Search

Traffic sign detection is critical for autonomous driving technology. However, accurately detecting traffic signs in complex traffic environments remains challenge despite the widespread use of one‐stage detection algorithms known for their real‐time processing capabilities. In this paper, the authors propose a traffic sign detection method based on YOLO v8. Specifically, this study introduces the Space‐to‐Depth (SPD) module to address missed detections caused by multi‐scale variations of traffic signs in traffic scenes. The SPD module compresses spatial information into depth channels, expanding the receptive field and enhancing the detection capabilities for objects of varying sizes. Furthermore, to address missed detections caused by complex backgrounds such as trees, this paper employs the Select Kernel attention mechanism. This mechanism enables the model to dynamically adjust its focus and more effectively concentrate on key features. Additionally, considering the uneven distribution of training data, the authors adopted the WIoUv3 loss function, which optimizes loss calculation through a weighted approach, thereby improving the model's detection performance across various sizes and frequencies of instances. The proposed methods were validated on the CCTSDB and TT100K datasets. Experimental results demonstrate that the authors' method achieves substantial improvements of 3.2% and 5.1% on the mAP50 metric compared to YOLOv8s, while maintaining high detection speed, significantly enhancing the overall performance of the detection system. The code for this paper is located at https://github.com/dusongjie/TSD‐YOLO‐Small‐Traffic‐Sign‐Detection‐Based‐on‐Improved‐YOLO‐v8 [ABSTRACT FROM AUTHOR]

Traffic signs detection is an important and challenging task in intelligent driving perception system. This paper proposes an improved lightweight traffic signs detection framework based on YOLOv5. Firstly, the YOLOv5's backbone is replaced with ShuffleNet v2, which simplifies the calculation complexity and reduces the parameters of backbone network. Secondly, aiming at the problem of inconspicuous traffic sign characteristics in complex road environment, we use the CA attention mechanism in this paper to improve the saliency of the object. Finally, aiming at the large-scale difference between the traffic signs and the high proportion of small objects, we design the BCS-FPN to fuse multi-scale features and improve the representation ability of the small-scale objects. The TT-100K dataset is also analyzed and the dataset is collated. We test on the collated TT-100K dataset for the improved YOLOv5 in this paper. And the results show that compared with YOLOv5s, the mAP of our algorithm is equivalent to that of YOLOv5s, and the speed is improved by 20.8%. This paper also has carried on the experiment on embedded devices, experimental results show that our framework in computing power less embedded devices has a better effect. [ABSTRACT FROM AUTHOR]

Introduction: Advanced traffic monitoring systems face significant challenges in vehicle detection and classification. Conventional methods often require substantial computational resources and struggle to adapt to diverse data collection methods. Methods: This research introduces an innovative technique for classifying and recognizing vehicles in aerial image sequences. The proposed model encompasses several phases, starting with image enhancement through noise reduction and Contrast Limited Adaptive Histogram Equalization (CLAHE). Following this, contourbased segmentation and Fuzzy C-means segmentation (FCM) are applied to identify foreground objects. Vehicle detection and identification are performed using EfficientDet. For feature extraction, Accelerated KAZE (AKAZE), Oriented FAST and Rotated BRIEF (ORB), and Scale Invariant Feature Transform (SIFT) are utilized. Object classification is achieved through a Convolutional Neural Network (CNN) and ResNet Residual Network. Results: The proposed method demonstrates improved performance over previous approaches. Experiments on datasets including Vehicle Aerial Imagery from a Drone (VAID) and Unmanned Aerial Vehicle Intruder Dataset (UAVID) reveal that the model achieves an accuracy of 96.6% on UAVID and 97% on VAID. Discussion: The results indicate that the proposed model significantly enhances vehicle detection and classification in aerial images, surpassing existing methods and offering notable improvements for traffic monitoring systems. [ABSTRACT FROM AUTHOR]

Summary: In the ever‐evolving realm of cloud computing, the challenge of intrusion detection has grown increasingly intricate and vital. With the proliferation of cyber‐attacks and the widespread use of virtualized environments, there is a pressing need for network security solutions that are not only innovative and robust but also easily comprehensible. These solutions must possess the ability to effectively detect malicious activities, provide visibility into network operations, adapt to changing requirements, and promptly alert stakeholders to any suspicious behavior. In this study, we introduce a groundbreaking approach known as fuzzy logic‐driven optimized long short‐term memory (LSTM)—FLOLSTM, specifically designed for hypervisor‐based environments. By integrating fuzzy logic with an optimized LSTM neural network, FLOLSTM aims to significantly enhance the detection of attack traffic within hypervisor networks by combining the interpretability and uncertainty management capabilities of fuzzy logic with the temporal pattern recognition prowess of LSTM. The research methodology involves meticulous data collection from hypervisor monitors, followed by rigorous cleaning and preprocessing to enhance data reliability. Subsequently, the preprocessed data is input into the FLOLSTM classifier to identify malicious activities. Furthermore, the performance of the LSTM is fine‐tuned using the waterwheel plant optimization (WPO) algorithm. Experimental evaluations compare the efficiency of the proposed FLOLSTM with existing techniques across various metrics including accuracy, recall, precision, F‐measure, specificity, false‐positive rate (FPR), and false‐negative rate (FNR). Overall, the FLOLSTM model represents a significant advancement in intrusion detection for cloud environments, offering a potent blend of interpretability, accuracy, and efficiency. Its superior performance underscores its capacity to enhance network security and effectively mitigate cyber threats in dynamic and virtualized settings, thereby making a substantial impact on the field of cyber security. [ABSTRACT FROM AUTHOR]

The unmanned aerial vehicle (UAV) city patrol is of great significance in ensuring the safety of residents' lives and properties, as well as maintaining the normal operation of the city. However, the detection of UAV images faces challenges such as numerous small-scale objects, complex backgrounds, and high requirements for detection speed. In response to these issues, we introduce a Real-time Small Object Detection network in UAV-vision (RTS-Net), tailored for UAV patrols. Initially, we introduce a multiscale feature fusion module (MFFM) designed to augment the expressiveness of features across scales, thereby enhancing the detection of smaller objects. Subsequently, leveraging attention mechanisms, we present the coordinated attention detection module (CADM), which bolsters the detection model's ability to accurately segregate objects from the background in expansive, complex scenarios. Lastly, a lightweight real-time feature extraction module (RFEM) is crafted to diminish model computational complexity and boost inference speed. On the UAV road patrol image dataset we constructed, our proposed method attains a detection accuracy of 89.9 % mAP, breaking previous records. It surpasses all prevailing detection methods, particularly for small-scale objects. Simultaneously, it achieves an inference speed of 163.9 FPS. The experimental results show that RTS-Net can satisfy the accurate and efficient detection of ground objects by various different UAV platforms in different complex scenarios. [ABSTRACT FROM AUTHOR]

