http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
송광열(Gwangyul Song),이기용(Kiyong Lee),이준웅(Joonwoong Lee),윤팔주(Paljoo Yoon) 한국자동차공학회 2007 한국자동차공학회 Symposium Vol.- No.-
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.
영상의 컬러 코렐로그램을 이용한 Free driving space 결정
최지혜(Ji-Hye Choi),송광열(Gwangyul Song),이준웅(Joonwoong Lee) 한국자동차공학회 2009 한국자동차공학회 학술대회 및 전시회 Vol.2009 No.11
This paper proposes an algorithm determining a free driving space for autonomously navigating ground vehicles. Based on color image processing, the algorithm is composed of two main concepts: 1) color correlogram and 2) random forests. As the outdoor images are on the many effects of various light sources, defining a feature descriptor is quite important. Chosen as a color feature, the color correlogram is built of hue and saturation components of a color image. The correlogram provides spatial relationships as well as color frequencies. In particular, well-known to the statistical learning method, random forests exploiting the correlograms classify the patches of a given road-image into patches of road surface and patches of non-road surface. Experimental results show that the proposed algorithm successfully generates free driving spaces in real-time.
카메라 기반 자율주행 인식률 평가를 위한 실도로 영상 DB 구축 및 평가 방안 연구
이정우(Jungwoo Lee),임동준(Dongjun Lim),송광열(Gwangyul Song),노형주(Hyeongju Noh),양현아(Hyunah Yang) 한국자동차공학회 2019 한국자동차공학회 부문종합 학술대회 Vol.2019 No.5
In this paper, we propose a method of generating and evaluating reference values for real - time road image DB construction and quantitative recognition rate evaluation of camera - based autonomous drive recognition rate. As a result of government project to develop camera based recognition algorithm for autonomous driving, we developed a system that can store LIDAR, DGPS and IVN information in real time which can provide original data and reference value of multi-channel camera module. The acquired DB performs the reference value generation operation separately for improving the algorithm performance and quantitatively evaluating the recognition rate. The reference values that are processed frame by frame are stored in the upper pixel of the image converged into one image file and provided to the algorithm developer. In the future, the real road image DB and reference value of the common format are used for algorithm development and performance verification.