• 방송공학회논문지
• /
• 제21권4호
• /
• pp.609-620
• /
• 2016

최근 발생빈도가 높은 차량 간 충돌사고를 미연에 방지하고 운전자의 편의를 증진하기 위한 전방 충돌 경고 시스템에 관한 연구가 활발히 진행되고 있다. 충돌 회피를 위한 차량 시스템에 자동으로 차량을 검출하는 기술은 필수적 요소이다. 기존의 학습 기반 차량 검출 방법들은 일반적으로 차량의 후면 전체를 학습하며, 외형이 다른 승용차와 트럭, SUV의 경우 클래스를 분류하여 학습해야 한다는 단점이 있다. 본 논문에서는 이러한 단점을 해결하기 위해 차종에 관계없이 후미등 하단 부의 외형은 유사하다는 점에 착안하여 하단부에 한해 Haar-like feature를 학습함으로써 전방 차량을 검출하는 방법을 제안하였다. 또한 검증단계로서 후미등 검출을 통해 실제 차량과 차량이 아닌 것들을 분류하고 후미등 검출이 어려운 작은 크기의 후보 영역은 HOG(Histogram Of Gradient) 특징과 SVM(Support Vector Machine) 분류기를 통해 검증하여 오검출률을 낮추었다. 도로 주변에 건물이 많은 복잡한 영상에서도 차종에 관계없이 95%에 해당하는 정확도를 보여 전방 차량 검출 성능이 개선된 것을 확인하였다.

• Steel and Composite Structures
• /
• 제46권6호
• /
• pp.731-744
• /
• 2023

Vehicle load information is an important role in operating and ensuring the structural health of cable-stayed bridges. In this regard, an efficient and economic method is proposed for vehicle load detection based on the observed cable tension and vehicle position using a graph neural network (GNN). Datasets are first generated using the practical advanced analysis program (PAAP), a robust program for modeling and considering both geometric and material nonlinearities of bridge structures subjected to vehicle load with low computational costs. With the superiority of GNN, the proposed model is demonstrated to precisely capture complex nonlinear correlations between the input features and vehicle load in the output. Four popular machine learning methods including artificial neural network (ANN), decision tree (DT), random forest (RF), and support vector machines (SVM) are refereed in a comparison. A case study of a cable-stayed bridge with the typical truck is considered to evaluate the model's performance. The results demonstrate that the GNN-based model provides high accuracy and efficiency in prediction with satisfactory correlation coefficients, efficient determination values, and very small errors; and is a novel approach for vehicle load detection with the input data of the existing monitoring system.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 추계학술대회 논문집 전기기기 및 에너지변환시스템부문
• /
• pp.238-240
• /
• 2006

For the method of train position detection, ground-based train position estimation mainly has been applied so far. Ground-based position detection is the way to detect train current positions by installing train position equipments on railroad lines. However, the ground-based methods should install detection equipments on each section, and can only be able to detect train positions from main command center. So this method has several disadvantages such as an discontinuous position detection, an increment in cost of installation and maintenance. To make possible continuous train position detection, and to minimize amount of the cost, the vehicle-based position detection method should be chosen to determine train positions by loading position equipments on vehicles. In this paper, to realize the vehicle-based train position detection method, configuration scheme of train position detection equipment is suggested by using GPS, inertial sensor, speed sensor and its performance is verified by simulations.

• 한국정보과학회논문지:시스템및이론
• /
• 제30권11호
• /
• pp.610-620
• /
• 2003

다양한 그림자는 비젼 기반 차량 검출에서 오류를 발생시키는 주요 원인이다. 본 논문에서는 노면 표시 기반 방법과 배경 빼기 및 에지(BS & Edge) 방법이라는 두 가지 방안을 차량 검출과 그림자 제거를 위해 제안하였다. 노변의 지형 지물들로 인해서 발생하는 그림자의 영향이 크게 증가하는 상황에서의 실험을 통해서 96% 이상의 차량 검출 정확도를 나타냄을 확인하였다. 전술한 두 가지 방법을 기반으로 하여, 차량 추적, 차량 계수, 차종 분류, 그리고 속도 측정을 수행하여 각 차로의 부하를 나타내는 데 사용되는 차량 흐름과 관련된 여러 가지 교통 파라미터를 추출하였다.