Train noise is a kind of green, non‐destructive and strong‐energy artificial seismic sources, which is widely used in railway safety monitoring, near‐surface imaging and urban underground space exploration. Distributed acoustic sensing is a new seismic acquisition technology, which has the advantages of dense sampling, simple deployment and strong anti‐electromagnetic interference ability. In recent years, distributed acoustic sensing has been gradually applied in the fields of urban traffic microseism monitoring, crack detection and underground space imaging. However, previous studies mainly focused on microseism interferometry using train event coda noise, and there is limited research on the workflow of interferometry imaging using distributed acoustic sensing–based heavy train events noise (with short coda windows), which produces an abundant of near‐source interference. Aiming at proving the effectiveness of this idea, we investigated a process workflow to get underground shear‐velocity structure based on distributed acoustic sensing recorded heavy traffic noise near Qinhuangdao train station. A weighted sliding absolute average method is used to weaken the strong amplitude to the coda wave level and reduce the near‐source influence. We demonstrated that the cross‐coherence interferometry method, after spectral whitening, has the best effect on sidelobe suppression in the virtual source surface wave shot gathers, through a comparative analysis of cross‐correlation and cross‐coherence results. For obtaining concentrated energy and strong continuity in phase velocity spectra, we selected the time windows with high spatial coherence and signal‐to‐noise ratio not less than 1.2 for stacking from 720 time windows in F–K domain. When dividing subarrays to extract pseudo‐two‐dimensional profile, we set the overlap rate between adjacent time windows to 80% to increase stacking times, enhancing the precision of phase velocity spectra and reducing the errors of picking dispersion curve. Our results show that heavy traffic train events noise (non‐pure coda) can be used to detect underground velocity structure with clear dispersion and high inversion reliability. This research provides a new processing flow for distributed acoustic sensing train noise imaging and can be applied in future urban underground space exploration. [ABSTRACT FROM AUTHOR]

As a fundamental element of the transportation system, traffic signs are widely used to guide traffic behaviors. In recent years, drones have emerged as an important tool for monitoring the conditions of traffic signs. However, the existing image processing technique is heavily reliant on image annotations. It is time consuming to build a high-quality dataset with diverse training images and human annotations. In this paper, we introduce the utilization of Vision–language Models (VLMs) in the traffic sign detection task. Without the need for discrete image labels, the rapid deployment is fulfilled by the multi-modal learning and large-scale pretrained networks. First, we compile a keyword dictionary to explain traffic signs. The Chinese national standard is used to suggest the shape and color information. Our program conducts Bootstrapping Language-image Pretraining v2 (BLIPv2) to translate representative images into text descriptions. Second, a Contrastive Language-image Pretraining (CLIP) framework is applied to characterize not only drone images but also text descriptions. Our method utilizes the pretrained encoder network to create visual features and word embeddings. Third, the category of each traffic sign is predicted according to the similarity between drone images and keywords. Cosine distance and softmax function are performed to calculate the class probability distribution. To evaluate the performance, we apply the proposed method in a practical application. The drone images captured from Guyuan, China, are employed to record the conditions of traffic signs. Further experiments include two widely used public datasets. The calculation results indicate that our vision–language model-based method has an acceptable prediction accuracy and low training cost. [ABSTRACT FROM AUTHOR]

In autonomous driving, environmental perception technology often encounters challenges such as false positives, missed detections, and low accuracy, particularly in detecting small objects and complex scenarios. Existing algorithms frequently suffer from issues like feature redundancy, insufficient contextual interaction, and inadequate information fusion, making it difficult to perform multi-task detection and segmentation efficiently. To address these challenges, this paper proposes an end-to-end multi-task environmental perception model named YOLO-Mg, designed to simultaneously perform traffic object detection, lane line detection, and drivable area segmentation. First, a multi-stage gated aggregation network (MogaNet) is employed during the feature extraction process to enhance contextual interaction by improving diversity in the channel dimension, thereby compensating for the limitations of feed-forward neural networks in contextual understanding. Second, to further improve the model's accuracy in detecting objects of various scales, a restructured weighted bidirectional feature pyramid network (BiFPN) is introduced, optimizing cross-level information fusion and enabling the model to handle object detection at different scales more accurately. Finally, the model is equipped with one detection head and two segmentation heads to achieve efficient multi-task environmental perception, ensuring the simultaneous execution of multiple tasks. The experimental results on the BDD100K dataset demonstrate that the model achieves a mean average precision (mAP50) of 81.4% in object detection, an Intersection over Union (IoU) of 28.9% in lane detection, and a mean Intersection over Union (mIoU) of 92.6% in drivable area segmentation. The tests conducted in real-world scenarios show that the model performs effectively, significantly enhancing environmental perception in autonomous driving and laying a solid foundation for safer and more reliable autonomous driving systems. [ABSTRACT FROM AUTHOR]

Accurate mapping is crucial for the successful deployment of unmanned ground vehicles (UGVs) in navigation tasks. However, maps derived from as-designed models often lack realism and fail to reflect dynamic changes on construction sites. To address this issue, a premapping method leveraging prior construction progress video is proposed to create up-to-date navigation maps. This method proactively accounts for dynamic changes, particularly those caused by transient objects that are present only during the progress video collection phase and may not exist at the time of UGV deployment. The proposed approach involves detecting, tracking, and segmenting these transient objects, followed by their removal using a flow-guided model. The cleaned frames are used to reconstruct an updated point cloud, which serves as the basis for defining a two-dimensional (2D) navigation map. By considering these changes during the premapping process, the method ensures that the resulting map is consistent with the actual site conditions. Validation in an indoor construction setting demonstrated the necessity of the proposed method and highlighted the impacts of transient objects on path planning outcomes. This method is beneficial for routine monitoring scenarios, for which it can update and transform prior video into a navigation map, enabling efficient advanced planning. Practical Applications: The practical application of this study primarily lies in enhancing the deployment efficiency of unmanned ground vehicles for routine monitoring of construction sites. The proposed premapping method, which leverages prior construction progress video, offers two key practical benefits. Firstly, as an offsite premapping approach, it enables the creation of maps without interfering with ongoing construction activities, minimizing on-site operations. This enhances UGV acceptance on site and maintains regular site workflows. Secondly, by proactively considering the dynamic changes caused by transient objects, the method ensures that the created map is up-to-date and provides a more comprehensive basis for efficient initial planning. This premapping approach is particularly beneficial for routine monitoring scenarios, because it reduces operational efforts by providing effective and up-to-date maps derived from construction progress video. By doing so, it enhances both the efficiency of UGV deployment and the effectiveness of advanced planning in dynamic construction environments. [ABSTRACT FROM AUTHOR]

This study examines the publication strategies of legacy news outlets on Twitter following the company's acquisition by Elon Musk. We used SCImago Media Rankings to identify the news organizations with the highest online audiences, namely The Guardian, The New York Times, USA Today, The Independent, and The Washington Post. Data was acquired through the Twitter Academic API v2 and the SEMrush platform at the beginning of 2023 and covers all of 2022. The database includes website audits that monitor the web traffic of a domain, with the aggregate data subsequently processed using R packages and Atlas.ti. Our analyses show that a higher presence on Twitter leads to a substantive increase in social traffic, as seen with The Independent. This is in sharp contrast to the other news outlets. These differences remain irrespective of their topical news coverage, broadly focused on the Russo-Ukrainian war, or compositional feature in their messages, which largely lack typical Twitter features like hashtags and @-mentions. We conclude our analysis by discussing the communication strategies of news organizations on Twitter following the company's acquisition by Elon Musk in 2022. [ABSTRACT FROM AUTHOR]

Traffic control through video/image processing is a trending research topic within Intelligent Transportation Systems. The number of moving vehicles, the density of vehicles at the junctions, traffic anomalies, and traffic flow directly influence real-time traffic congestion. The primary goal of this work is to design an architecture that is simple and fast enough to use in real-time heterogeneous traffic scenes. The proposed methodology uses user-defined zones for determining vehicle occupancy and count based on traffic surveillance video. The angular integral projection function accompanies the background subtraction method for better localization. Traffic anomaly detection is implemented by incorporating intelligent sensor technology. Also, the skipping of frames at fixed intervals has increased the processing speed with minimal data loss. The compatibility with any Internet Protocol camera further distinguishes this approach from other state-of-the-art methods. The proposed algorithm has been tested on a publicly available traffic surveillance video dataset and real-time surveillance feeds. Experimental results from the real-time intersection scenes show an overall accuracy of 97.14% and an average processing speed of 92.91%. The accuracy of the detection module in the custom-made dataset is 98.73%. [ABSTRACT FROM AUTHOR]

