Number Of Traffic Signs Research Articles

Using traditional and intelligent organizational means to solve traffic problems in Baghdad/Maysaloun Intersection

The widespread use of vehicle has led to new problems in the city because of the need for these vehicles in the organization of movement and parking, especially in areas characterized by increased congestion in the morning and evening peak, which is one of the problems of the great urban growth that hit the city of Baghdad. Accordingly, this study dealt with identifying the traditional and intelligent organizational means to regulate traffic, and the most important specifications and requirements for their distribution within cities. An applied case study of Maysaloun intersection in Al-Ghadeer municipality in Baghdad city was studied for identifying the streets that suffer from the lack of traditional organizational means and the imbalance in their spatial distribution. Through field surveys of the study area, it was noted that the streets suffer from congestion in the morning and evening rush hours resulting from a spatial distribution deficit and a large deficit in the number of traffic signs that have a significant role in regulating traffic and traffic management. Service levels as a street condition were compared with US HCM standards using the HCS2010 program. It was found that all streets of the study area with a service level (LOS) are lower than level (C) and therefore this condition is met by the applicability of Variable message sign (VMS). Finally, a digital photo map using a satellite image of the study area has been prepared as a proposal showing the streets where the requirements for the deployment of intelligent signs (VMS) and basic traffic signs are available to regulate and control the traffic movement and to fill the deficit and imbalance in the number and spatial distribution.

Traffic sign recognition by combining global and local features based on semi‐supervised classification

The legibility of traffic signs has been considered from the beginning of design, and traffic signs are easy to identify for humans. For computer systems, however, identifying traffic signs still poses a challenging problem. Both image-processing and machine-learning algorithms are constantly improving, aimed at better solving this problem. However, with a dramatic increase in the number of traffic signs, labelling a large amount of training data means high cost. Therefore, how to use a small number of labelled traffic sign data reasonably to build an efficient and high-quality traffic sign recognition (TSR) model in the Internet-of-things–based (IOT-based) transport system has been an urgent research goal. Here, the authors propose a novel semi-supervised learning approach combining global and local features for TSR in an IOT-based transport system. In their approach, histograms of oriented gradient, colour histograms (CH), and edge features (EF) are used to build different feature spaces. Meanwhile, on the unlabelled samples, a fusion feature space is found to alleviate the differences between different feature spaces. Extensive evaluations on a collection of signs from the German Traffic Sign Recognition Benchmark (GTSRB) dataset shows that the proposed approach outperforms the others and provides a potential solution for practical applications.

The Impact of Traffic Sign Quality on the Frequency of Traffic Accidents

As part of an overall traffic signalling system, traffic signs warn the traffic participants, give information about the restrictions, prohibitions and obligations and provide additional information needed for a safe and undisturbed traffic flow. The traffic sign quality is expressed by its retroreflection and it is particularly pronounced in conditions of reduced visibility. The aim of this paper is to analyse how traffic signs quality affects the frequency of traffic accidents in low visibility conditions. For this purpose, traffic accidents and the quality of traffic signs were analysed on 130 state roads in the Republic of Croatia between 2013 and 2015. In the analysis several variables were used: the number of traffic accidents occurred under reduced visibility conditions, length of the road, AADT (Annual Average Daily Traffic) and the number of traffic signs that do not meet the minimum prescribed values of retroreflection. Statistical analysis showed a positive correlation between the accidents and unsatisfactory traffic signs, i.e. that with the increase in the number of traffic sings that do not have a satisfactory level of retroreflection, an increase in the number of accidents in reduced visibility conditions is expected.

Fast Branch Convolutional Neural Network for Traffic Sign Recognition

In this paper, we propose a novel framework for speeding up the test-time of traffic sign recognition, which is named Branch Convolution Neural Network. It is the first time to introduce a branch-output mechanism into a deep Convolution Neural Network. Our model has an accuracy as high as a deep convolution neural network model, while it performs faster at the same condition during test stage. It is a significantly accelerated framework for designing a real-time deep neural network system. We present a detail process to change a regular pre-trained Convolution Neural Network into a Branch Convolution Neural Network: train several simple branch classifiers, bias classifiers and optimize branches. Experiment applied on GTSRB shows that large number of traffic signs are unnecessary to go through all layers in a deep model and they can be separated out in a relative shallow neural network. This framework speeds up the recognition progress, while keeping the accuracy within an extremely minor drop.

Using stationary image based data collection method for evaluation of traffic sign condition

Abstract Transportation asset management helps monitor the transportation systems and optimize the construction, operation, and maintenance of assets. Many state Department of Transportations (DOTs) have already established asset management systems for high cost and low quantity assets, e.g., bridge and tunnel assets. However, due to the sheer number of traffic signs deployed by DOTs, statewide sign inventory and condition information are not well developed. Currently, using handheld devices is the most selected method by agencies to measure signs. To address safety challenge and high cost of data collection, an innovative stationary image based method has recently been proposed. This paper discusses the advantages and disadvantages of such image based method over using handheld devices in terms of the accuracy, possibility and consistency of data, speed, safety, maintenance, and cost. At its completion, this study provides suggestions to tackle the issues associated with image based method.

Effects of Memory Retrieval Load on Drivers’ Reproduction Performance Under Route Searching

Experiments were conducted using a personal computer to study the human factors involved in information processing during route searching. The primary objective was to investigate the effects of differences in the amount of memory retrieval load in terms of a given task. Two route-searching tasks were combined with two memory-reproducing tasks, for four types of tasks. The subjects used either a map guidance system or a voice guidance system to trace a predetermined route to the destination in a simulated driving test. In addition, the subjects were asked to perform memory-reproducing tasks using working memory. In these tasks, the subjects memorized the same number of traffic signs, but there was a difference in the memory retrieval process that the tasks demanded. The subjects were 62 students at Hokkaido University. Multiple analysis of variance was used to analyze the interaction effects and the main effects of tasks. The results indicated that the accuracy of route searching, the response time of route searching, and the accuracy of reproducing memorized traffic signs increased as the memory retrieval load decreased. This suggests that the memory retrieval process affects the performance of drivers’ information processing.