Search

• Quantified the amount of information on the tunnel entrance signage. • Validation of the effectiveness and timeliness of different information volumes. • Facility layouts more in line with desired results were proposed. Driver's perception of speed is the basis of driving safety, and installing speed limit signs at tunnel entrances is an intuitive means of controlling a driver's driving speed. Therefore, tunnel entrance signs should be set up effectively to ensure each signage can fulfil its intended effect. Quantifying the information volume conveyed by traffic signs, discovering the impact of different information volumes on driving speeds, and understanding the effects of typical signs are the prerequisites for the effective use of signage. This study collected the speed variations of 40 drivers driving at nine mountainous highway tunnels with different entrance traffic signs through a naturalistic driving real vehicle test. The effect of low-volume intermittent information at tunnel entrances is identified in terms of the black hole effect on drivers (BHD), white hole effect on drivers (WHD), velocity fluctuation trend (VFT), and effect of speed limit signs (ELS). The study's results confirm that the effect of speed limit signage with traffic sign information volume (TSIV) in the range of 8.904 to 33.318 bit is significant. In comparison, the sign will lose its effectiveness in guiding drivers to control their speed when TSIV is greater than 58.641 bits. Using low-volume information speed limit signs or tunnel warning signs repeated twice before entering a tunnel can alert drivers and regulate their driving speeds. The speed limit sign (sign ①) for each type of vehicle is set individually, and the combined speed limit and no lane change sign (sign ②) used in two stacks have a strong speed control effect. Moreover, the sign ②, when used alone, has the best timeliness; the effective distance of speed control can reach 1522 m; using it before short tunnels or long tunnels less than 1500 m can achieve better effectiveness. [ABSTRACT FROM AUTHOR]

Mobile mapping systems (MMS) have gained increasing interest as a cost-effective means of collecting geospatial data, catering to the digital mapping needs of various domains such as advanced driver assistance systems (ADAS) and intelligent transportation systems (ITS). In the generated maps, the location and class information of traffic signs are particularly crucial for the aforementioned applications. However, the extensive and complex nature of data collected by MMS makes it challenging to infer the location and class of traffic signs. Consequently, researchers have developed artificial intelligence-based methods for processing traffic sign data. In this study, a Crowdsourced Geographical Data Collection System (CGDCS) which is designed for the inference of traffic sign location and class information using artificial intelligence is introduced. CGDCS is a lightweight system that operates on mobile devices, leveraging the MobileNet architecture to detect and classify traffic signs present in real-time camera images, thereby transferring the location and class information of the signs to a database. The study demonstrates that CGDCS is more practical and efficient than traditional methods involving manual processing, semi-traditional methods based on the extraction of shape and color features of traffic signs, and AIbased methods that process field data in high-performance computers using high computer vision and machine learning techniques. [ABSTRACT FROM AUTHOR]

Over the coming years, the advancement of driverless transport systems for people and goods that are designed to be used on fixed routes will revolutionize the transportation system. Therefore, for a safe transportation system, detecting and recognizing traffic signals based on computer vision has become increasingly important. Deep learning approaches, particularly convolutional neural networks, have shown exceptional performance in various computer vision applications. The goal of this research is to precisely detect and recognize traffic signs that are present on the streets using computer vision and deep learning techniques. Previous work has focused on symbol-based traffic signals, where popular single-task learning models have been trained and tested. Therefore, several comparisons have been conducted to select accurate single-task learning models. For further improvement, these models are employed in a multi-task learning approach. Indeed, multi-task learning algorithms are built by sharing the convolutional layer parameters between the different tasks. Hence, for the multi-task learning approach, different experiments have been carried out using pre-trained architectures like, for instance, InceptionResNetV2 and DenseNet201. A range of traffic signs and traffic lights are employed to validate the designed model. An accuracy of 99.07% is achieved when the entire network has been trained. To further enhance the accuracy of the model for traffic signs obtained from the street, a region of interest module is added to the multi-task learning module to accurately extract the traffic signs available in the image. To check the effectiveness of the adopted methodology, the designed model has been successfully tested in real-time on a few Riyadh highways. [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.)

