• Journal of the Korea Convergence Society
• /
• v.12 no.2
• /
• pp.1-6
• /
• 2021

In the modern society, traffic problems are occurring as vehicle ownership increases. In particular, the incidence of highway traffic accidents is low, but the fatality rate is high. Therefore, a technology for detecting an abnormality in a vehicle is being studied. Among them, there is a vehicle anomaly detection technology using deep learning. This detects vehicle abnormalities such as a stopped vehicle due to an accident or engine failure. However, if an abnormality occurs on the road, it is possible to quickly respond to the driver's location. In this study, we propose a deep learning-based vehicle anomaly detection using road CCTV data. The proposed method preprocesses the road CCTV data. The pre-processing uses the background extraction algorithm MOG2 to separate the background and the foreground. The foreground refers to a vehicle with displacement, and a vehicle with an abnormality on the road is judged as a background because there is no displacement. The image that the background is extracted detects an object using YOLOv4. It is determined that the vehicle is abnormal.

• International Journal of Highway Engineering
• /
• v.9 no.4
• /
• pp.83-92
• /
• 2007

In 2004, the number of traffic crashes and deaths in Korea are 220,755 and 6,563, respectively. Korea Road Traffic Safety Authority reported that the number of traffic accidents occupies over 25% out of total accidents, and found that traffic crash probability is extremely high at intersections since intersections have various traffic conflict points. A Safety study using Traffic Conflict Technique is much more useful than a study using reported traffic accident data. Existing traffic conflict research hardly considered conflict severity occurring at intersections. So, the study developed new criteria considering conflict severity. Analytic methods precisely detecting crashing points using field surveying data, and applied an application of our new criteria. Opposing left-turn conflict criteria was devided by three groups(high severe conflict, middle severe conflict, and less severe conflict) based on conflict boundary by means of a standard vehicle length. After analyzing field surveying data(3hours), we found totally 41 opposing left-turn conflicts. 3 cases are high severe conflict, and another 10 cases are middle severe conflicts, and the other cases are less severe. Studies related in conflict severity are considerably important to evaluate intersection's detailed safety index, and existing studies(purely conflict counting does not consider severity) have a limitation to clearly determine the level of safety of intersections for an application.

• International Journal of Highway Engineering
• /
• v.13 no.4
• /
• pp.177-185
• /
• 2011

In order to reduce traffic accidents at long road sections involving high risks of traffic accidents, including tunnels, bridges or curves, the National Police Agency have decided a tentative installation and operation of an automated point-to-point speed enforcement system based on cameras at January, 2007. The first system was established at the direction to Kangreung of Dunnae Tunnel at December, 2007. Currently, the automated point-to-point speed enforcement system operates at 11 sections including eight sections on expressways and three sections on national highways. However, there have been few researches on the effects of the system upon the reduction of traffic accidents in a scientific way. To this end, this research effort was initiated to evaluate the effect of the automated point-to-point speed enforcement system on traffic safety by comparing the number of traffic accidents before and after the installation of the system using a comparison-group(C-G) method. Three-year-long traffic accident data for the expressways were collected in order to evaluate the system. As a result, the installation of the system was found to reduce traffic accidents by 49.97% in average.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.2
• /
• pp.142-151
• /
• 2018

In the era of the 4th Industrial Revolution, research and development on autonomous vehicles have been actively conducted all over the world. Under these international trends, the Ministry of Land, Infrastructure and Transport is actively promoting the development of autonomous vehicles aiming at commercialization of autonomous vehicles at level 3 or higher by 2020. In the level 3 autonomous vehicle, it is essential to transfer control between the driver and the vehicle according to driving situations. Prior to the full-fledged autonomous vehicle age, this study developed a representative scenario for the safety evaluation on take-over on expressways. To accomplish this, we developed a highway driving scenario first, and then developed six control transition scenarios based on 2014 highway traffic accident data and take-over data. The variables to be considered in the developed scenarios are divided into drivers, vehicles, and environmental factors. A total of 36 variables are selected.

