• Journal of Korean Society of Transportation
• /
• v.20 no.6
• /
• pp.81-98
• /
• 2002

본 연구는 Wilcoxon Rank Sum Test 기법을 이용한 자동 돌발상황 검지 모형을 개발하는 것이다. 본 연구의 수행을 위하여 고속도로에 설치된 루프 차량 검지기(Loop Vehicle Detection System)에서 수집된 점유율 데이터를 사용하였다. 기존의 검지모형은 산정하기가 까다로운 임계치에 의하여 돌발상황을 검지하는 방식이었다. 반면 본 연구 모델은 위치와 시간대 교통 패턴에 관계없이 모형을 일정하게 적용하며, 지속적으로 돌발상황 지점과 상·하류의 교통패턴을 비교 검정 기법인 Wilcoxon Rank Sum Test 기법을 사용하여 돌발상황 검지를 수행하도록 하였다. 연구모형의 검증을 위한 테스트 결과 시간과 위치에 관계없이 정확하고 빠른 검지시간(돌발 상황 발생 후 2∼3분)을 가짐을 알 수 있었다. 또한 기존의 모형인 APID, DES, DELOS모형과 비교검증을 위하여 검지율 및 오보율 테스트를 수행한 결과 향상된 검지 능력(검지율 : 89.01%, 오보율 : 0.97%)을 나타남을 알 수 있었다. 그러나 압축파와 같은 유사 돌발상황이 발생되면 제대로 검지를 하지 못하는 단점을 가지고 있으며 향후 이에 대한 연구가 추가된다면 더욱 신뢰성 있는 검지모형으로 발전할 것이다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.4
• /
• pp.97-113
• /
• 2023

Swift and appropriate responses in emergency situations like objects falling on the road can bring convenience to road users and effectively reduces secondary traffic accidents. In Korea, current intelligent transportation system (ITS)-based detection systems for emergency road situations mainly rely on loop detectors and CCTV cameras, which only capture road data within detection range of the equipment. Therefore, a new detection method is needed to identify emergency situations in spatially shaded areas that existing ITS detection systems cannot reach. In this study, we propose a ResNet-based algorithm that detects and classifies emergency situations from vehicle camera footage. We collected front-view driving videos recorded on Korean highways, labeling each video by defining the type of emergency, and training the proposed algorithm with the data.

• Journal of Korean Society of Transportation
• /
• v.19 no.6
• /
• pp.105-118
• /
• 2001

Incident detection rate and false alarm rate are the key measures tot estimating the performance of automatic incident detection algorithms. It is, however inherently very difficult to improve the two measures simultaneously. The main purpose of this study is to present some methods for solving the problem. For this, an incident detection algorithm has been designed in this study. The algorithm is consisted of two functions, one for detecting incident and another for detecting congestion. A logic for distinguishing non-recurrent congestion from recurrent congestion was employed in the algorithm. The new algorithm basically requires speed, flow, and occupancy data for defining incident situation, but the algorithm is able to perform this task without one of the three parameters. The performance of the algorithm has been evaluated by using the field data collected from Interstate Highway 880 in Bay Area, California. The empirical analysis results are very promising and thus, the algorithm proposed may be very useful for the analysts. This paper presents some empirical test results for the performance of California incident detection algorithm, only for the reference purpose.

• Journal of Korean Society of Transportation
• /
• v.22 no.6
• /
• pp.175-196
• /
• 2004

This study focuses on improving the performance of freeway incident detection by introducing some new measures to reduce false alarms in developing a new incident detection model. The model consists of the 5 major components through which a series of decision makings in determining the given traffic flow condition are made. The decision making process was designed such that the causes of traffic congestions can be accurately classified into several types including incidents and bottlenecks according to their unique characteristics. The model performance was tested and found to be compatible with that of the existing well-recognized models in terms of the detection rate and detection time. It should noted that the model produced much less false alarms than most of the existing models. The study results prove that the initial objective of the study was satisfied as it was an experimental trial to improve the false alarm rate for the incident detection model to be more pactically usable for traffic management purposes.