Timely and accurate detection and recognition of traffic lights are critical for Autonomous Vehicles (AVs) to avoid crashes due to red light running. This paper integrates a new robust machine learning based solution by combining a Convolutional Neural Network (CNN) with computer vision techniques to achieve a real-time traffic light detector. The proposed detection and recognition algorithm is capable of recognizing traffic lights on low-power small-form platforms, which are lightweight, portable, and can be mounted on AVs in daylight scenarios. The LISA open-source dataset is utilized with augmentation methods to increase the accuracy of the solution. The proposed approach achieves 93.42% of accuracy at a speed of 30.01 Frames Per Second (FPS) on an NVIDIA Jetson Xavier platform without using hardware accelerators such as FPGA. This solution is expected to promote the quicker adoption and wider deployment of AVs by increasing the chances of avoiding crashes and ultimately saving lives. [ABSTRACT FROM AUTHOR]

Real-time traffic data is fundamental for active traffic monitoring and control. Traditionally, traffic data are collected using location-based sensors and spatial sensors. However, both sensors have well-known limitations due to installation, operations, maintenance costs, and environmental factors. This study develops a methodology to use Wi-Fi sensors for traffic state characterization on urban roads to overcome these limitations. We examine the received signal strength indicator (RSSI) patterns and identify three distinct RSSI signature patterns. These patterns are used to develop methodologies to estimate (a) Whether the position of the end of the queue is upstream or downstream of the detector, (b) Whether the traffic conditions in the vicinity of the detector are uniformly uncongested or uniformly congested, and (c) The maximum queue length and the time is taken for the queue to grow to the maximum extent. The estimates from the methodology are validated with empirical data that showed good concurrence with field conditions, and the methods proposed in this article have the potential to estimate the traffic conditions using sparse data from Wi-Fi sensors. [ABSTRACT FROM AUTHOR]

With the growth in services supplied over the internet, network infrastructure has become more exposed to cyber-attacks, particularly Distributed Denial of Service (DDoS) attacks, which can easily cause the disruption of services. The key factor for fighting against these attacks is the earlier separation and detection of the traffic in networks. In this paper, a novel approach, named Half Autoencoder-Stacked DNNs (HAE-SDNN) model, is proposed. We suggest using a Stacked Deep Neural Networks (SDNN) model. as a deep learning model, in order to detect DDoS attacks. Our approach allows feature selection from a preprocessed dataset using a Half AutoEncoder (HAE), resulting in a final set of important features. These features are subsequently used to train the DNNs that are stacked together by applying Softmax layer to combine their outputs. Experiments were performed on a benchmark cybersecurity dataset, named CICDDoS2017, containing various DDoS attack types. The experimental results demonstrate that the introduced model attained an overall accuracy rate of 99.95%. Moreover, the HAE-SDNN model outperformed existing models, highlighting its superiority in accurately classifying attacks. [ABSTRACT FROM AUTHOR]

Autonomous vehicles have limited computational resources and thus require efficient control systems. The cost and size of sensors have limited the development of self-driving cars. To overcome these restrictions, this study proposes an efficient framework for the operation of vision-based automatic vehicles; the framework requires only a monocular camera and a few inexpensive radars. The proposed algorithm comprises a multi-task UNet (MTUNet) network for extracting image features and constrained iterative linear quadratic regulator (CILQR) and vision predictive control (VPC) modules for rapid motion planning and control. MTUNet is designed to simultaneously solve lane line segmentation, the ego vehicle's heading angle regression, road type classification, and traffic object detection tasks at approximately 40 FPS for 228 × 228 pixel RGB input images. The CILQR controllers then use the MTUNet outputs and radar data as inputs to produce driving commands for lateral and longitudinal vehicle guidance within only 1 ms. In particular, the VPC algorithm is included to reduce steering command latency to below actuator latency, preventing performance degradation during tight turns. The VPC algorithm uses road curvature data from MTUNet to estimate the appropriate correction for the current steering angle at a look-ahead point to adjust the turning amount. The inclusion of the VPC algorithm in a VPC-CILQR controller leads to higher performance on curvy roads than the use of CILQR alone. Our experiments demonstrate that the proposed autonomous driving system, which does not require high-definition maps, can be applied in current autonomous vehicles. [ABSTRACT FROM AUTHOR]

In the fast modernised world, usage of vehicles increased day by day, leads to vehicle traffic and congestions. Traditional way of monitoring traffic is less efficient and requires man power. Besides, safety of traffic controllers is the major concern in the manual monitoring. For that reason, effective prediction of vehicular traffic and monitoring the level of congestion is significant in VANET (Vehicular Adhoc Network) for mitigating the delays and danger of accidents. In order to make clear prediction of vehicles and the collision free path in the network, numerous existing algorithms have been provided for prediction and classification. However, the conventional techniques have faced complications in satisfying the accuracy while making predictions of path images. The accuracy in prediction along with the faster computational time have been addressed as a drawback which hinders in making appropriate travel path decisions. In order to overcome such challenges, proposed system employed Yolo v5 algorithm, GGA (Greedy based Genetic Algorithm) and Deep CNN (Convolutional Neural Network) with LSTM (Long Short Term Memory) for traffic monitoring and congestion control in VANET. The Yolo v5 algorithm is utilised for the vehicle prediction mechanism, for the capability of detection with high speed and accuracy. GGA is utilised for the feature selection for the capability of handling numerous features in data and to increase the computational speed. The Deep CNN with LSTM is utilised for Classification for the capability of handling larger datasets and to enhance the accuracy. Though, DCNN is the effective algorithm for classification, it has few limitations like slow processing time. To resolve this, DCNN is utilised with LSTM for enhancing the speed in the projected system. Vehicle detection dataset is used in the proposed system. Further, the experimental evaluations and comparative analysis exhibits the efficiency of the system in terms of accuracy, precision, recall and F1 score when compared to several existing algorithms. The proposed model holds the potential in efficient vehicular congestion management with expected accuracy and it is intended to contribute in to the path identification for vehicles in VANET applications. [ABSTRACT FROM AUTHOR]

Small-object detection has become a hot issue in complex traffic scenes. A global context multilevel fusion attention detection method for small-object detection is proposed in this paper. First, a global context feature fusion network model is designed with cross-stage partial DarkNet (CSPDarkNet) as the backbone to capture the global context semantic information and refine the local information. To further refine the local information, a hierarchical hybrid attention module is designed that uses global average pooling to obtain the H direction weight matrix, fuse it with the W direction weight matrix, fuse the results with the channel direction weight matrix, and finally obtain a feature map with multidimensional weights. Second, to increase the receptive field of the multidimensional weighted feature map, atrous convolution is added in the spatial pyramid pooling (SPP) module. To improve the small-object detection accuracy, a 160 × 160 small-object detection head is added. Finally, the Focal-EIOU (efficient intersection over union) loss function is adopted to better converge in the training process. Full experiments have been carried out in the open traffic datasets Cityscapes and KITTI, and this paper proposed model is the best-performing method. [ABSTRACT FROM AUTHOR]

