scholarly journals Traffic Accident Data Generation Based on Improved Generative Adversarial Networks

Sensors ◽  
2021 ◽  
Vol 21 (17) ◽  
pp. 5767
Author(s):  
Zhijun Chen ◽  
Jingming Zhang ◽  
Yishi Zhang ◽  
Zihao Huang
Keyword(s):  
Traffic Accident ◽  
Traffic Accidents ◽  
Original Data ◽  
Urban Traffic ◽  
Generative Adversarial Networks ◽  
Generation Model ◽  
Data Generation ◽  
Road Traffic Safety ◽  
Adversarial Networks ◽  
Accident Data

For urban traffic, traffic accidents are the most direct and serious risk to people’s lives, and rapid recognition and warning of traffic accidents is an important remedy to reduce their harmful effects. However, research scholars are often confronted with the problem of scarce and difficult-to-collect accident data resources for traffic accident scenarios. Therefore, in this paper, a traffic data generation model based on Generative Adversarial Networks (GAN) is developed. To make GAN applicable to non-graphical data, we improve the generator network structure of the model and used the generated model to resample the original data to obtain new traffic accident data. By constructing an adversarial neural network model, we generate a large number of data samples that are similar to the original traffic accident data. Results of the statistical test indicate that the generated samples are not significantly different from the original data. Furthermore, the experiments of traffic accident recognition with several representative classifiers demonstrate that the augmented data can effectively enhance the performance of accident recognition, with a maximum increase in accuracy of 3.05% and a maximum decrease in the false positive rate of 2.95%. Experimental results verify that the proposed method can provide reliable mass data support for the recognition of traffic accidents and road traffic safety.

Intrusion detection of railway clearance from infrared images using generative adversarial networks

2020 ◽  
pp. 1-13
Author(s):  
Yundong Li ◽  
Yi Liu ◽  
Han Dong ◽  
Wei Hu ◽  
Chen Lin
Keyword(s):  
Intrusion Detection ◽  
Synthetic Data ◽  
Generative Adversarial Networks ◽  
Generation Model ◽  
Single Shot ◽  
Data Generation ◽  
Infrared Images ◽  
Adversarial Networks ◽  
Training Samples ◽  
Rgb Images

The intrusion detection of railway clearance is crucial for avoiding railway accidents caused by the invasion of abnormal objects, such as pedestrians, falling rocks, and animals. However, detecting intrusions using deep learning methods from infrared images captured at night remains a challenging task because of the lack of sufficient training samples. To address this issue, a transfer strategy that migrates daytime RGB images to the nighttime style of infrared images is proposed in this study. The proposed method consists of two stages. In the first stage, a data generation model is trained on the basis of generative adversarial networks using RGB images and a small number of infrared images, and then, synthetic samples are generated using a well-trained model. In the second stage, a single shot multibox detector (SSD) model is trained using synthetic data and utilized to detect abnormal objects from infrared images at nighttime. To validate the effectiveness of the proposed method, two groups of experiments, namely, railway and non-railway scenes, are conducted. Experimental results demonstrate the effectiveness of the proposed method, and an improvement of 17.8% is achieved for object detection at nighttime.

scholarly journals Identification of potential traffic accident hot spots based on accident data and GIS

2020 ◽  
Vol 325 ◽  
pp. 01005
Author(s):  
Hongge Zhu ◽  
Yuntong Zhou ◽  
Yanyan Chen
Keyword(s):  
Traffic Safety ◽  
Hot Spots ◽  
Traffic Accident ◽  
Traffic Accidents ◽  
Road Traffic ◽  
Hot Spot ◽  
Geographic Information ◽  
Causality Analysis ◽  
Road Traffic Safety ◽  
Accident Data

The problem of road traffic safety has been widely concerned in recent years. The identification of traffic accident hot spots can effectively improve the road traffic safety and let the traffic managers formulate targeted improvement measures and suggestions. The traditional identification method of accident hot spot does not consider the spatial attribute of the accident, so it has some limitations in the identification of traffic accident hot area. Therefore, this paper first proposes a method to identify the hot spot of traffic accidents based on geographic information system (GIS). The mathematical model and machine learning model are used to explore the correlation between traffic accidents and spatial characteristics from macro and micro aspects. Finally, taking Beijing as an example, the feasibility of the research method is proved by using the accident data of Beijing in 2015 and the geographic information of Beijing. The research results of this paper can realize the spatial effective transformation of accident records, comprehensively consider the micro and macro attributes of the accident itself, realize the automatic and efficient identification of the accident hot spot. In addition, the causality analysis results between each attribute and the distribution of accident hot spots can help decision makers to formulate safety and sustainable road strategies.

