• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2938-2940
• /
• 1999

도로 색과 대비를 이루는 계수선을 도로위에 그려놓고 영상을 이용하여 통과 차량을 계수 하는 영상기반 교통량 측정 기술을 제안하였다. 기존의 컴퓨터 영상을 이용한 차량 계수 장치들은 대부분 도로와 차량의 gray level차이를 이용하는 방식이기 때문에 도로와 차량간의 구분이 쉽지 않고 그림자와 차량의 식별도 어려우며 조명에 의해 착오를 일으키기 쉬운 문제를 안고 있었다. 제안한 계수방법은 도로 위에 도로 색과 뚜렷한 대비를 이루는 띠를 도로를 가로질러 그려 놓은 다음 이를 통과하는 차량에 의해 띠의 형상이 절단되는지의 여부를 판단하여 차량을 계수 하는 방법이다. 이를 위해서 계수선을 포함하는 영상의 창을 설치하고 수시로 에지를 추출한 후, 에지의 절단 여부를 검사한다. 이때 계수선의 색과 동일한 차량이 통과할 경우에는 계수선의 에지는 절단되어 도로방향으로 연장되게 되며 계수선의 색과 다른 차량인 경우에는 계수선의 에지가 계수선의 내부로 연장되게 된다. 계수선 상에 그림자가 있는 경우에도 도로와 계수선 간의 색상변화가 뚜렷하므로 계수선의 에지 검출에는 영향을 받지 않게 된다. 제안한 알고리즘을 실험실에 설치된 모형도로와 모형 차량들을 이용하여 실험하고 결과를 제시하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.935-938
• /
• 2019

본 논문에서는 YOLOv3 기반의 차량 인식 및 계수 방안 연구를 진행한다. 이를 위해, 단일 단계 방식의 구조를 가진 YOLOv3 를 활용하여 영상 속 차량 인식 및 계수 하는 실험을 수행한다. 위의 실험을 통해 영상 속 차량의 수와 밝기에 인식률 정도와 차량 계수 평균 값 추출을 하였다. 실험 결과, 영상 속의 차량의 수와 밝기에 상관없이 90%이상의 차량에 대한 높은 인식률을 보였으며 1초 당 약 20 번씩 인식되는 차량의 수에 대한 계수의 평균 값을 추출하여 텍스트 파일 형태로 저장하여 시간에 따른 차량 계수를 확인하였다. 본 기술은 현대 사회의 교통 체증을 해결하기 위한 교통 흐름 예측에 활용될 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.562-564
• /
• 2003

지능형 교통 시스템(intelligent traffic system)의 필요성은 컴퓨터비전기술의 발전에 따라 성장해오고 있다. 지능형 교통 시스템의 바람직한 방법들 중 하나는 비전처리를 통하여 자동차의 수를 계수하는 것이다. 차량이 포함된 도로 이미지들은 디지털 카메라에 의하여 촬영된다. 우선적으로 이미지에서 도로의 중앙선 오른쪽부분은 제외한다. 단순히 개개의 차량을 분리해 나는 것은 겹치는 문제, 그림자, 조명과 같은 많은 문제점을 가지고 있다. 차량계수를 목적으로 이미지를 세밀히 조사하다가 보니 낮에는 차량의 앞 범퍼 끝의 그림자로 밤에는 헤드라이트 불빛으로 차량을 식별할 수 있다는 것을 알 수 있었다. 이 논문에서는 그림자분석을 통한 차량계수를 다루었다. 이 특징을 사용하여 차량을 계수하여 만족할만한 정확성을 성취하는 것이 이 논문의 목표이다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.14 no.4
• /
• pp.52-61
• /
• 2015

PCE(Passenger Car Equivalents) is used to analysis of road capacity and LOS(Level of Service). In this study calculates PCE by number of lane and 12 vehicle type by MOLIT(Minister of Land, Infra Structure and Transport) using individual vehicle data. The results of the calculation, PCEs are increased when high vehicle classification level, many number of lanes and weekend. Heavy vehicle factors are smaller than KHCM on 4, 6 lane. Also, In this study estimates of PCE variation model by heavy vehicle percentage. Impact of Heavy vehicles on PCEs is the most sensitive on 2 lane. The results of the study, heavy vehicles low impact on PCE on multi-lane and business trips are a little in weekend.

• Journal of Navigation and Port Research
• /
• v.31 no.1 s.117
• /
• pp.107-113
• /
• 2007

This research addressed the problem of describing how the operating characteristics of passenger car and large vehicle differ qualitatively and quantitatively through the analysis of field survey data. A formulation that estimates passenger car equivalents used in this paper is derived by microscopic headway method. Regression analysis was used to focus on the effect of vehicle type on intervehicular spacings and the modeling technique for the statistical analysis was detailed.

