http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Optical Flow를 이용한 측후방 차량인식 시스템 개발
성준용(Junyong Sung),이경복(Kuengbok Lee),한민홍(Minhong Han) 한국자동차공학회 2006 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-
We have developed rear-side vehicle recognition system using Optical Flow Algorithm. This system detect rapidly approaching vehicle at side lane. This vehicle recognition method understand vehicle region if Optical Flow is x-axis increasing direction at left back screen or x-axis decreasing direction at right back screen. we look for centroid axis at vehicle region. This system tell driver danger if rear side vehicle penetrate within established dangerous region. The experiments show that rear-side vehicle recognition using Optical Flow can be utilized as a device which assists safety driving by detecting rear side vehicle under warning the driver of dangerous situation during lane change.
이경복(Kyungbok Lee),김영욱(Yongwok Kim),한민홍(Minhong Han) 한국자동차공학회 2006 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-
A traffic accident being happened on the road is mostly caused by rear-end collisions. In particular, the proportion of the big accident is high on the express highway. An information delivery system between cars is suggested as a method for decreasing the proportion of such big accident on the express highway. A traffic accident can be remarkably reduced if a rear car can notice the danger information of a front car in advance. If a GPS receiver and RF module are established in all the cars and the position coordinate and danger information received from the GPS receiver are sent to all the rear cars within the communication area, each of the rear cars can notice the opponent position, headway direction, and danger information by using the position coordinate and danger information received from the front car. Finally all the car drivers within the RF communication area can perceive the front danger information in advance. Accordingly, the present system delivers the front danger information to the rear car by means of a system between cars with low expense so that it can be easily employed as a device for safe driving on the express highway.
안수진(Soojin An),이경복(Kuengbok Lee),한민홍(Minhong Han) 한국자동차공학회 2006 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-
In this paper, we propose lane detection using Clustering as powerful algorithm. The process of land detection algorithm using Clustering that we are proposed has six steps bigly. first, divide the picture into right part and left part after the image imformations were received by camera then dedect each edge pixel through Sobel algorithm. second, make the clusters Using distance of pixel coordinates for removing noise. third, to eliminate noise, limited the cluster height. there again, formulate a straight line equation about all clusters. forth, find out every width between right side of clusters' pixel value and left side of cluster's pixel value after determine up-pixel value and low-pixel value on every cluster at last, according to difference virtual width between actual lane'width, judge true or false. we tested this algorithm various image such as the inside road and the northern river side road in Seoul. The percentage of correct lane detection is over 98%.
성준용(Jun-Yong Sung),변재민(Jae-Min Byun),노명찬(Myung-Chan Roh),김성훈(Sung-Hoon Kim),김중배(Joon-Bae Kim),한민홍(MinHong Han) 대한전자공학회 2010 대한전자공학회 학술대회 Vol.2010 No.6
We have developed an obstacle detection method using two cameras arranged vertically inside a vehicle. One camera is installed horizontally toward the front of a vehicle while another camera is installed, tilted toward the road surface, at a position which is higher than the first camera. From each image obtained from the two cameras, the farthest obstacle distance from the vehicle is calculated, and their calculated distance values are compared. If there exists a substantial difference in distance values, an obstacle is assumed to exist protruding from the road surface, otherwise no obstacle exists. The method is in use satisfactorily for adapted cruise control(ACC) purpose of a fully autonomous vehicle.