Smart cities are now embracing the new frontier of urban living, with advanced technology being used to enhance the quality of life for residents. Many of these cities have developed transportation systems that improve efficiency and sustainability, as well as quality. Integrating cutting-edge transportation technology and data-driven solutions improves safety, reduces environmental impact, optimizes traffic flow during peak hours, and reduces congestion. Intelligent transportation systems consist of many systems, one of which is traffic sign detection. This type of system utilizes many advanced techniques and technologies, such as machine learning and computer vision techniques. A variety of traffic signs, such as yield signs, stop signs, speed limits, and pedestrian crossings, are among those that the traffic sign detection system is trained to recognize and interpret. Ensuring accurate and robust traffic sign recognition is paramount for the safe deployment of self-driving cars in diverse and challenging environments like the Arab world. However, existing methods often face many challenges, such as variability in the appearance of signs, real-time processing, occlusions that can block signs, low-quality images, and others. This paper introduces an advanced Lightweight and Efficient Convolutional Neural Network (LE-CNN) architecture specifically designed for accurate and real-time Arabic traffic sign classification. The proposed LE-CNN architecture leverages the efficacy of depth-wise separable convolutions and channel pruning to achieve significant performance improvements in both speed and accuracy compared to existing models. An extensive evaluation of the LE-CNN on the Arabic traffic sign dataset that was carried out demonstrates an impressive accuracy of 96.5% while maintaining superior performance with a remarkably low inference time of 1.65 s, crucial for real-time applications in self-driving cars. It achieves high accuracy with low false positive and false negative rates, demonstrating its potential for real-world applications like autonomous driving and advanced driver-assistance systems. [ABSTRACT FROM AUTHOR]

Introduction: The design of traffic signs should consider human cognitive abilities to enhance drivers’ understanding of the signs. Cognitive features, as one of the crucial principles of ergonomics, are among the influential factors in the design of signs. The present study aimed to evaluate the effect of graphic elements of eight new sign designs based on cognitive features on conveying the message “prohibition of using mobile phones.” Material and Methods: This study was conducted in six driving schools in Tehran in 2013. One hundred seventy-four participants, with an average age of 23.5 and a standard deviation of six years, participated in this study. Participants were then presented with the designed signs through a colored questionnaire. They were instructed to evaluate the signs’ cognitive features including simplicity, concreteness, meaningfulness, and semantic closeness—using a Likert scale ranging from 0 to 100. Results: The results revealed that the average score of the cognitive features of the designed signs is higher than other traffic, industrial and pharmaceutical signs. In this study, “semantic closeness” was the best cognitive feature for predicting the message of the signs. The sign with the “hands-free” element had the best performance in transferring the message. Conclusion: This research aimed to identify the most effective of eight proposed signs for banning mobile phone use while driving. Participants rated the sign featuring a button phone with a hands-free symbol as the top choice. Although most of the mobile phones in the market are of the touch screen type and the use of button phones has decreased a lot, the symbol of these phones as the dominant symbol still effectively conveys messages.

Intelligent Transportation System (ITS) aims to strengthen the connection between vehicles, roads, and people. As the important road information in ITS, intelligent detection of traffic signs has become an important part in the intelligent vehicle. In this paper, a lightweight vehicle mounted multi-scale traffic sign detector is proposed. First, guided by the attention fusion algorithm, an improved feature pyramid network is proposed, named AFPN. Assign weights according to the importance of information and fuse multi-dimensional attention maps to improve feature extraction and information retention capabilities. Second, a multi-head detection structure is designed to improve the positioning and detection capability of the detector. According to the target scale, the corresponding detection head is constructed to improve the target detection accuracy. The experimental results show that compared with other state-of-the-art methods, the proposed method not only has excellent detection accuracy with 50.3% for small targets and 64.8% for large targets but also can better trade-off detection speed and detection accuracy. Furthermore, the proposed detector is deployed on the Jetson Xavier NX and integrated with the vehicle-mounted camera, inverter, and LCD to realize real-time traffic sign detection on the vehicle terminal, and the speed reaches 25.6 FPS. [ABSTRACT FROM AUTHOR]