• International Journal of Highway Engineering
• /
• v.14 no.2
• /
• pp.29-36
• /
• 2012

In this study, frequently passing vehicles with two, three, four, and five axles were chosen through traffic volume analysis in Kyung-In Expressway in order to analyze how the road roughness and vehicle speed affect on the dynamic loads for roads in various vehicle classes. Dynamic loads according to chosen vehicles are estimated by TruckSim program. Dynamic load amplification factor is ratio between dynamic and static loads, and it is also determined for each vehicle classes. From the result of dynamic loads estimated by the dynamic load amplification factor, it is shown that for three-axles vehicle, when IRI is 3.5 and vehicle speed is 100km/hr, asphalt pavements receive additional 36% of static loads in maximum. The analysis of the amplification factor according to each vehicle classes also indicates that the amplification factor increases as the distance between the axles becomes smaller and each axle receives more loads.

• Journal of Korean Society for Geospatial Information Science
• /
• v.14 no.1 s.35
• /
• pp.57-63
• /
• 2006

The purpose of this study is to correct data on 3 intersection ranging from Kwangchun INT. to Nongsung INT. by the means of VTR recording and site survey and to measure the responsiveness of performance index by diversifying the platoon dispersion factor in signal progression simulation. The results are as follows : 1. The value of platoon dispersion factor was 0.28-0.33. The value in up stream is lower than that of in down stream even on the same intersection. 2. The platoon index showed big changes, though performance index didn't according to platoon dispersion factor. Therefore the value of platoon dispersion factor which is inner variable in T7F can be fixed for 0.34. 3. There was only little divergence in performance index changes according to platoon dispersion factor in designing the progression or T7F model.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.23 no.1
• /
• pp.45-52
• /
• 2010

This paper estimated the impact factor using the finite element program to confirm the dynamic behavior of new type of bridges constructed by introduction of new vehicles and compared the design criteria about the impact factor applied to domestic as well as each country. The study estimated effects of the impact factor according to pipe truss types modeled as respectively 34m, 44m, 54m and span length. The vehicle models are vehicle for bimodal tram of two axis and three axis which passes on actual bridge and dump truck model proposed by Park Young suk(1997). Each vehicle is estimated the impact factor according to velocity from 10 to 100(km/h) and examined. Also, the study investigated and compared the design regulation of domestic and a foreign country based on the impact factor on span center calculated in accordance with vehicle and span length.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.7 no.4
• /
• pp.313-321
• /
• 2005

Drag coefficient is one of the critical design factors to quantify the piston effect in vehicle tunnels. Several problems are raised on the drag coefficient currently applied for the ventilation system design; unverified adoption of the projected frontal area of the vehicle from the foreign study in the past, and lack of consideration for the slip-streaming effect. This study aims at better estimation of the traffic-induced ventilation force in the local tunnels. Values for the projected frontal area of the vehicles running on the local roads at present are proposed and results of an extensive CFD study are studied on the slip-streaming effects in various traffic conditions to quantify $K_{blockage}$ and the drag coefficient in the domestic tunnels.

• Journal of Korean Society of Steel Construction
• /
• v.23 no.2
• /
• pp.199-210
• /
• 2011

The cables in cable-stayed bridges are under high stress and are very sensitive to vibration due to their small section areas compared with other members. Therefore, it is reasonable to evaluate the cable impact factor by taking into account the dynamic effect due to moving-vehicle motion. In this study, the cable impact factors were evaluated via moving-vehicle-load analysis, considering the design parameters, i.e., vehicle weight, cable model, road surface roughness, vehicle speed, longitudinal distance between vehicles. For this purpose, two steel composite cable-stayed bridges with 230- and 540-m main spans were selected. The results of the analysis were then compared with those of the influence line method that is currently being used in design practice. The road surface roughness was randomly generated based on ISO 8608, and the convergence of impact factors according to the number of generated road surfaces was evaluated to improve the reliability of the results. A9-d.o.f. tractor-trailer vehicle was used, and the vehicle motion was derived from Lagrange's equation. 3D finite element models for the selected cable-stayed bridges were constructed with truss elements having equivalent moduli for the cables, and with beam elements for the girders and the pylons. The direct integration method was used for the analysis of the bridge-vehicle interaction, and the analysis was conducted iteratively until the displacement error rate of the bridge was within the specified tolerance. It was acknowledged that the influence line method, which cannot consider the dynamic effect due to moving-vehicle motion, could underestimate the impact factors of the end-cables at the side spans, unlike moving-vehicle-load analysis.