Generation of Synthetic Data with Conditional Generative Adversarial Networks

2020 ◽  
Author(s):  
Belén Vega-Márquez ◽  
Cristina Rubio-Escudero ◽  
Isabel Nepomuceno-Chamorro
Keyword(s):  
Research Work ◽  
Synthetic Data ◽  
Original Data ◽  
Classification Problem ◽  
Generative Adversarial Networks ◽  
Data Generation ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Adversarial Networks ◽  
Original Dataset

Abstract The generation of synthetic data is becoming a fundamental task in the daily life of any organization due to the new protection data laws that are emerging. Because of the rise in the use of Artificial Intelligence, one of the most recent proposals to address this problem is the use of Generative Adversarial Networks (GANs). These types of networks have demonstrated a great capacity to create synthetic data with very good performance. The goal of synthetic data generation is to create data that will perform similarly to the original dataset for many analysis tasks, such as classification. The problem of GANs is that in a classification problem, GANs do not take class labels into account when generating new data, it is treated as any other attribute. This research work has focused on the creation of new synthetic data from datasets with different characteristics with a Conditional Generative Adversarial Network (CGAN). CGANs are an extension of GANs where the class label is taken into account when the new data is generated. The performance of our results has been measured in two different ways: firstly, by comparing the results obtained with classification algorithms, both in the original datasets and in the data generated; secondly, by checking that the correlation between the original data and those generated is minimal.

scholarly journals Mixed-Type Data Generation Method Based On Generative Adversarial Networks

2021 ◽  
Author(s):  
Ning Wei ◽  
Longzhi Wang ◽  
Guanhua Chen ◽  
Yirong Wu ◽  
Shuifa Sun ◽  
...  
Keyword(s):  
Policy Research ◽  
Real Data ◽  
Training Data ◽  
Mixed Data ◽  
Generative Adversarial Networks ◽  
Generation Model ◽  
Data Generation ◽  
Continuous Space ◽  
The Real ◽  
Adversarial Networks

Abstract Data-driven based deep learing has become a key research direction in the field of artificial intelligence. Abundant training data is a guarantee for building efficient and accurate models. However, due to the privacy protection policy, research institutions are often limited to obtain a large number of training data, which would lead to a lack of training sets circumstance. In this paper, a mixed data generation model (mixGAN) based on generative adversarial networks (GANs) is proposed to synthesize fake data that have the same distribution with the real data, so as to supplement the real data and increase the number of available samples. The model first pre-trains the autoencoder which maps given dataset into a low-dimensional continuous space. Then, the Generator constructed in the low-dimension space is obtained by training it adversarially with Discriminator constructed in the original space. Since the constructed Discriminator not only consider the loss of the continuous attributes but also the labeled attributes, the generator nets formed by the Generator and the decoder can effectively learn the intrinsic distribution of the mixed data. We evaluate the proposed method both in the independent distribution of the attribute and in the relationship of the attributes, and the experiment results show that the proposed generate method has a better performance in preserve the intrinsic distribution compared with other generation algorithms based on deep learning.

scholarly journals Improvement and Application of Generative Adversarial Networks Algorithm Based on Transfer Learning

2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Fangming Bi ◽  
Zijian Man ◽  
Yang Xia ◽  
Wei Liu ◽  
Wenjia Yang ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Loss Function ◽  
Network Architecture ◽  
Data Distribution ◽  
Original Data ◽  
Generative Adversarial Networks ◽  
Standard Normal Distribution ◽  
Generation Model ◽  
Generative Adversarial Network ◽  
Adversarial Networks

Generative adversarial networks are currently used to solve various problems and are one of the most popular models. Generator and discriminator are characteristics of continuous game process in training. While improving the quality of generated pictures, it will also make it difficult for the loss function to be stable, and the training speed will be extremely slow compared with other methods. In addition, since the generative adversarial networks directly learns the data distribution of samples, the model will become uncontrollable and the freedom of the model will become too large when the original data distribution is constantly approximated. A new transfer learning training idea for the unsupervised generation model is proposed based on the generation network. The decoder of trained variational autoencoders is used as the network architecture and parameters to generative adversarial network generator. In addition, the standard normal distribution is obtained by sampling and then input into the model to control the degree of freedom of the model. Finally, we evaluated our method on using the MNIST, CIFAR10, and LSUN datasets. The experiment shows that our proposed method can make the loss function converge as quickly as possible and increase the model accuracy.