This study investigates the critical role of retroreflectivity in traffic signs, particularly in the context of autonomous vehicles (AVs), where accurate detection is paramount for road safety. Retroreflectivity, influencing visibility and legibility, is essential for ensuring safe road conditions. The study aims to assess traffic sign retroreflectivity using handheld retroreflectometers and LiDAR data, offering a comprehensive comparison of results with a specific focus on the RA1 and RA2 traffic sign classes. In a real-world setting, an AV equipped with LiDAR sensors, GPS units, and a stereo camera collects data on traffic signs, including point cloud attributes such as intensity and density. Simultaneously, a handheld retroreflectometer measures retroreflectivity coefficients from identified traffic signs. While retroreflectometers provide precision, they face limitations regarding time-consuming measurements and handling large or elevated signs. In contrast, LiDAR systems efficiently evaluate retroreflective features for numerous signs without such constraints. Despite both methods consistently yielding accurate retroreflectivity, the study reveals a limited correlation between LiDAR point cloud data and handheld retroreflectivity coefficients. The implications of these findings are significant, particularly in the selection and maintenance of retroreflective materials in traffic signs, with direct repercussions on overall road safety. The results offer valuable insights into leveraging LiDAR technology to enhance AVs' detection capabilities. Recommendations for further research include exploring factors influencing LiDAR intensity, establishing a more accurate relationship between intensity and retroreflectivity, correcting the point cloud during intensity calibration, and testing empirical prediction models with a larger sample size. These endeavors aim to generate a robust regression graph and determine correlation coefficients, providing a more nuanced understanding of the intricate relationship between LiDAR data and handheld retroreflectivity coefficients in the context of traffic sign assessment.

Traffic signs are the globally essential map features for the era of autonomous driving and smart cities. Traffic sign recognition is a difficult task due to their multiple shapes, sizes, color, occlusions and complicated driving scenes. For automatic traffic signs and classification, a robust and efficient system is needed with a highly accurate prediction rate. Therefore, a novel goat-based convolutional neural–Kalman framework (GbCN-KF) is proposed to detect the traffic signs for smart cities. Primarily, the input traffic sign images are noise-filtered by the Kalman function and the relative features for the recognition process were extracted. Further, traffic signs are recognized by matching the features with the trained set and classified using the goat fitness function. The system is tested with the BTSC and GTSRB database. The performance score was evaluated for the datasets and compared with the prevailing recognition models. The model recorded a high accuracy percentage of 99.89% and 99.94% for the tested BTSC and GTSRB datasets. [ABSTRACT FROM AUTHOR]

The importance of how cognitive processes (Reasoning) influence on the understanding and the mental representation of road-side elements in form of the movement effects on the path driving must be studied in-depth. Literature shows that the key point to explain the influence of attention on movement is the meaning of the object being processed; literature also shows that the Reasoning with those cues we pay attention to has an effect on driving. By using a driving-simulation task, traffic signs have been tested on their effect on movement. Results show that the least-representative-of-their-meaning signs deviate participants´ movement path more intensively than the most-representative-of-their-meaning signs. Conclusions: Traffic accidents are well-known for their. The results here reported help to improve the cognitive Ergonomics of every road element, specially, traffic signs. [ABSTRACT FROM AUTHOR]

Copyright of China Sciencepaper is the property of China Sciencepaper 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.)

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. 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.)

Traffic signs are part of road equipment that is very important for motorists because they can provide direction while on the highway, and if there is damage, repair or replacement must be carried out immediately because it can cause traffic accidents. Data collection for damaged traffic signs is still done using the manual method, so it takes a long time. Therefore, a web- and android-based application was designed that implements the Analytical Hierarchy Process (AHP) algorithm in determining the priority of repair or replacement of traffic signs on the route of South Sulawesi Province. As a result of this research, the public can report the type of damage and its location via Android, and then the officer processes the data so that it displays the type of damage that is a priority for repair or replacement. Implement the Analytical Hierarchy Process algorithm into the application for prioritization of traffic sign improvement using two (two) web-based and Android platforms. System design using UML produces use cases (2 actors, admin, and user) and class diagrams (15 admin classes and 4 user classes). The black box used as a test produced 40 modules, of which all were in line with expectations