With the rapid promotion of autonomous driving technology, it is extremely important to scientifically anticipate the related technologies and analyze their possible impact on urban road systems. The accuracy of detection and localization of traffic elements of autonomous driving is closely related to the ability of autonomous driving devices to make control decisions and the safety of autonomous driving. The study designs a new high-speed road driving scheme based on autonomous driving by analyzing the challenges related to urban traffic that may be brought about by unmanned driving. On the basis of the faster R-CNN algorithm, the context information around the target is introduced to locate and detect small-scale traffic signs. A new pedestrian detection model is designed, which is based on the feature pyramid network and introduces the SE module to highlight the features of the visible part of the pedestrian and reduce the missed detection rate caused by inter-class occlusion. The improved traffic sign detection framework improves the detection accuracy by 18.91% compared to the original faster R-CNN, while the enhanced pedestrian inspection method improves the detection accuracy by 14.00%. For both traffic sign detection and pedestrian detection accuracy and speed are improved compared to the original method. [ABSTRACT FROM AUTHOR]

Seizures produce autonomic symptoms, mainly sympathetic but also parasympathetic in origin. Within this context, the vagus nerve is a key player as it carries information from the different organs to the brain and vice versa. Hence, exploiting vagal neural traffic for seizure detection might be a promising tool to improve the efficacy of closed-loop Vagus Nerve Stimulation. This study developed a VENG detection algorithm that effectively detects seizures by emphasizing the loss of spontaneous rhythmicity associated with respiration in acute intrahippocampal Kainic Acid rat model. Among 20 induced seizures in six anesthetized rats, 13 were detected (sensitivity: 65%, accuracy: 92.86%), with a mean VENG-detection delay of 25.3 ± 13.5 s after EEG-based seizure onset. Despite variations in detection parameters, 7 out of 20 seizures exhibited no ictal VENG modifications and remained undetected. Statistical analysis highlighted a significant difference in Delta, Theta and Beta band evolution between detected and undetected seizures, in addition to variations in the magnitude of HR changes. Binomial logistic regression analysis confirmed that an increase in delta and theta band activity was associated with a decreased likelihood of seizure detection. This results suggest the possibility of distinct seizure spreading patterns between the two groups which may results in differential activation of the autonomic central network. Despite notable progress, limitations, particularly the absence of respiration recording, underscore areas for future exploration and refinement in closed-loop stimulation strategies for epilepsy management. This study constitutes the initial phase of a longitudinal investigation, which will subsequently involve reproducing these experiments in awake conditions with spontaneous recurrent seizures. [ABSTRACT FROM AUTHOR]

This study aims to examine, ascertain, and measure the extent of research conducted on real-time surveillance video analytics (RTSVA) and uncover the patterns, advancements, and progression in the SCOPUS database. An electronic search methodology was employed to identify the most relevant scholarly papers. Each analytical tool was analyzed using the R program code and VOS Viewer software. Multiple factors were explored, including co-authorship, co-citation, conceptual structure, co-word occurrence, trend themes analysis, thematic map, and visualization analysis. It is observed the study includes data from 653 sources, such as journals, books, and conference proceedings. There are 1041 documents, with an annual growth rate of 21%. IEEE Access has the highest impact, and Multimedia Tools and Applications has the second highest source impact. The most significant author among the research group is Chen Y, with a fractionalization value of 6.59 (27 articles), followed by Wang J (15 articles) with a value of 3.80. The keywords related to the video analytics domain along with its frequency of occurrence, are "Edge computing," which has appeared 158 times, and video surveillance has appeared 95 times. The Thematic map is divided into four quadrants: Niche, Motors, Emerging, and Basic themes. Traffic surveillance, blockchain, and COVID-19 are under Niche themes. It is also necessary to explore the sociological and ethical implications linked to using video surveillance systems. This study has the potential to aid both novice and experienced researchers in identifying novel research domains, suitable sources, and opportunities for collaboration. The results pertaining to relational and evaluate methodologies could be helpful to the new researchers. [ABSTRACT FROM AUTHOR]

With the rapid advancement of the Internet of Things, network security has garnered increasing attention from researchers. Applying deep learning (DL) has significantly enhanced the performance of Network Intrusion Detection Systems (NIDSs). However, due to its complexity and "black box" problem, deploying DL-based NIDS models in practical scenarios poses several challenges, including model interpretability and being lightweight. Feature selection (FS) in DL models plays a crucial role in minimizing model parameters and decreasing computational overheads while enhancing NIDS performance. Hence, selecting effective features remains a pivotal concern for NIDSs. In light of this, this paper proposes an interpretable feature selection method for encrypted traffic intrusion detection based on SHAP and causality principles. This approach utilizes the results of model interpretation for feature selection to reduce feature count while ensuring model reliability. We evaluate and validate our proposed method on two public network traffic datasets, CICIDS2017 and NSL-KDD, employing both a CNN and a random forest (RF). Experimental results demonstrate superior performance achieved by our proposed method. [ABSTRACT FROM AUTHOR]

As the primary variable load of bridges, vehicle load is an important parameter for bridge health monitoring. However, traditional Weigh-in-Motion (WIM) systems and the commonly used method of placing sensors on the bridge are challenging to apply in load monitoring for many small and medium-sized bridges. Therefore, this paper proposes a bridge vehicle load identification method based on traffic surveillance video data. Leveraging the surveillance video data on the bridge, without introducing additional hardware devices, the displacement of target points is detected through sub-pixel level image detection algorithms, enabling non-contact measurement of bridge structural response through imaging. A spatiotemporal relationship model of structural displacement, vehicle load, and load distribution is established to solve for vehicle load. Finally, model bridge tests under various loading conditions and engineering practice experiments are conducted to validate the feasibility of the method. The results of the model bridge tests show that the structural displacement measured using traffic video measurement has a deviation of less than 10% compared to the measurements obtained using contact displacement sensors (LVDT), and it can accurately reflect the displacement characteristics of the structure. The results of the field tests demonstrate that the average estimation deviation for heavy vehicle loads ranging from 12 to 18 tons is approximately 18%, meeting the engineering requirements. The proposed method can provide load statistical information for the extensive health monitoring of small and medium-sized bridges and offer a new technical pathway for obtaining bridge load information. [ABSTRACT FROM AUTHOR]

Due to the impact of scale variation of vehicle targets and changes in traffic environments in large-scale traffic monitoring systems, vehicle target detection methods often face challenges. To improve the adaptability of detection methods to these variations, we proposed an enhanced YOLOv7 for traffic systems (ETS-YOLOv7). To mitigate the effects of complex environments, we introduced the convolutional block attention module (CBAM) into the YOLOv7 framework, which filters important features in both channel and spatial dimensions, thereby enhancing the model's capability to recognize traffic object features. To address the influence of aspect ratio variations in vehicle targets, we replaced the original complete intersection over union (CIoU) with wise intersection over union v3 (WIoUv3), eliminating the aspect ratio consistency loss and enhancing the model's ability to generalize and its overall performance. Additionally, we employed the compact layer aggregation networks (CLAN) module to replace the efficient layer aggregation networks (ELAN) module, reducing redundant computations and improving computational efficiency without compromising model accuracy. The proposed method was validated on the large-scale traffic monitoring dataset UA-DETARC, achieving a mean average precision (mAP0.5–0.95) of 90.2%, which is a 3% improvement over the original YOLOv7. The frames per second (FPS) reached 149, demonstrating that the proposed model is highly competitive in terms of detection efficiency and vehicle detection accuracy compared to other advanced object detection methods. [ABSTRACT FROM AUTHOR]

