http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
이준웅(J.W.Lee),장인수(I.S.Jang),강동중(D.J.Kang),권인소(I.S.Kweon) 한국자동차공학회 1997 한국자동차공학회 춘 추계 학술대회 논문집 Vol.1997 No.11_2
This paper presents an algorithm for road lane recognition. The algorithm composes of the estimation of 1D function of the histogram of gradient magnitude with respect to the image gradient direction by a recursive filter, determination of a band region as the search region for the road lane, construction of 10 function of the histogram of gradient magnitude of edge pixels in the band region, and recognition of road lane by the convolution of two functions. We also initially estimate the vanishing point and the location of road lane for host vehicle based on calibrated camera parameters. To prevent the failure of road lane: detection due to the disappearing of a lane mark and the lateral motion of host vehicle, this algorithm utilizes the recursive ftltering. The newly proposed approach provides the reliable detection of road lanes of natural outdoor scenes even in the presence of various noises.
김형우(H.W.Kim),이운근(U.K.Yi),이준웅(J.W.Lee),방중철(J.C.Bang) 한국자동차공학회 1997 한국자동차공학회 춘 추계 학술대회 논문집 Vol.1997 No.11_2
The development of an active safety system for a vehicle is indispensable considering the worldwide trend of developing safety system, the traffic conditions and the increasing rate of traffic accidents. This paper describes the configuration of a laser radar system, provides the several methods for measuring the distance between the leading and host vehicles and suggests a new method for measuring the distance. The method is cost-effective and increases the distance resolution. It is the primary goal to obtain more accurate distance data by improving and modifying the disadvantages occurred from the classical integration method.
장인수(I.S.Chang),금동교(D.K.Keum),이준웅(J.W.Lee),남궁억(E.NamGoong) 한국자동차공학회 1998 한국자동차공학회 춘 추계 학술대회 논문집 Vol.1998 No.11_2
This paper presents a fusion algorithm of lane information obtained by the image processing and the vehicle location information detected by a radar system for CWS (collision warning system). In the algorithm, the lane information expressed on the image plane has been transformed to the 3D space in which radar data is defined. Experimental results show that the proposed fusion algorithm provides good evidences for detecting the target vehicle running in front on the same lane.<br/>