Over the coming years, the advancement of driverless transport systems for people and goods that are designed to be used on fixed routes will revolutionize the transportation system. Therefore, for a safe transportation system, detecting and recognizing traffic signals based on computer vision has become increasingly important. Deep learning approaches, particularly convolutional neural networks, have shown exceptional performance in various computer vision applications. The goal of this research is to precisely detect and recognize traffic signs that are present on the streets using computer vision and deep learning techniques. Previous work has focused on symbol-based traffic signals, where popular single-task learning models have been trained and tested. Therefore, several comparisons have been conducted to select accurate single-task learning models. For further improvement, these models are employed in a multi-task learning approach. Indeed, multi-task learning algorithms are built by sharing the convolutional layer parameters between the different tasks. Hence, for the multi-task learning approach, different experiments have been carried out using pre-trained architectures like, for instance, InceptionResNetV2 and DenseNet201. A range of traffic signs and traffic lights are employed to validate the designed model. An accuracy of 99.07% is achieved when the entire network has been trained. To further enhance the accuracy of the model for traffic signs obtained from the street, a region of interest module is added to the multi-task learning module to accurately extract the traffic signs available in the image. To check the effectiveness of the adopted methodology, the designed model has been successfully tested in real-time on a few Riyadh highways.

Smart cities are now embracing the new frontier of urban living, with advanced technology being used to enhance the quality of life for residents. Many of these cities have developed transportation systems that improve efficiency and sustainability, as well as quality. Integrating cutting-edge transportation technology and data-driven solutions improves safety, reduces environmental impact, optimizes traffic flow during peak hours, and reduces congestion. Intelligent transportation systems consist of many systems, one of which is traffic sign detection. This type of system utilizes many advanced techniques and technologies, such as machine learning and computer vision techniques. A variety of traffic signs, such as yield signs, stop signs, speed limits, and pedestrian crossings, are among those that the traffic sign detection system is trained to recognize and interpret. Ensuring accurate and robust traffic sign recognition is paramount for the safe deployment of self-driving cars in diverse and challenging environments like the Arab world. However, existing methods often face many challenges, such as variability in the appearance of signs, real-time processing, occlusions that can block signs, low-quality images, and others. This paper introduces an advanced Lightweight and Efficient Convolutional Neural Network (LE-CNN) architecture specifically designed for accurate and real-time Arabic traffic sign classification. The proposed LE-CNN architecture leverages the efficacy of depth-wise separable convolutions and channel pruning to achieve significant performance improvements in both speed and accuracy compared to existing models. An extensive evaluation of the LE-CNN on the Arabic traffic sign dataset that was carried out demonstrates an impressive accuracy of 96.5% while maintaining superior performance with a remarkably low inference time of 1.65 s, crucial for real-time applications in self-driving cars. It achieves high accuracy with low false positive and false negative rates, demonstrating its potential for real-world applications like autonomous driving and advanced driver-assistance systems.

Deontic norms are expected to impose individuals' control over their behavior. In this paper, we address such norms presented in traffic signs and test their influence over executive control functions. For Experiment 1, we develop a traffic flanker task in which the typical neutral arrows are replaced with traffic prohibition/obligation signs. Experiment 2 isolated the deontic aspect of the signs using simple arrows on red, blue, and green backgrounds and either primed them to be interpreted as traffic signs or as elements of a gaming console controller. Results in both studies show evidence of controlling context interferences more efficiently when dealing with deontic (traffic) signs than with simple arrows (Experiment 1) or with similar perceptive targets when primed with a deontic context than with a gaming context (Experiment 2). In both studies, obligation/blue signs mitigate flanker effects less than prohibition/red signs. Stimuli color affects the alertness of the cognitive system, with the color red being, by itself, a cue for increased control. Based on temporal analysis, we further discuss these results as evidence of an increase in proactive control that aims to prevent the occurrence of undesirable influence. [ABSTRACT FROM AUTHOR]

Introduction: Traffic signs can be functional if they are visible and legible, especially when they alert about grave hazards to vulnerable road users such as cyclists. However, what is the current state of traffic signs directed at cyclists? This knowledge is essential for road infrastructure, especially in a city like Bogotá, which presents growing mobility challenges for cyclists. This paper aims to characterize the state of the signs directed at cyclists in Bogotá and evaluate the conformity level according to current guidelines. Methods: A reliable method was developed to evaluate existing signs. The method assesses the physical and functional characteristics of signs to verify if they are adequate from the human factors perspective and if they facilitate perception, are obeyed, and conform regarding dimensions, location, and function. With the method, an observational study was applied to characterize and evaluate signs in urban public spaces. Results: Eighty traffic signs were characterized in ten high-risk sectors for cyclists in Bogotá. Fifty-five percent of the signs are not in good condition, and often damaged; 18% are not coherent with the environment. Thirty-eight percent do not comply with the appropriate board size. The “Get off the bike” sign has the highest violation rate. Conclusion: Traffic signs aimed at cyclists present problems focused on compatibility with guidelines and compliance by cyclists. The observation indicates no effective maintenance plans. The coherence problems indicate insufficient infrastructure design planning. The Road Signaling System requires standardization in its design, implementation, and operation to promote perception, comprehension, and compliance.