• Journal of Auto-vehicle Safety Association
• /
• v.3 no.2
• /
• pp.17-21
• /
• 2011

Vehicle safety device such as active headrest and rear detection system has been developing as people are interested about rear end collision more than head on or than front. However, there is no any standard or criterion in order to evaluate vehicle safety device for rear end collision. Also there is no test protocol about rear end collision in vehicle experiment. Therefore, this research developed scenario for experiment about rear end collision in vehicle experiment. Also this research evaluated dangerousness about vehicle test and fitness about re-enacting rear end collision using scenario developed using commercial software (PC-Crash) which can re-enact vehicle collision in virtual vehicle experiment. Scenario developed according to statistics from National Highway Traffic Safety Administration and German In-Depth Accident Study. Scenario has twelve cases which composed of Re-LVS (Rear end Leading Vehicle Stop), Re-LVM (Rear end Lead Vehicle Moving) and scenario for evaluation about malfunction of active headrest.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.20 no.6
• /
• pp.39-45
• /
• 2012

More than 80 percent of traffic accidents related with lane departure believed to be the result of crossing the lane due to either negligence or drowsiness of the driver. Lane-departure related accident in the highway usually involve high fatality. Even though LDWS is believed to prevent accident 25% and reduce fatalities by 15% respectively, its effectiveness in performance is yet to be confirmed in many aspects. In this study, the vehicle lateral locations relative to warning zone envelop (earliest and latest warning zone) defined in ISO standard, ECE and NHTSA regulations are compared with respect to various factors including delays, vehicle speed and vehicle heading angle with respect to the lane. Since LDWS is designed to be activated at the speed over 60 km/h, vehicle speed range for the study is set to be from 60 to 100 km/h. The vehicle heading angle (yaw angle) is set to be up to 5 degree away from the lane (abrupt lane change) considering standard for lane change test using double lane-change test specification. The TLC is calculated using factors like vehicle speed, yaw angle and reaction time. In addition, the effect of vehicle type has been considered to assess LDWS safety.

• International Journal of Highway Engineering
• /
• v.10 no.4
• /
• pp.199-211
• /
• 2008

This study deals with the side right-angle collisions of 4-legged signalized intersections in Cheongju. The goals are to analyze the characteristics of accidents and to find out the accident factors that affect severity using ordered probit model. In pursuing the above, the study uses the data of 580 side right-angle collisions occurred at the 181 intersections(2004-2005). The analyses show that more accidents were occurred in the nighttime and in going straight. The main cause was analyzed to be the red-light violation. Also, the main results of modeling are the following, First, the likelihood ratio index is 0.094 and t-ratio values that explain goodness of fit are significant. Second, minor road traffic volumes, minor road lanes, major road left-turn lanes, major road left-turn signal, major road yellow signal time, cross angle, major and minor road speed limits are significant factors affecting crash severities at signalized intersections.

• Korean Journal of Environment and Ecology
• /
• v.25 no.5
• /
• pp.649-657
• /
• 2011