• Journal of Korean Society of Transportation
• /
• v.22 no.6
• /
• pp.77-88
• /
• 2004

This study aims at developing the model that is able to detect the compression wave, which is included as a similar situation in incidents, that causes false applicable to the similar character such as incidents in the incident detection model for expressways. In this study, it has been checked whether the number of false alarms is decreased or not by modularizing this model for being able to applicable to other models such as DES and DELOS, etc. which do not perform the compression wave test based on the compression wave test process of APID model which has been being used in the expressway traffic management system currently. The evaluation in this study focuses on the sensitivity of the model and the results analysis is performed classified by each polling cycle. And how well these models are working is evaluated by each polling cycle. In addition to this, the detection rate, the false alarm rate and the average detection time in both the existing models and the model in this study are calcuated. As a result of appling the model in this study, it is found that the false alarm rate is improved through the reasonable decrease in the number of false alarm frequencies and there are not remarkable changes concerning the detection rate and the average detection time. To sum up, it is expected that a good number of improvement effects will be occurred when this model is applied to the actual expressway traffic management system.

• Journal of Korean Society of Transportation
• /
• v.24 no.2 s.88
• /
• pp.127-137
• /
• 2006

This Paper Presents a novel incident detection method for using the speed-density difference between detectors. When a incident is happened the downstream traffic condition is mostly lower speed and higher density and the upstream is higher speed and lower density In respect of such characteristics, we can suggest a method for detecting an incident based on the speed-density information provided from detectors. The incident detection method is tested by using the accident data collected from the Cheonan-Nonsan Highway. The results show that suck an incident detection method can analyze the position of accident by comparing the speed-density difference between detectors. The work described in this paper is only at on early stage, in the sense that there are several areas to be further investigated in application of this method. such as setting a critical value for judging the incident. a level of accident wave, a varying Pattern between the incident locations and detectors, etc.

• 한국ITS학회:학술대회논문집
• /
• 2004.11a
• /
• pp.175-180
• /
• 2004

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.3
• /
• pp.38-50
• /
• 2023

Recent advances in artificial intelligence (AI) technology have enabled the integration of AI technology into image sensors, such as Closed-Circuit Television (CCTV), to detect specific traffic incidents. However, most incident detection methods have been carried out using fixed equipment. Therefore, there have been limitations to incident detection for all roadways. Nevertheless, the development of mobile image collection and analysis technology, such as image sensors and edge-computing, is spreading. The purpose of this study is to estimate the reducing effect of the incident detection time according to the introduction level of mobile image collection and analysis equipment (or connected vehicles). To carry out this purpose, we utilized data on the number of incidents collected by the Suwon branch of the Gyeongbu expressway in 2021. The analysis results showed that if the market penetration rate (MPR) of connected vehicles is 4% or higher for two-lane expressway and 3% or higher for three-lane expressways, the incident detection time was less than one minute. Furthermore, if the MPR is 0.4% or higher for two-lane expressways and 0.2% or higher for three-lane expressways, the incident detection time decreased compared to the average incident detection time announced by the Korea Expressway Corporation for both two-lane and three-lane expressways.

• Journal of Korean Society of Transportation
• /
• v.26 no.4
• /
• pp.229-239
• /
• 2008

Most algorithms for detecting incidents have been developed under the premise that congestion must happen whenever an incident occurs. For that reason, the performance of these algorithms could not be guaranteed in cases where congestion did not happen due to traffic operations with low flows despite the occurrence of an incident. The objective of this paper is to develop an automatic incident detection algorithm using a new diagram that can reliably detect the incident under various conditions of traffic operations including a low volume state. Compared with the McMaster Algorithm, the proposed algorithm in this paper was evaluated with three different cases in which the incidents occur in traffic operations with a low volume state, a relatively high volume state, and a recurrent congestion state. It is shown that the new algorithm has a capability to identify the flow characteristics of incidents for all the three cases and is much better than McMaster algorithm in terms of detection rate and false alarm rate.

• Journal of Korean Society of Transportation
• /
• v.34 no.2
• /
• pp.146-157
• /
• 2016

The objective of this study is to actualize a statistical model of the 3-detector simplification model, which was proposed to detect outbreak situations by Daganzo in 1997 and to verify the statistical appropriacy thereof. This study presents the calculation process of the 3-detector simplification model and realizes the process using a statistics program. Firstly, the model was applied using data on detector of the main highways on which there is no entrances or exits. Moreover, in order to statistically verify the 3-detector simplification model, accumulative traffics for 30 seconds period, which reflects the dynamic changes of traffics due to shock wave, were estimated for outbreak traffics and steady flow, and the error of acquired data was statistically compared with that of the actual accumulative traffics. As a result, the error ratio between steady and incident cumulative flows has reached its maximum after 2-3 hours from an accident. Moreover, the incident traffic flows by accidents and the stade flows are heterogeneous in terms of their dispersion and means.