on sensor input Electronic Control Unit (ECU) delivers the information to driver in terms of warnings, alarms, etc. or system takes the longitudinal and lateral control to bring the vehicle to safer zone. Focus of this paper is to discuss the hardware and software technologies used for the development of ADAS systems and its challenges.

In recent years, computer vision technology has developed rapidly, as one of its important research directions, object detection has received widespread attention due to its high accuracy. Meanwhile, object detection has many application fields, such as intelligent transportation, medical and health, security systems, etc. Traditional object detection methods have limitations when applied to complex real-world scenarios. To improve the shortcomings of conventional methods, deep learning based object detection algorithms have significantly improved the efficiency of object detection and become a research hotspot in object detection. This article summarizes the algorithms into two-stage and one-stage object detection algorithms based on the technical processes and structural differences in handling object detection tasks. Firstly, several common two-stage and one-stage object detection algorithms and their applications in real-world scenarios are introduced. Then their data sets and algorithm performance are analyzed and compared. Finally, according to the results of the comparison, the existing problems of the two-stage object detection algorithm and the one-stage object detection method are discussed, and their future development directions are pointed out.

In recent years, with the gradual development of artificial intelligence, as one of the important branch applications, intelligent transportation is also developing rapidly, and the research of autonomous vehicles has also attracted more and more attention. With the increase in vehicle ownership and the increase in road congestion, the improvement of transportation efficiency has always been an urgent problem. With the development of autonomous vehicles, the impact of the application of autonomous vehicles on travel efficiency has become an important research subject. This article uses sumo to build an urban road network to simulate real road scenes, and to study the impact of autonomous vehicles on traffic flow by controlling the operating parameters of autonomous vehicles in the traffic flow.

Pedestrians, as vulnerable road users (VRUs), lack effective protective measures in traffic accidents, making them highly susceptible to injuries. Accurate prediction of pedestrian behavior is crucial for road safety, traffic management systems, advanced driver-assistance systems (ADAS), and autonomous vehicle development. Against this backdrop, this paper proposes a pedestrian group crossing intention prediction model that integrates spatiotemporal features to enhance the accuracy of pedestrian behavior prediction in autonomous driving scenarios. The model combines spatiotemporal features, including pedestrian pose key points, 2D positional trajectories, and group information. Experimental results on the JAADbeh and JAADall datasets demonstrate that the proposed model achieves superior performance in terms of accuracy, precision, and F1-Score. Notably, on the complex and large-scale JAADall dataset, the model achieves an accuracy of 0.82, highlighting its robustness. Furthermore, the findings reveal that incorporating pedestrian group information improves prediction accuracy, especially in group pedestrian scenarios, where this feature significantly enhances detection performance. This study provides a reliable pedestrian intention prediction framework for autonomous driving and intelligent transportation systems, while also laying a foundation for future exploration of diverse non-visual features and complex scenarios.

The presence of line-of-sight (LOS) and non-line-of-sight conditions in an intelligent transportation system (ITS) is critical, as it directly impacts the localisation and overall system performance. Distinguishing LOS, first-order and higher-order multipaths (HOMP) is hindered by the dense multipath (MP) propagation. The computational resource requirement and limited accuracy associated with traditional iterative methods have stimulated the use of deep learning (DL). This work proposes a novel, lightweight, dual limb attention (DLA) based DL network termed as “MOVENetx64”, for robust identification of MPs in ITS. DLA mechanism stacked with convolutional neural networks and long-short-term memory layers forms the prominent feature extractor, which enables feeding raw, unprocessed data into the DL pipeline. A novel alternating loss strategy is designed to nullify the effect of the extreme imbalance associated with the target output classes. Ray-tracing-based vehicular datasets for City, Suburban, and Highway are generated. A combination of time, received power, angular characteristics, and phase spectrum forms the input feature-set. The proposed MOVENetx64 showcases MP classification accuracy of (98.71%,99.41%,96.44%), and achieves the least HOMP prediction error of (0.6%,0.17%,2.28%) on the datasets. These results validate that the proposed MOVENetx64 is scalable and computationally efficient for distinguishing multipaths efficiently and enabling reliable vehicular communication in ITS.