Traffic signs are updated quickly, and there image acquisition and labeling work requires a lot of manpower and material resources, so it is difficult to provide a large number of training samples for high-precision recognition. Aiming at this problem, a traffic sign recognition method based on FSOD (few-shot object learning) is proposed. This method adjusts the backbone network of the original model and introduces dropout, which improves the detection accuracy and reduces the risk of overfitting. Secondly, an RPN (region proposal network) with improved attention mechanism is proposed to generate more accurate target candidate boxes by selectively enhancing some features. Finally, the FPN (feature pyramid network) is introduced for multi-scale feature extraction, and the feature map with higher semantic information but lower resolution is merged with the feature map with higher resolution but weaker semantic information, which further improves the detection accuracy. Compared with the baseline model, the improved algorithm improves the 5-way 3-shot and 5-way 5-shot tasks by 4.27% and 1.64%, respectively. We apply the model structure to the PASCAL VOC dataset. The results show that this method is superior to some current few-shot object detection algorithms. [ABSTRACT FROM AUTHOR]

The current computational advance allows the development of technological solutions using tools, such as mobile robots and programmable electronic systems. We present a design that integrates the Khepera IV mobile robot with an NVIDIA Jetson Xavier NX board. This system executes an algorithm for navigation control based on computer vision and the use of a model for object detection. Among the functionalities that this integration adds to the Khepera IV in generating guided driving are trajectory tracking for safe navigation and the detection of traffic signs for decision-making. We built a robotic platform to test the system in real time. We also compared it with a digital model of the Khepera IV in the CoppeliaSim simulator. The navigation control results show significant improvements over previous works. This is evident in both the maximum navigation speed and the hit rate of the traffic sign detection system. We also analyzed the navigation control, which achieved an average success rate of 93 % . The architecture allows testing new control techniques or algorithms based on Python, facilitating future improvements. [ABSTRACT FROM AUTHOR]

This article proposes an algorithm for recognizing road signs based on a determination of their color and shape. It first searches for the edge segment of the road sign. The boundary curve of the road sign is defined by the boundary of the edge segment. Approximating the boundaries of a road sign reveals its shape. The hierarchical road sign recognition system forms classes in the form of a sign. Six classes are at the first level. Two classes contain only one road sign. Signs are classified by the color of the edge segment at the second level of the hierarchy. The image inside the edge segment is cut at the third level of the hierarchy. The sign is then identified based on a comparison of the pattern. A computer experiment was carried out on two collections of road signs. The proposed algorithm has a high operating speed and a low percentage of errors. [ABSTRACT FROM AUTHOR]

Hundreds of people are injured or killed in road accidents. These accidents are caused by several intrinsic and extrinsic factors, including the attentiveness of the driver towards the road and its associated features. These features include approaching vehicles, pedestrians, and static fixtures, such as road lanes and traffic signs. If a driver is made aware of these features in a timely manner, a huge chunk of these accidents can be avoided. This study proposes a computer vision-based solution for detecting and recognizing traffic types and signs to help drivers pave the door for self-driving cars. A real-world roadside dataset was collected under varying lighting and road conditions, and individual frames were annotated. Two deep learning models, YOLOv7 and Faster RCNN, were trained on this custom-collected dataset to detect the aforementioned road features. The models produced mean Average Precision (mAP) scores of 87.20% and 75.64%, respectively, along with class accuracies of over 98.80%; all of these were state-of-the-art. The proposed model provides an excellent benchmark to build on to help improve traffic situations and enable future technological advances, such as Advance Driver Assistance System (ADAS) and self-driving cars. [ABSTRACT FROM AUTHOR]