scholarly journals Assessing inequality, irregularity, and severity regarding road traffic safety during COVID-19

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Lei Lin ◽  
Feng Shi ◽  
Weizi Li
Keyword(s):  
New York ◽  
Los Angeles ◽  
Traffic Safety ◽  
Traffic Accidents ◽  
Road Traffic ◽  
Human Mobility ◽  
Critical Conditions ◽  
Road Traffic Safety ◽  
Accident Data ◽  
The Impact

AbstractCOVID-19 has affected every sector of our society, among which human mobility is taking a dramatic change due to quarantine and social distancing. We investigate the impact of the pandemic and subsequent mobility changes on road traffic safety. Using traffic accident data from the city of Los Angeles and New York City, we find that the impact is not merely a blunt reduction in traffic and accidents; rather, (1) the proportion of accidents unexpectedly increases for “Hispanic” and “Male” groups; (2) the “hot spots” of accidents have shifted in both time and space and are likely moved from higher-income areas (e.g., Hollywood and Lower Manhattan) to lower-income areas (e.g., southern LA and southern Brooklyn); (3) the severity level of accidents decreases with the number of accidents regardless of transportation modes. Understanding those variations of traffic accidents not only sheds a light on the heterogeneous impact of COVID-19 across demographic and geographic factors, but also helps policymakers and planners design more effective safety policies and interventions during critical conditions such as the pandemic.

scholarly journals ROAD SAFETY CAMPAIGNS TO REDUCE TRAFFIC ACCIDENTS FOR YOUNG ROAD USERS

ICCD ◽  
2019 ◽  
Vol 2 (1) ◽  
pp. 601-606
Author(s):  
Widodo Budi Dermawan ◽  
Dewi Nusraningrum
Keyword(s):  
School Environment ◽  
Road Safety ◽  
Traffic Accident ◽  
Traffic Accidents ◽  
Road Traffic ◽  
School Administrators ◽  
Road User ◽  
The Road ◽  
Road Users ◽  
Accident Data

Every year we lose many young road users in road traffic accidents. Based on traffic accident data issued by the Indonesian National Police in 2017, the number of casualties was highest in the age group 15-19, with 3,496 minor injuries, 400 seriously injured and 535 deaths. This condition is very alarming considering that student as the nation's next generation lose their future due to the accidents. This figure does not include other traffic violations, not having a driver license, not wearing a helmet, driving opposite the direction, those given ticket and verbal reprimand. To reduce traffic accident for young road user, road safety campaigns were organized in many schools in Jakarta. This activity aims to socialize the road safety program to increase road safety awareness among young road users/students including the dissemination of Law No. 22 of 2009 concerning Road Traffic and Transportation. Another purpose of this program is to accompany school administrators to set up a School Safe Zone (ZoSS), a location on particular roads in the school environment that are time-based speed zone to set the speed of the vehicle. The purpose of this paper is to promote the road safety campaigns strategies by considering various campaign tools.

scholarly journals MOBILE-WEB-BASE VOLUNTEERED GEOGRAPHIC INFORMATION APPLICATION AND GEOMETRIC ACCURACY ANALYSIS FOR TRAFFIC ACCIDENTS

2020 ◽  
Vol XLIV-4/W3-2020 ◽  
pp. 375-378
Author(s):  
H. K. Sevinc ◽  
I. R. Karas ◽  
E. Demiral
Keyword(s):  
Traffic Accident ◽  
Traffic Accidents ◽  
Geographic Information ◽  
Google Earth ◽  
Geographical Information ◽  
Accuracy Analysis ◽  
Geometric Accuracy ◽  
Web Based ◽  
Mobile Web ◽  
Accident Data

Abstract. The users can contribute to geographic information through platforms such as Wikimapia and OpenStreetMap. They can also generate data by themselves with their applications in cyber worlds like Google Earth. This study is primarily designed to be a guide regarding Volunteered Geographical Information (VGI) and to evaluate the geometric accuracy of data collected from volunteers on application. The main purpose of this study is to present basic information about Volunteered Geographical Information (VGI), why users are tending to use VGI, the accuracy of the data entered by the user, to examine the examples of use in various fields, to learn about geographic information systems and to compare this phenomenon and also by developing a VGI application to examine the similarity between the actual data and the data collected from volunteer users. A mobile and web-based application have been developed to collect traffic accident data from volunteer users. The geometric accuracy analysis was performed by comparing the data collected with this application with the data obtained from the General Directorate of Security.

Sign in / Sign up

Export Citation Format

Share Document