• 전기전자학회논문지
• /
• 제23권1호
• /
• pp.235-241
• /
• 2019

본 논문에서는 농업 기계 시스템에서 사용하기 위한 딥러닝 알고리즘 기반의 프레임 내의 관심 영역 유사성을 이용한 새로운 후방 접근 차량 검출 알고리즘을 제안한다. 농업 기계 시스템은 후방에서 접근하는 차량만 검출해야 한다. 지나가는 자동차가 검출되면 혼란을 야기할 수 있다. 논문에서는 차량 검출을 위해 딥러닝에서 뛰어난 검출률을 나타내는 Faster R-CNN 모델을 사용하였다. 딥러닝은 뒤에서 접근하는 차량뿐만 아니라 지나가는 차량도 검출하므로 긍정오류 차량을 배제해야 한다. 본 논문에서 이를 해결하기 위해 검출된 프레임에서 관심 영역에 대한 유사성과 평균 에러를 피라미드 형태로 이용하여 접근하는 자동차만 검출하는 알고리즘을 제안하였다. 실험을 통하여 제안된 방법이 평균 98.8%의 높은 검출률을 나타내었다.

• 한국정보기술학회 영문논문지
• /
• 제10권2호
• /
• pp.167-175
• /
• 2020

In this paper, we propose a traffic accident detection on vehicle-mounted camera. In the proposed method, the minimum bounding box coordinates the central coordinates on the bird's eye view and motion vectors of each vehicle object, and ego-motions of the vehicle equipped with dash-cam are extracted from the dash-cam video. By using extracted 4 kinds features as the input of Bi-LSTM (bidirectional LSTM), the accident probability (score) is predicted. To investigate the effect of each input feature on the probability of an accident, we analyze the performance of the detection the case of using a single feature input and the case of using a combination of features as input, respectively. And in these two cases, different detection models are defined and used. Bi-LSTM is used as a cascade, especially when a combination of the features is used as input. The proposed method shows 76.1% precision and 75.6% recall, which is superior to our previous work.

• 제어로봇시스템학회논문지
• /
• 제21권2호
• /
• pp.102-107
• /
• 2015

In this paper, we propose a technique of drowsiness detection using a histogram for vehicle safety. The drowsiness of vehicle drivers is often the main cause of many vehicle accidents. Therefore, the checking of eye images in order to detect the drowsiness status of a driver is very important for preventing accidents. In our suggested method, we analyse the changes of a histogram of eye region images which are acquired using a CCD camera. We develop a drowsiness detection system using this histogram change information. The experimental results show that the proposed method enhances the accuracy of detecting drowsiness to nearly 97%, and can be used to prevent accidents due to driver drowsiness.

• 제어로봇시스템학회논문지
• /
• 제15권4호
• /
• pp.390-395
• /
• 2009

In this paper, we design an on-road vehicle detection system based on the combination of a camera and a LIDAR system. In the proposed system, the candidate area is selected from the LIDAR data using a grouping algorithm. Then, the selected candidate area is scanned by an SVM to find an actual vehicle. The morphological edged images are used as features in a camera. The principal components of the edged images called eigencar are employed to train the SVM. We conducted experiments to show that the on-road vehicle detection system developed in this paper demonstrates about 80% accuracy and runs with 20 scans per second on LIDAR and 10 frames per second on camera.

• 한국자동차공학회논문집
• /
• 제17권1호
• /
• pp.1-11
• /
• 2009

This paper proposes an algorithm capable of detecting vehicles in front or in rear using a monocular camera installed in a vehicle. The vehicle detection has been regarded as an important part of intelligent vehicle technologies. The proposed algorithm is mainly composed of two parts: 1)hypothesis generation of vehicles, and 2)hypothesis verification. The hypotheses of vehicles are generated by the analysis of vertical and horizontal edges and the detection of symmetry axis. The hypothesis verification, which determines vehicles among hypotheses, is done by the AdaBoost algorithm. The proposed algorithm is proven to be effective through experiments performed on various images captured on the roads.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 1999년도 추계학술대회논문집
• /
• pp.315-320
• /
• 1999

Obstacle Detection System(ODS) is a essential system for automated vehicle, such as AGV(Automatic Guided Vehicle), mobile robot. Automated vehicle must have a capability to detect and to avoid obstacles to guarantee a safe driving condition. To implement obstacle detection system, virtual bumper concept adapted. Like real bumper in a car, such as in the truck, it protects vehicle from collision using laser distance sensor. When an obstacle(such as other vehicle, building, etc) intrudes this virtual bumper area, a virtual force is calculated and produces necessary strategy to be able to avoid collision. In this paper, simplified virtual bumper concept is presented, and various problems when happens to implement are discussed.