Traffic sign retroreflectivity is critical for nighttime visibility, an important factor in driver safety. Current methods of sign retroreflectivity assessment are expensive, time-consuming, dangerous, or subjective. There is an urgent need to explore an alternative method that is cost-effective, safe, objective, and can be operated during daytime or nighttime. One such method is mobile lidar. However, a methodology utilizing lidar cloud data for practical retroreflectivity condition assessment is still lacking because of the inability to numerically correlate lidar retro-intensity readings to the retroreflectivity standard set by the Manual on Uniform Traffic Control Devices (MUTCD). In addition, there is also a need to explore sign deterioration behavior using real-world lidar data. In this study, we (1) propose a practical, categorical traffic sign condition assessment using lidar data; (2) establish a preliminary correlation between the retro-intensity and retroreflectivity readings to determine the minimum retro-intensity thresholds for condition assessment of different sheeting types and colors; (3) validate the accuracy of the assessment by comparing it with standard nighttime visual inspection outcomes; (4) demonstrate the practical implementation through a feasibility study at Georgia Interstate 285; and (5) reveal the retro-intensity deterioration trends using historical lidar cloud data. The results show that the proposed methodology can reliably yield results comparable to manual measurements, potentially reducing sign retroreflectivity condition assessment effort, increasing the transportation agency’s productivity, and filling gaps where manual assessment is not possible. Additionally, the retro-intensity deterioration trends can help transportation agencies to understand the long-term behavior of sign retro-intensity and predict the optimal timing for sign replacement. [ABSTRACT FROM AUTHOR]

Quantum machine learning (QML) has attracted significant research attention over the last decade. Multiple models have been developed to demonstrate the practical applications of the quantum properties. In this study, we first demonstrate that the previously proposed quanvolutional neural network (QuanvNN) using a randomly generated quantum circuit improves the image classification accuracy of a fully connected neural network against the Modified National Institute of Standards and Technology (MNIST) dataset and the Canadian Institute for Advanced Research 10 class (CIFAR-10) dataset from 92.0% to 93.0% and from 30.5% to 34.9%, respectively. We then propose a new model referred to as a Neural Network with Quantum Entanglement (NNQE) using a strongly entangled quantum circuit combined with Hadamard gates. The new model further improves the image classification accuracy of MNIST and CIFAR-10 to 93.8% and 36.0%, respectively. Unlike other QML methods, the proposed method does not require optimization of the parameters inside the quantum circuits; hence, it requires only limited use of the quantum circuit. Given the small number of qubits and relatively shallow depth of the proposed quantum circuit, the proposed method is well suited for implementation in noisy intermediate-scale quantum computers. While promising results were obtained by the proposed method when applied to the MNIST and CIFAR-10 datasets, a test against a more complicated German Traffic Sign Recognition Benchmark (GTSRB) dataset degraded the image classification accuracy from 82.2% to 73.4%. The exact causes of the performance improvement and degradation are currently an open question, prompting further research on the understanding and design of suitable quantum circuits for image classification neural networks for colored and complex data. [ABSTRACT FROM AUTHOR]

Summary: The detection and classification of traffic signs is a major challenge for self‐driving vehicles. The task can be narrowed down to detecting and classifying Indian Cautionary Traffic Signs (ICTS). In this proposed work, the difficulty of detecting and identifying Indian Cautionary Traffic Signs is addressed, and sincere attempts have been made to attain a possible solution using a Generative Adversarial Network (GAN), Improved Mask R‐CNN, and GrabCut algorithms. Data augmentation is done using Cascade Pyramid Generative Adversarial Network (CP‐GAN) to upsize the data set. The Mask R‐CNN with certain adoptions termed Improved mask RCNN is used in conjunction with the Grab‐Cut method to handle ICTS detection and identification through automatic end‐to‐end learning. Initially, Improved Mask R‐CNN generates a pixel‐by‐pixel segmentation mask for each item in the input sample. Masks developed using Improved Mask R‐CNN are not always clean, that is, some background pixels are often seen in the foreground segmentation. Hence, the generated masks are refined using the Grab Cut algorithm to enhance image segmentation. This combined approach works well in isolating the traffic signs from the ground truth images. Improved Mask R‐CNN attained better performance in the overall performance of traffic sign detection in the Indian data‐set. This method recognizes 40 different types of cautionary traffic signs from the unique Indian data set. The results are provided for complicated traffic sign categories that have not been addressed before. The proposed technique is trained and evaluated on ICTS, GTSDB, STSD, LISA, and DITS data sets, and it is also compared against cutting‐edge object recognition methods like Mask RCNN, Faster RCNN, SSD, and YOLOv3. [ABSTRACT FROM AUTHOR]