In intelligent traffic monitoring systems, the significant distance between cameras and their non-overlapping fields of view leads to several issues. These include incomplete tracking results from individual cameras, difficulty in matching targets across multiple cameras, and the complexity of inferring the global trajectory of a target. In response to the challenges above, a deep learning-based vehicle tracking algorithm called FairMOT-MCVT is proposed. This algorithm con-siders the vehicles' characteristics as rigid targets from a roadside perspective. Firstly, a Block-Efficient module is designed to enhance the network's ability to capture and characterize image features across different layers by integrating a multi-branch structure and depth-separable convolutions. Secondly, the Multi-scale Dilated Attention (MSDA) module is introduced to improve the feature extraction capability and computational efficiency by combining multi-scale feature fusion and attention mechanisms. Finally, a joint loss function is crafted to better distinguish between vehicles with similar appearances by combining the trajectory smoothing loss and velocity consistency loss, thereby considering both position and velocity continuity during the optimization process. The proposed method was evaluated on the public UA-DETRAC dataset, which comprises 1210 video sequences and over 140,000 frames captured under various weather and lighting conditions. The experimental results demonstrate that the FairMOT-MCVT algorithm significantly enhances multi-target tracking accuracy (MOTA) to 79.0, IDF1 to 84.5, and FPS to 29.03, surpassing the performance of previous algorithms. Additionally, this algorithm expands the detection range and reduces the deployment cost of roadside equipment, effectively meeting the practical application requirements. [ABSTRACT FROM AUTHOR]

The social internet of vehicle (SIoV) is a specialized network combining intelligent sensing devices and vehicular communications to address traffic monitoring and resource management challenges in smart cities. Ensuring efficient and sustainable green computing with global network stability is crucial, especially in the dynamic environment of vehicular mobility. The software-defined-SIoV (SD-SIoV) architecture separates control and forwarding planes for centralized management. The architecture addresses green traffic data dissemination with heterogeneous traffic data by formulating control plane nodes' election as an NP-Hard optimization problem, considering parameters, e.g., transmission distance, node's residual energy, load imbalance, and mobility factor. The architecture incorporates the random way-point mobility (RWPM) model for simulating nodes' mobility. The proposed improved energy-efficient gray wolf optimization (IEEGWO) algorithm enhances energy-efficiency by intelligently electing and re-electing optimal control plane nodes, jointly addressing load imbalance and fault-tolerance issues, ultimately improving green computing and communication performance in SD-SIoV. Comparative analysis with state-of-the-art demonstrates that IEEGWO provides significant green computing benefits in a real-time SIoV scenario [ABSTRACT FROM AUTHOR]

For air traffic controllers, the extent of their cognitive workload can significantly impact their cognitive function and response time, consequently influencing their operational efficiency or even resulting in safety incidents. In order to enhance the accuracy and efficiency in determining the cognitive workload of air traffic controllers, a cognitive workload detection method for air traffic controllers based on mRMR and fewer EEG channels was proposed in this study. First of all, a set of features related to gamma waves was initially proposed; subsequently, an EEG feature evaluation method based on the mRMR algorithm was employed to pinpoint the most relevant indicators for the detection of the cognitive workload. Consequently, a model for the detection of the cognitive workload of controllers was developed, and it was optimized by filtering out channel combinations that exhibited higher sensitivity to the workload using the mRMR algorithm. The results demonstrate that the enhanced model achieves the accuracy and stability required for practical applications. Notably, in this study, only three EEG channels were employed to achieve the highly precise detection of the cognitive workload of controllers. This approach markedly increases the practicality of employing EEG equipment for the detection of the cognitive workload and streamlines the detection process. [ABSTRACT FROM AUTHOR]

As an indispensable part of the vehicle environment perception task, road traffic marking detection plays a vital role in correctly understanding the current traffic situation. However, the existing traffic marking detection algorithms still have some limitations. Taking lane detection as an example, the current detection methods mainly focus on the location information detection of lane lines, and they only judge the overall attribute of each detected lane line instance, thus lacking more fine-grained dynamic detection of lane line attributes. In order to meet the needs of intelligent vehicles for the dynamic attribute detection of lane lines and more perfect road environment information in urban road environment, this paper constructs a fine-grained attribute detection method for lane lines, which uses pixel-level attribute sequence points to describe the complete attribute distribution of lane lines and then matches the detection results of the lane lines. Realizing the attribute judgment of different segment positions of lane instances is called the fine-grained attribute detection of lane lines (Lane-FGA). In addition, in view of the lack of annotation information in the current open-source lane data set, this paper constructs a lane data set with both lane instance information and fine-grained attribute information by combining manual annotation and intelligent annotation. At the same time, a cyclic iterative attribute inference algorithm is designed to solve the difficult problem of lane attribute labeling in areas without visual cues such as occlusion and damage. In the end, the average accuracy of the proposed algorithm reaches 97% on various types of lane attribute detection. [ABSTRACT FROM AUTHOR]

Accurate detection of vehicles at nighttime is essential for transportation monitoring and management. However, annotating nighttime vehicle data is challenging, and vehicle features differ significantly between day and night, introducing difficulties in nighttime detection using pretrained models trained on daytime data. In this study, the nighttime vehicle detection performance is improved by employing a patchwise contrastive learning technique to enhance the representation of informative features for various traffic instances. An object detection network with reduced computational complexity and hyperparameters is utilized to conduct vehicle detection at night. Extensive experiments have been performed using images acquired from a section of Jingshi Road in Jinan, China. The impacts of learning rates and crop sizes are discussed. Three commonly adopted indicators, including mean average precision (mAP), precision, and recall, have been used to evaluate the training performance of the adopted FreeAnchor detector. Experimental results indicate that using a crop size of 320 and a learning rate of 2e-4, the developed generative adversarial network (GAN) achieves the best performance in image translation. Moreover, with a ratio of 60% real images to 40% fake images in model training, the FreeAnchor detector achieves the highest mAP of 96.6%. Visualized results for both image translation and nighttime vehicle detection demonstrate improved performance, underscoring the effectiveness of the proposed framework. This study paves the way for leveraging GAN-based networks to assist in vehicle detection under nighttime conditions. [ABSTRACT FROM AUTHOR]

This study proposes an embedded traffic sign detection system, YOLOV5-MCBS, based on an enhanced YOLOv5 algorithm. This system aims to mitigate the impact of traditional target detection algorithms' high computational complexity and low detection accuracy on traffic sign detection performance, thereby improving accuracy and real-time performance. Our primary objective is to develop a lightweight network that enhances detection accuracy, enabling real-time detection on embedded systems. First, to minimize computation and model size, we replaced the original YOLOv5 algorithm's backbone feature network with a lightweight MobileNetV3 network. Subsequently, we introduced the convolutional block attention module into the neck network to optimize the feature fusion stage's attention and enhance model detection accuracy. Concurrently, we employed the bidirectional feature pyramid network in the neck layer for multi-scale feature fusion. Additionally, we incorporated a small target detection layer into the original network output layer to enhance detection performance. What's more, we transplanted the enhanced algorithm into a Raspberry Pi embedded system to validate its real-time detection performance. Finally, we conducted computer simulations to assess our algorithm's performance by comparing it with existing target detection algorithms. Experimental results suggest that the enhanced algorithm achieves an average precision mean (mAP @ 0.5) value of 95.3% and frames per second value of 91.1 on the embedded system. [ABSTRACT FROM AUTHOR]