This study had targeted the Manjong Junction~Hongchun interchange section of Jungang highway in order to analyze the efficiency and improvement of the highway wild-life fences. Being analyzed wild-life fence as an effective facility for the prevention of road-kill, it had founded that the wild animals did not jump over even the two layer guardrail, concrete barrier and noise barrier, fence for rock fall. For the section with high road-kill decrease effect after installing the wild-life fence, the effect was higher at wild-life fence with the height of 1.0m where the door had not been installed and the well connected points of road facility than the area with fence height of 1.5m. The road-kill decrease effect was also high at the well managed areas even if the door has not been installed. Also, road-kill had occurred by concentrating around the end of wild-life fence after installing the fence, Moreover road-kill had also occurred around the cutting section. The door of wild-life fence had higher amount of road-kill occurrence as the installation interval was closer. it was analyzed that the door management has a lot of effect on road-kill decrease. The fence for rock fall, two layer guardrail and concrete barrier having the effect of wild-life fence installed on the road would have to be installed by connecting with wild-life fences through proper facility improvement. Although the door should not be installed if possible, it should be installed as automatic door or gravity door to prevent the door from leaving the door open. An escape route has to be formed for the prompt escape away from the road for the animals entered through the ending section of the fence. The eco-corridor has to be made by restoring the soil layer of dual purpose eco-corridor forming a planting area. Also, the dead body after the accident has to be disposed to the outer section of the road immediately in order to prevent the secondary road-kill by the predator from the road-kill. The fence has to be installed as 500m or longer in both ways, in other words 1,000m or longer, from the targeted spot of minimum accident prevention while connecting up to the bridge or box culvert, etc that are next road facilities if possible to guide wild animals safely to the eco-corridor.

• International Journal of Highway Engineering
• /
• v.18 no.5
• /
• pp.75-82
• /
• 2016

PURPOSES : This study was initiated to estimate the benefits from the campaign to prevent drowsy driving crashes on expressways. The study was conducted by the Korea Expressway Corporation using a contingent valuation method. METHODS : First, a questionnaire was designed for a preliminary survey. From the survey's results, the initial willingness to pay for the campaign was determined by averaging different amounts of payments chosen under virtual scenarios in the survey. The willingness to pay data was used to find a first bid price for the open-ended method used for the second survey. After that, a primary questionnaire was designed and conducted using a single dichotomous choice question (SDBCQ). Drivers at expressway resting areas were asked their willingness to pay for the campaign. Based on statistical analysis using data collected from the second survey, the mean willingness to pay was estimated using a probability utility function. Finally, the benefit from the campaign was calculated using the estimated willingness to pay and accident data on expressways. CONCLUSIONS : Based on the result from the contingent valuation method, the benefit from the campaign to prevent drowsy driving crashes was estimated to be 170.6 won per expressway trip. The benefit is to be paid as an additional toll. In addition, the traffic crash cost estimate is about 2,209,680,000 won less than the cost during the same period in 2014.

• International Journal of Highway Engineering
• /
• v.14 no.5
• /
• pp.165-177
• /
• 2012

PURPOSES: The number of traffic accidents in 2010 was 226,978 in Korea, a high percentage of which up to 12.61% was due to drunk driving. As it is expected that the number of traffic accidents will increase because of the drastic increase of the number of vehicle registrations and the prevalent drinking cultures, it is necessary to understand the driving characteristics of drunken drivers to lower the increasing rate. METHODS: This study, therefore, comparatively analyzes the two groups - one group before drinking and the other after drinking - based on the graph, and implements the correlation in each scenario(1,2,3). scenario 1. appearance of jaywalkers; scenario 2. appearance of an illegal left-turning car; and scenario 3. appearance of a vehicle and a person as obstacles to the driver after an accident. RESULTS: The comparative analysis of speed shows that the group after drinking was 50km/h faster than the group before drinking in Scenario 1, 20km/h in Scenario 2, and 15km/h in Scenario 3 respectively. In the comparative analysis of acceleration, the average level of the group after drinking was 0.15 higher than that of the group before drinking in Scenario 1, 0.30 in Scenario 2, and 0.15 in Scenario 3. In the comparative analysis of deceleration, the average level of the group after drinking was about 0.4 lower than that of the group before drinking in Scenario 1, 0.35 in Scenario 2, and 0.2 in Scenario 3 respectively. In the comparative analyses, the item of speed, acceleration and deceleration was of significance for each group in Scenarios. CONCLUSIONS: The comparative analysis demonstrated that there is a difference between the group before drinking and the group after drinking. In the analysis of correlation in each group, it was proved that the drunken group was of significance.