The traffic signal recognition model plays a significant role in the intelligent transportation model, as traffic signals aid the drivers to driving the more professional with awareness. The primary goal of this paper is to proposea model that works for the recognition and detection of traffic signals. This work proposes the pre-processing and segmentation approach applying machine learning techniques are occurred recent trends of study. Initially, the median filter&histogram equalization technique is utilized for pre-processing the traffic signal images, and also information of the figures being increased. The contrast of the figures upgraded, and information about the color shape of traffic signals are applied by the model. To localize the traffic signal in the obtained image, then this region of interest in traffic signal figures are extracted. The traffic signal recognition and classification experiments are managed depending on the German Traffic Signal Recognition Benchmark- (GTSRB). Various machine learning techniques such as Support Vector Machine (SVM), Extreme Learning Machine (ELM), Linear Discriminant Analysis (LDA), Principal Component Analysis (PCA), Convolutional neural network (CNN)- General Regression Neural Network (GRNN) is used for the classification process. Finally, the obtained results will be compare in terms of the performance metrics like accuracy, F1 score, kappa score, jaccard score, sensitivity, specificity, recall, and precision. The result shows that CNN-GRNN with ML techniques by attaining 99.41% accuracy compare to other intelligent methods. In this proposed technique is used for detecting and classifying various categories of traffic signals to improve the accuracy and effectiveness of the system. [ABSTRACT FROM AUTHOR]

Aiming at the decrease in the accuracy of traffic sign recognition due to dim light in the night environment, this paper proposes an improved you only look once version 5 (YOLOv5) algorithm to reduce carbon emissions. An improved adaptive histogram equalization method is designed to adjust the brightness and contrast of the image and highlight the detail information of traffic signs. In response to the higher requirements of the driving assistance system on the recognition model processing speed, the model is lightened and the standard convolution method of the backbone network is designed as a depth-separable convolution method, which greatly reduces the number of model parameters. To address the problem of feature loss during model learning, an improved feature pyramid AAM SPPF path aggregation network (AS-PAN) structure is proposed to enhance the learning capability of the model by adding an adaptive attention module to the Neck head and a spatial pyramid pooling module before its P3 and P4 outputs. Finally, the traditional non-maximum suppression (NMS) generates prediction frames by replacing the traditional NMS with weighted frame fusion weighted boxes fusion (WBF), which changes all possibility target frames from discard to fusion. Experiments demonstrate that the improved algorithm achieves improved detection accuracy, decreased processing time for a single image and low carbon emissions in the traffic sign recognition process compared with the original YOLOv5 algorithm in a self-built nighttime environment dataset. [ABSTRACT FROM AUTHOR]

Autonomous driving is a promising way to future safe, efficient, and low-carbon transportation. Real-time accurate target detection is an essential precondition for the generation of proper following decision and control signals. However, considering the complex practical scenarios, accurate recognition of occluded targets is a major challenge of target detection for autonomous driving with limited computational capability. To reveal the overlap and difference between various occluded object detection by sharing the same available sensors, this paper presents a review of detection methods for occluded objects in complex real-driving scenarios. Considering the rapid development of autonomous driving technologies, the research analyzed in this study is limited to the recent five years. The study of occluded object detection is divided into three parts, namely occluded vehicles, pedestrians and traffic signs. This paper provided a detailed summary of the target detection methods used in these three parts according to the differences in detection methods and ideas, which is followed by the comparison of advantages and disadvantages of different detection methods for the same object. Finally, the shortcomings and limitations of the existing detection methods are summarized, and the challenges and future development prospects in this field are discussed.