In order to facilitate communication, wireless networks are built from a collection of nodes that may be either static or dynamic. They are acquiring a lot of popularity in the area of research due to the fact that they are ad hoc in nature, and the number of users of mobile devices is rising day by day. Because of the ease with which these networks may be deployed in challenging and unsupervised rural places, the exchange of information has been a reality since the invention of these networks. Mobile ad hoc networks are simple to set up because of the properties that allow for self-organization and the fact that the medium is wireless. A lack of centralized fixed infrastructure, flexibility to frequent change in topologies, and other features like these are some of the other things that draw people's attention to wireless networks. Wireless networks are vulnerable to a wide range of assaults since their nodes are able to move around and their topologies are constantly changing. In addition, MANET operates in an environment that is both open and dynamic, which leaves it subject to a variety of threats from other types of network assaults. Routing protocols are almost always the target of one form or another of the same general category of attacks. Eavesdropping, causing damage, changing routing information, deleting routing information, manipulating information, advertising phoney routes, and misrouting information are all potential components of these assaults. The circumstances may make it difficult to maintain confidentiality in any communications. There are many different kinds of assaults, and each one may damage wireless networks on a different tier of the communication stack and bring the performance of the network down. Eavesdropping, jamming, traffic analysis and monitoring, denial of service attacks, grey hole attacks, black hole attacks, and wormhole assaults are a few examples of the many sorts of attacks that fall under this category. Ad-hoc networks are more susceptible to security breaches than traditional wired and wireless networks due to the usage of open wireless medium, dynamic topology, and dispersed and cooperative channel sharing. The wormhole attack on dispersed wireless networks is being described here by the person who conducted this study. Because this assault is so potent, it is very difficult to identify it before it has ever been launched. The invader may simply initiate it without having knowledge of the network or compromising any authorized nodes, which is a need for launching it. During a wormhole attack, a malicious node in one part of the network takes control of the packets and tunnels them to another hostile node in a different part of the network, which then repeats the packets locally. The thesis aims to do two things at the same time: (a) To simulate a variety of possible wormhole assaults on the MANET network (b) To investigate the functionality and efficiency of the proposed secure routing protocol within the context of these simulated attacks on the network. [ABSTRACT FROM AUTHOR]

[Display omitted] • MCTracker integrates motion correlation and mix-scale feature enhancement to track small, dense, and blurry objects effectively. • CDFE, IFMCM, and MSEB structures enhance contrast encoding, motion correlation, and multi-scale feature integration. • STFM resolves temporal-spatial fusion inconsistencies, boosting fusion expressive capability for temporal-spatial features. The video data captured by satellites, colloquially referred to as "gazing," holds significant value for analyzing object status and dynamically tracking movements. Such data and their associated applications play pivotal roles in disaster response, urban traffic monitoring, national defense, and security operations. The multi-object tracking task in satellite video surveillance is inherently more complex than single-object tracking, presenting challenges such as tracking dense small objects and limited capability in multi-object tracking in ambiguous environments. This study proposes a satellite video multi-object tracking (MCTracker) method that considers inter-frame motion correlation and multi-scale cascaded feature enhancement, achieving outstanding performance in multi-object tracking tasks. This study builds upon a differentially encoded backbone network incorporating both global and local information, proposing an Inter-frame Motion Correlation Module (IFMCM) to enhance inter-frame dynamic continuity. A Mixed-scale Enhancement Block (MSEB) is designed to address the challenge of detecting and tracking small objects under multiscale effects. Additionally, a Spatio-temporal Fusion Module (STFM) is introduced to improve the feature fusion representation capability among modules. Through experiments conducted on two open-source datasets, AIR-MOT and SAT-MTB, it has been demonstrated that the proposed MCTracker exhibits efficient synergistic consistency in detection and ReID tasks. On the AIR-MOT dataset, the object tracking accuracy, as measured by MOTA, reaches 59.4 %, achieving optimal performance. In the SAT-MTB dataset's four-class multiple object tracking task, particularly in tracking small-scale objects such as cars and ships, the proposed method demonstrates comprehensive optimal performance, yielding robust tracking results. The satellite video-based multiple object tracking approach proposed in this study holds significant reference value for further exploration of object motion states and trajectory tracking. [ABSTRACT FROM AUTHOR]

With the popularity of the Internet and the increase in the level of information technology, cyber attacks have become an increasingly serious problem. They pose a great threat to the security of individuals, enterprises, and the state. This has made network intrusion detection technology critically important. In this paper, a malicious traffic detection model is constructed based on a decision tree classifier of entropy and a proximal policy optimisation algorithm (PPO) of deep reinforcement learning. Firstly, the decision tree idea in machine learning is used to make a preliminary classification judgement on the dataset based on the information entropy. The importance score of each feature in the classification work is calculated and the features with lower contributions are removed. Then, it is handed over to the PPO algorithm model for detection. An entropy regularity term is introduced in the process of the PPO algorithm update. Finally, the deep reinforcement learning algorithm is used to continuously train and update the parameters during the detection process, and finally, the detection model with higher accuracy is obtained. Experiments show that the binary classification accuracy of the malicious traffic detection model based on the deep reinforcement learning PPO algorithm can reach 99.17% under the CIC-IDS2017 dataset used in this paper. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Chongqing University of Technology (Natural Science) is the property of Chongqing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Autonomous vehicles face challenges in small-target detection and, in particular, in accurately identifying traffic lights under low visibility conditions, e.g., fog, rain, and blurred night-time lighting. To address these issues, this paper proposes an improved algorithm, namely KCS-YOLO (you only look once), to increase the accuracy of detecting and recognizing traffic lights under low visibility conditions. First, a comparison was made to assess different YOLO algorithms. The benchmark indicates that the YOLOv5n algorithm achieves the highest mean average precision (mAP) with fewer parameters. To enhance the capability for detecting small targets, the algorithm built upon YOLOv5n, namely KCS-YOLO, was developed using the K-means++ algorithm for clustering marked multi-dimensional target frames, embedding the convolutional block attention module (CBAM) attention mechanism, and constructing a small-target detection layer. Second, an image dataset of traffic lights was generated, which was preprocessed using the dark channel prior dehazing algorithm to enhance the proposed algorithm's recognition capability and robustness. Finally, KCS-YOLO was evaluated through comparison and ablation experiments. The experimental results showed that the mAP of KCS-YOLO reaches 98.87%, an increase of 5.03% over its counterpart of YOLOv5n. This indicates that KCS-YOLO features high accuracy in object detection and recognition, thereby enhancing the capability of traffic light detection and recognition for autonomous vehicles in low visibility conditions. [ABSTRACT FROM AUTHOR]

Sound event detection involves identifying sound categories in audio and determining when they start and end. However, in real-life situations, sound events are usually not isolated. When one sound event occurs, there are often other related sound events that take place as co-occurrences or successive occurrences. The timing relationship of sound events can reflect their characteristics. Therefore, this paper proposes a sound event detection method for traffic scenes based on a graph convolutional network, which considers this timing relationship as a form of multimodal information. The proposed method involves using the acoustic event window method to obtain co-occurrences or successive occurrences of relationship information in the sound signal while filtering out possible noise relationship information. This information is then represented as a graphical structure. Next, the graph convolutional neural network is improved to balance relationship weights between neighbors and itself and to avoid excessive smoothing. It is used to learn the relationship information in the graph structure. Finally, the convolutional recurrent neural network is used to learn the acoustic feature information of sound events, and the relationship information of sound events is obtained by multi-modal fusion to enhance the performance of sound event detection. The experimental results show that using multi-modal information with the proposed method can effectively improve the performance of the model and enhance the perception ability of smart cars in their surrounding environment while driving. [ABSTRACT FROM AUTHOR]