With the rapid development of intelligent transportation systems (ITS), mission-critical vehicular services such as vehicle platooning, safety alarming, and remote driving play a key role in the future ITS. In order to support these emerging services, low-latency and high reliability transmission should be ensured. In the 4G Long Term Evolution (LTE) and 5G New Radio (NR) vehicle-to-everything (V2X) systems, it is very difficult to meet the latency requirement since group of orthogonal frequency division multiplexing (OFDM) symbols are processed in a form of the resource block. In this paper, we propose a novel low-latency packet transmission scheme for V2X systems, referred to as partial sample transmission (PST). In PST, we map the transmit information into subcarrier positions and then decode it using a small fraction of received samples at the receiver. To perform the efficient decoding, we propose a deep learning (DL)-based PST decoding. From the numerical evaluations on the V2X system, we demonstrate that the proposed PST technique outperforms conventional transmission schemes in terms of the block error rate (BLER) and the signaling latency.

Background of study: Blood cell analysis is vital for diagnosing medical conditions, but traditional manual methods are laborious and error-prone. Deep learning, especially YOLO models, offers automated solutions for medical image analysis. However, the real-world effectiveness of the latest YOLOv11 in blood cell detection is not thoroughly investigated, as general object detection improvements may not translate to biomedical images due to their unique characteristics.Aims and scope of paper: This study systematically compares YOLOv10 and YOLOv11 on a public blood cell detection dataset to assess if YOLOv11's advancements provide tangible benefits for blood cell classification. The goal is to identify the most effective model for accurate and efficient detection in microscopic images, guiding AI-driven diagnostic tool selection.Methods: Both models were trained and tested under identical conditions using the Kaggle Blood Cell Detection Dataset (RBCs, WBCs, Platelets). Images were resized to 640x640 pixels. Performance metrics included mAP (mAP@50 and mAP@50–95), Precision, Recall, F1-score, speed, model complexity, and training time.Result: YOLOv11n consistently showed higher accuracy (mAP50: 0.9279 vs. 0.9120; mAP50-95: 0.6524 vs. 0.6347), particularly for RBCs and WBCs. However, YOLOv11n had longer inference (11.35 ms/image) and postprocessing times (8.64 ms/image) compared to YOLOv10n (7.00 ms/image and 0.90 ms/image). YOLOv11n trained faster (0.311 hours vs. 0.375 hours), with a smaller model size (5.5 MB vs. 5.8 MB), fewer parameters, and reduced computational complexity.Conclusion: YOLOv11n offers superior accuracy and improved training efficiency, making it suitable for medical image object detection where precision is paramount. The increased inference and postprocessing times indicate a performance-speed trade-off. Model selection should balance these factors based on deployment context.

The rapid advancement of deep learning has significantly transformed the development of next-generation autonomous vehicles, enhancing both safety and efficiency. This paper explores the integration of deep learning techniques, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and reinforcement learning, in perception, decision-making, and control systems of autonomous vehicles. By leveraging vast datasets and real-time processing, deep learning enables precise object detection, path planning, and adaptive driving strategies. Furthermore, the implementation of sensor fusion techniques combining LiDAR, radar, and cameras enhances situational awareness, reducing the risk of accidents. Despite these advancements, challenges such as computational complexity, adversarial robustness, and ethical considerations remain key research areas. This study provides an overview of the current state-of-the-art deep learning applications in autonomous vehicles and discusses future directions toward fully autonomous, safer, and more efficient transportation systems.

The timely and precise identification of traffic signs is essential for maintaining the effectiveness and safety of contemporary roads, particularly in light of the increasing number of self-driving cars. Conventional image processing methods have faced challenges because to the intricate and fluctuating variables present in real-world settings, including various signage, erratic weather, and inconsistent illumination. This study utilizes recent breakthroughs in deep learning, particularly the YOLOv8 (You Only Look Once version 8) model, to tackle these difficulties. YOLOv8 incorporates cutting-edge neural network architectural advancements, such as an anchor-free detection methodology, adaptive spatial feature pooling, and dynamic neural configurations. In order to further increase detection efficiency and accuracy, this study presents two innovative models, YOLOv8-DH and YOLOv8-TDHSA. These models make use of improvements such decoupled heads and transformer-based self-attention mechanisms. Experimental results indicate that the suggested models substantially surpass current deep learning models, attaining enhanced performance across multiple measures, including accuracy, recall, F-score, and mean average precision (mAP). This research enhances traffic sign detecting technology, facilitating the development of safer and more intelligent transportation systems.