This research demonstrates the percentage of drivers and pedestrians understanding the traffic sign in Dhaka, Bangladesh. The survey was conducted among 634 drivers inside and outside of Dhaka city. Moreover, 508 pedestrians participated in the survey within Dhaka city. In comparison, there were 863 male respondents and 279 female respondents among 1142 respondents. The survey took the form of multiple-choice questions that included the picture attached to each sign. However, the survey questionnaires included a few questions regarding driver gender, age, educational qualification, and driving experience. Similarly, the survey questionnaires for pedestrians had also been discussed questions regarding pedestrian gender, age, educational qualification, and job status. The overall traffic sign understanding of drivers was 68.68%. Moreover, the comprehensive traffic signs understanding of pedestrians was 64.5%. The findings showed that the drivers had a medium degree of understanding of the traffic sign’s meaning. However, the study results showed that efforts are needed to educate the drivers and pedestrians about the proper interpretation and reaction to traffic signals. [ABSTRACT FROM AUTHOR]

Road maintenance is a key aspect of road safety and resilience. Traffic signs are an important asset of the road network, providing information that enhances safety and driver awareness. This paper presents a method for the recognition and geolocation of vertical traffic signs based on artificial intelligence and the use of a low-cost mobile mapping system. The approach developed includes three steps: First, traffic signals are detected and recognized from imagery using a deep learning architecture with YOLOV3 and ResNet-152. Next, LiDAR point clouds are used to provide metric capabilities and cartographic coordinates. Finally, a WebGIS viewer was developed based on Potree architecture to visualize the results. The experimental results were validated on a regional road in Avila (Spain) demonstrating that the proposed method obtains promising, accurate and reliable results. [ABSTRACT FROM AUTHOR]

According to road traffic accident (hereinafter referred to as RTA) statistics, the vulnerable road users are pedestrians in Latvia. The aim of this study is to investigate and analyse technical equipment used on non-signalled pedestrian crossings (zebra crossings) in Latvia and to make suggestions for measures that would increase road traffic safety on zebra crossings. RTAs involving collisions with pedestrians were filtered from the Ministry of the Interior database for a three-year period from 2016 to 2018. Thirty-two zebra crossings with a higher number of accidents with pedestrians were observed on the spot during the daylight and at night in several cities of Latvia. The main emphasis during the observation was placed on traffic signs and zebra road marking performance. Pedestrian crossings were observed from car driver’s view by taking photographs during day-time and night-time observations. Most attention was paid to road sign and road marking visibility from driver’s seat position. Retroreflection coefficient R’ was measured for each pedestrian crossing road sign. It was found that the condition and performance of traffic organisation equipment were not maintained on a regular basis and the life cycle of some traffic signs had well expired. Many road signs do not comply with minimum requirements, and road markings have weak visibility during wet weather conditions. It is recommended to improve visibility of pedestrian crossings from driver’s view in the urban areas by increasing rain vision for road markings and higher retroreflection class for traffic signs.

Road signs recognition plays an important role in improving traffic safety for both drivers and pedestrians. To ensure this recognition, many approaches are proposed by researchers. To overcome the limitations of the existing methods, Deep Learning approaches are used. This type of approaches achieves high recognition performances, and is also less sensitive to real world adverse conditions. However, they are in contrast very computationally expensive. From this perspective, the objective of this work is to adopt an approach that aims to reduce the computational complexity of these networks, in order to ensure a fast and efficient classification of traffic signs, especially for low and limited resources environments.

Copyright of Journal of Traffic & Transportation Research / Trafik ve Ulaşım Araştırmaları Dergisi is the property of Journal of Traffic & Transportation Research 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.)

Against the background of studies on redundancy in law that completely omit the visual element in law and of studies on traffic signs that are laconic about their redundancies, the present study proposes more focused investigation into the redundancies of traffic signs. After presentation of the broader context of existing studies on traffic signs and on redundancy in law, and following a discussion of the direct inspiration for embarking upon research into this topic, the article moves to present and discuss six proposed types of redundancies of signs. Utilizing Franciszek Studnicki's distinction between sign-types and sign-realizations, and given that traffic signs exist in various complicated relationships with each other, with written formulations in legal texts, and with the environments in which they are placed, the study comments on six types of redundancy: (1) sign-type–basic task(s) of road traffic law; (2) of sign-type's elements; (3) sign-type–sign-type; (4) sign-type–legal text; (5) sign-realization–sign-realization; and (6) sign-realization–environment. Analysis of each type is supported through examples, various subdivisions, and additional lines of inquiry. The study has value for both strictly theoretical and more practical enterprises related both to traffic signs themselves and to the wider realm of visualization of norms. [ABSTRACT FROM AUTHOR]