Vehicle tracking is one of computer vision's most important applications, with applications ranging from robotics and traffic monitoring to autonomous vehicle navigation and many more. Even with the significant advancements in recent research, issues like occlusion, fluctuating illumination, and fast motion still need to be addressed, calling for more investigation and creativity in this field. This study performs a thorough examination of various vehicle-tracking approaches and suggests a thorough classification scheme that divides them into four main categories: strategies that rely on features, segmentation, estimate, or learning. Two wellknown methods are highlighted specifically in the estimation-based category: particle filters and Kalman filters. [ABSTRACT FROM AUTHOR]

Image defogging is an essential technology used in traffic safety monitoring, military surveillance, satellite and remote sensing image processing, medical image diagnostics, and other applications. Current methods often rely on various priors, with the dark-channel prior being the most frequently employed. However, halo and bright-field color distortion issues persist. To further improve image quality, an adaptive image-defogging algorithm based on bright-field region detection is proposed in this paper. Modifying the dark-channel image improves the abrupt changes in gray value in the traditional dark-channel image. By setting the first and second lower limits of transmittance and introducing an adaptive correction factor to adjust the transmittance of the bright-field region, the limitations of the dark-channel prior in extensive ranges and high-brightness areas can be significantly alleviated. In addition, a guide filter is utilized to enhance the initial transmittance image, preserving the details of the defogged image. The results of the experiment demonstrate that the algorithm presented in this paper effectively addresses the mentioned issues and has shown outstanding performance in both objective evaluation and subjective visual effects. [ABSTRACT FROM AUTHOR]

Featured Application: This work might be applied to long-term bridge health monitoring, including missing data reconstruction and bridge service status investigation, with short-term monitoring data. Response reconstruction is essential in bridge health monitoring for recovering missing data and evaluating service status. Previous studies have focused on reconstructing responses at specific cross-sections using data from adjacent sections. To address this challenge, time-series prediction methods have been employed for response reconstruction. However, these methods often struggle with the inherent complexities of long-term time-varying traffic conditions, posing practical challenges. In this study, we analyzed the theoretical physical characteristics of high-frequency bridge dynamics within a simplified vehicle–bridge interaction (VBI) system. Our analysis revealed that the relationship between high-frequency bridge responses across different cross-sections is time-invariant and only dependent on the bridge's mode shape. This relationship remains unaffected by time-varying factors such as traffic loading and environmental conditions like air temperature. Based on these physical characteristics, we propose the backpropagation artificial neural network (BP-ANN) method for response reconstruction. The validity of these physical characteristics was confirmed through finite element models, and the effectiveness of the proposed method was demonstrated using field test data from a continuous bridge. Our verification results show that the BP-ANN method enables effective utilization of short-term monitoring data for long-term bridge health monitoring, without necessitating real-time adjustments for factors such as traffic conditions or air temperature. [ABSTRACT FROM AUTHOR]

Small object detection in traffic sign applications often faces challenges like complex backgrounds, blurry samples, and multi-scale variations. Existing solutions tend to complicate the algorithms. In this study, we designed an efficient and simple algorithm network called StarCAN-PFD, based on the single-stage YOLOv8 framework, to accurately recognize small objects in complex scenarios. We proposed the StarCAN feature extraction network, which was enhanced with the Context Anchor Attention (CAA). We designed the Pyramid Focus and Diffusion Network (PFDNet) to address multi-scale information loss and developed the Detail-Enhanced Conv Shared Detect (DESDetect) module to improve the recognition of complex samples while keeping the network lightweight. Experiments on the CCTSDB dataset validated the effectiveness of each module. Compared to YOLOv8, our algorithm improved mAP@0.5 by 4%, reduced the model size to less than half, and demonstrated better performance on different traffic sign datasets. It excels at detecting small traffic sign targets in complex scenes, including challenging samples such as blurry, low-light night, occluded, and overexposed conditions, showcasing strong generalization ability. [ABSTRACT FROM AUTHOR]

The sensor network is an important part of the real-time monitoring system of traffic flow in tourist areas. In this study, we built a traffic flow prediction model based on machine learning to assist the construction and optimization of a sensor network. Using the sensor network, we developed a real-time monitoring system of traffic flow in tourist areas. Factors affecting traffic flow in the study area (Hongze Lake, China) were determined through interviews with 50 tourists. Using the scores of the factors, machine learning models such as random forest, decision tree, and support vector machine were constructed to predict traffic flow, and the result was used to design an appropriate sensor network. The predictions of three machine learning methods were compared to build a traffic flow prediction model. After comparing the predicted results with the tested ones, the performance of the prediction model was validated. Referring to such results, we selected the area near a square and bridges and the intersections of roads as key areas to install sensors for monitoring traffic flow. The system is being implemented in the study area to renovate the area to promote further growth of the tourism industry. [ABSTRACT FROM AUTHOR]

Traffic surveillance is crucial for road safety and efficiency. This study examines object detection techniques tailored for Indian traffic, highlighting the challenges faced. The aim is to track vehicles, pedestrians, and cyclists using drone-captured data in Indian cities. Urban areas in India grapple with traffic problems due to rapid growth and inadequate infrastructure. Effective monitoring can address these. Recent advancements in object detection, a critical computer vision task, show promise for this application. Employing the YOLOv8 model, trained on our drone-collected dataset, which improved detection accuracy for the Indian Traffic Scenario, is shown. This dataset labelling of vehicle bounding boxes underwent preprocessing using Gaussian Filter, resizing, normalization, and augmentation. Testing showcased the model's real-world applicability in areas like traffic management and autonomous driving. The proposed model thus enhances vehicle detection systems, fostering better safety and decision-making, yielding promising results, as evidenced by 0.86 mAP50 after training and testing the model under own UAV Dataset. [ABSTRACT FROM AUTHOR]

Air traffic controllers' mental workload significantly impacts their operational efficiency and safety. Detecting their mental workload rapidly and accurately is crucial for preventing aviation accidents. This study introduces a mental workload detection model for controllers based on power spectrum features related to gamma waves. The model selects the feature with the highest classification accuracy, β + θ + α + γ, and utilizes the mRMR (Max-Relevance and Min-Redundancy) algorithm for channel selection. Furthermore, the channels that were less affected by ICA processing were identified, and the reliability of this result was demonstrated by artifact analysis brought about by EMG, ECG, etc. Finally, a model for rapid mental workload detection for controllers was developed and the detection rate for the 34 subjects reached 1, and the accuracy for the remaining subjects was as low as 0.986. In conclusion, we validated the usability of the mRMR algorithm in channel selection and proposed a rapid method for detecting mental workload in air traffic controllers using only three EEG channels. By reducing the number of EEG channels and shortening the data processing time, this approach simplifies equipment application and maintains detection accuracy, enhancing practical usability. [ABSTRACT FROM AUTHOR]

Benefiting from the robust feature representation capability of convolutional neural networks (CNNs), the object detection technology of intelligent high-altitude UAV remote sensing has been developed rapidly. In this field, the adversarial examples (AEs) pose serious security risks and vulnerabilities to deep learning-based systems. Due to the limitation of object size, image degradation, and scene brightness, adding adversarial disturbances to small and dense objects is extremely challenging. To study the threat of AE for UAV object detection, a dynamic bi-level integrated attack (DBI-Attack) is proposed for intensive multi-scale UAV object detection. Firstly, we use the dynamic iterative attack (DIA) method to generate perturbation on the classification level by improving the momentum iterative fast gradient sign method (MIM). Secondly, the bi-level adversarial attack method (BAAM) is constructed to add global perturbation on the decision level for completing the white-box attack. Finally, the integrated black-box attack method (IBAM) is combined to realize the black-box mislabeling and fabrication attacks. We experiment on the real drone traffic vehicle detection datasets to better evaluate the attack effectiveness. The experimental results show that the proposed method can achieve mislabeling and fabrication attacks on the UAV object detectors in black-box conditions. Furthermore, the adversarial training is applied to improve the model robustness. This work aims to call more attention to the adversarial and defensive aspects of UAV target detection models. [ABSTRACT FROM AUTHOR]