The emergence of quantum computing presents significant challenges for intelligent transportation systems (ITS), which rely on traditional public-key cryptography, demanding both strong security and low-latency operation. This highlights an urgent requirement for post-quantum cryptographic (PQC) algorithms to guarantee long-term protection. However, a direct integration of PQC into ITS may introduce latency and inefficiency, as static implementations cannot adapt to dynamic traffic conditions and may apply excessive security in low-risk contexts, resulting in unnecessary computational overhead, particularly within resource-constrained edge computing environments. To mitigate these issues, the PDRL-PQCS (Predictive Deep Reinforcement Learning for Post-Quantum Cryptography Selection) framework is proposed. This approach employs a Transformer-based module to generate multi-step predictions of traffic flow and computational load, thereby providing proactive contextual awareness. These predictions subsequently inform a deep reinforcement learning (DRL) agent, which dynamically selects optimal parameter configurations for the CRYSTALS-Kyber PQC algorithm in real time, enabling adaptive security-performance balancing across the heterogeneous nodes of an ITS edge computing network. Through this two-stage approach, PDRL-PQCS adapts PQC parameters within defined security bounds to optimally balance security assurance with operational efficiency. An extensive experimental evaluation assesses the framework under three representative traffic scenarios with diverse resource availability and congestion levels. Results indicate that PDRL-PQCS achieves 10%–20% higher overall performance compared to baseline methods across various challenging conditions.

Intelligent Transportation Systems (ITS) are increasingly shifting toward data-driven and artificial intelligence-based frameworks that integrate sensing, prediction, and control to enhance mobility efficiency, safety, and adaptability. This dissertation contributes to the development of next-generation ITS through three complementary studies focusing on flexible traffic monitoring, real-time traffic forecasting, and stability-aware vehicle control. The first study develops a multi-drone framework for wide-view corridor traffic analysis and origin-destination (OD) estimation. By coordinating multiple unmanned aerial vehicles, as demonstrated with two drones in this study, the framework enables continuous vehicle tracking across extended corridors. A deep learning-based pipeline combining vehicle detection and tracking automatically reconstructs vehicle trajectories and matches them across drone views. Field experiments conducted at a freeway weaving section achieved over 91% accuracy in OD flow estimation compared with ground-truth data, demonstrating the framework’s effectiveness for scalable and high-resolution traffic monitoring. The second study proposes a day-specific spatial-temporal graph convolutional network (Day-STGCN) for real-time traffic forecasting. Building upon the classical STGCN architecture, the model integrates rolling historical time-dependent features with day-type categorization to capture both short-term variations and long-term recurring patterns. Using real-world sensor data, the model significantly improves prediction accuracy over baseline graph neural networks, providing robust day-aware forecasting that reflects realistic daily traffic dynamics. The third study investigates cooperative adaptive cruise control (CACC) and its effects on traffic string stability under varying market penetration rates of connected and automated vehicles (CAVs). Analytical derivations establish explicit stability conditions linking controller parameters, time headway, and control intervals in both homogeneous and heterogeneous traffic flows. The results reveal that appropriately designed CACC controllers enhance string stability even in mixed environments with different types of vehicles. Collectively, these studies present an integrated framework that advances ITS through data-driven sensing, learning, and control, offering theoretical and practical foundations for more intelligent, responsive, and stable transportation systems.

A new method of vehicle identification in intelligent traffic command system is proposed, which is directed to some problems in the vehicle detection method of complex traffic conditions in intelligent traffic systems. The method extracts the vehicle structure sub-features according to the structural characteristics of the vehicle itself, and then combines the sub¬features. The identification of the vehicle is achieved by the identification of the combined features. Through experimental analysis, the method not only has a fast processing speed, but also has a high vehicle recognition accuracy. Therefore, this method has broad application prospects and value in intelligent traffic command system.