With the rapid development of intelligent transportation systems, lane detection and traffic sign recognition have become critical technologies for achieving full autonomous driving. These technologies offer crucial real-time insights into road conditions, with their precision and resilience being paramount to the safety and dependability of autonomous vehicles. This paper introduces an innovative method for detecting and recognizing multi-lane lines and intersection stop lines using computer vision technology, which is integrated with traffic signs. In the image preprocessing phase, the Sobel edge detection algorithm and weighted filtering are employed to eliminate noise and interference information in the image. For multi-lane lines and intersection stop lines, detection and recognition are implemented using a multi-directional and unilateral sliding window search, as well as polynomial fitting methods, from a bird's-eye view. This approach enables the determination of both the lateral and longitudinal positioning on the current road, as well as the sequencing of the lane number for each lane. This paper utilizes convolutional neural networks to recognize multi-lane traffic signs. The required dataset of multi-lane traffic signs is created following specific experimental parameters, and the YOLO single-stage target detection algorithm is used for training the weights. In consideration of the impact of inadequate lighting conditions, the V channel within the HSV color space is employed to assess the intensity of light, and the SSR algorithm is utilized to process images that fail to meet the threshold criteria. In the detection and recognition stage, each lane sign on the traffic signal is identified and then matched with the corresponding lane on the ground. Finally, a visual module joint experiment is conducted to verify the effectiveness of the algorithm. [ABSTRACT FROM AUTHOR]

Traffic accidents are prone to occur in mixed-traffic flows of nonmotorized and motorized vehicles. Particularly, electric two-wheeled vehicles face risks when they move laterally to the motor vehicle lane to avoid obstacles. To decrease accident risks caused by lateral offset in mixed-traffic flows, this study proposed a lateral offset detection method based on deep learning technology. A vehicle lateral offset risk recognition model of You Only Look Once Version 8 (YOLOv8) was established using actual traffic scenes in Xuzhou as a dataset. The model was trained by loading the preprocessed and labelled dataset, selecting the model architecture suitable for target detection in traffic scenes, setting training parameters, and executing. Then, the training results were analyzed based on images, such as labelled images, training result plots, and model curves, and the accuracy and reliability of the model were evaluated using standard metrics, such as precision, recall, and average precision. Results demonstrate that the offset model of YOLOv8 can detect the lateral offset risk of vehicles from traffic surveillance videos. The proposed model has robustness in category judgment. The detection precision of lateral offset in the model is 0.638, the recall rate is 0.626, the mean average precision calculated at an intersection over union (IoU) threshold of 0.50 (mAP50) is 0.679, and the mean average precision calculated at an IoU threshold of 50%-95% (mAP50-95) is 0.358. The study provides an important reference for the construction and performance evaluation of lateral movement models of vehicles in mixed-traffic flows. [ABSTRACT FROM AUTHOR]

Current practices in traffic sign monitoring heavily rely on manual inspections, a method that is both time-consuming and prone to human error. This leads to inefficiencies in the management and maintenance of these critical roadside assets. The objective of this work is to overcome these limitations by proposing a method for automated change detection in traffic signs using low-density LiDAR data. The proposed solution integrates noise elimination, point cloud restructuring, and cross-scan KD-tree generation, followed by the application of unsupervised machine learning techniques for change identification. The effectiveness of this method was verified by testing across three different highways with varying point cloud resolutions. For robust testing, an algorithm was also designed to simulate a broad range of different damage scenarios in traffic signs of different types, sizes, and placements. Testing in different scenarios along almost 15 km of the road revealed impressive results with accuracy and F1 score metrics ranging from 92% to 100%. Moreover, the algorithm was also extremely efficient with an average runtime of just 115" per km of fully automated unattended processing. The change detection potential of the proposed algorithm extends beyond traffic signs, as it could be adapted for many highway elements, enhancing the efficiency of transportation asset management and highway maintenance programmes. The findings indicate that this approach not only fills a significant gap in the current traffic sign monitoring and asset management practice but also offers a promising, comprehensive solution towards automated, cost-effective, and precise monitoring and maintenance of traffic signs, thus addressing a major challenge in this area. [ABSTRACT FROM AUTHOR]

The efficient and accurate identification of traffic signs is crucial to the safety and reliability of active driving assistance and driverless vehicles. However, the accurate detection of traffic signs under extreme cases remains challenging. Aiming at the problems of missing detection and false detection in traffic sign recognition in fog traffic scenes, this paper proposes a recognition algorithm for traffic signs based on pix2pixHD+YOLOv5-T. Firstly, the defogging model is generated by training the pix2pixHD network to meet the advanced visual task. Secondly, in order to better match the defogging algorithm with the target detection algorithm, the algorithm YOLOv5-Transformer is proposed by introducing a transformer module into the backbone of YOLOv5. Finally, the defogging algorithm pix2pixHD is combined with the improved YOLOv5 detection algorithm to complete the recognition of traffic signs in foggy environments. Comparative experiments proved that the traffic sign recognition algorithm proposed in this paper can effectively reduce the impact of a foggy environment on traffic sign recognition. Compared with the YOLOv5-T and YOLOv5 algorithms in moderate fog environments, the overall improvement of this algorithm is achieved. The precision of traffic sign recognition of the algorithm in the fog traffic scene reached 78.5%, the recall rate was 72.2%, and mAP@0.5 was 82.8%. [ABSTRACT FROM AUTHOR]

With the continuous advancement of the economy and technology, the number of cars continues to increase, and the traffic congestion problem on some key roads is becoming increasingly serious. This paper proposes a new vehicle information feature map (VIFM) method and a multi-branch convolutional neural network (MBCNN) model and applies it to the problem of traffic congestion detection based on camera image data. The aim of this study is to build a deep learning model with traffic images as input and congestion detection results as output. It aims to provide a new method for automatic detection of traffic congestion. The deep learning-based method in this article can effectively utilize the existing massive camera network in the transportation system without requiring too much investment in hardware. This study first uses an object detection model to identify vehicles in images. Then, a method for extracting a VIFM is proposed. Finally, a traffic congestion detection model based on MBCNN is constructed. This paper verifies the application effect of this method in the Chinese City Traffic Image Database (CCTRIB). Compared to other convolutional neural networks, other deep learning models, and baseline models, the method proposed in this paper yields superior results. The method in this article obtained an F1 score of 98.61% and an accuracy of 98.62%. Experimental results show that this method effectively solves the problem of traffic congestion detection and provides a powerful tool for traffic management. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicle (UAV)-based object detection methods are widely used in traffic detection due to their high flexibility and extensive coverage. In recent years, with the increasing complexity of the urban road environment, UAV object detection algorithms based on deep learning have gradually become a research hotspot. However, how to further improve algorithmic efficiency in response to the numerous and rapidly changing road elements, and thus achieve high-speed and accurate road object detection, remains a challenging issue. Given this context, this paper proposes the high-efficiency multi-object detection algorithm for UAVs (HeMoDU). HeMoDU reconstructs a state-of-the-art, deep-learning-based object detection model and optimizes several aspects to improve computational efficiency and detection accuracy. To validate the performance of HeMoDU in urban road environments, this paper uses the public urban road datasets VisDrone2019 and UA-DETRAC for evaluation. The experimental results show that the HeMoDU model effectively improves the speed and accuracy of UAV object detection. [ABSTRACT FROM AUTHOR]