http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
전병태 한국지식정보기술학회 2018 한국지식정보기술학회 논문지 Vol.13 No.2
Many studies using the latest nighttime vehicle images are under way. Research on nighttime imaging has been conducted in the field of ITS and is being used to collect vehicle information. When the unusual headlight is shining on the eyes, it is said to be invisible for 4.5 seconds. In the case of a vehicle traveling at 80 km, it will run at 100 m without any defenses. Thus, in the case of an unsteady headlight, the driver of the other side of the opponent may have a serious problem in driving even if the driver takes the light for only a few seconds. In this paper, we propose a method to extract headlight area information from input nighttime vehicle image and determine whether it is a normal / abnormal headlight. After the binarization is performed on the whole image, the large area is extracted to the headlight area through the condition check. The extracted headlight area is compared with the pre-calculated normal headlight area size to determine normal / abnormal. Experimental results show that the detection of unsteady headlights is good. This study can be applied to search for illegal headlights in the future.
전병태,김병천 한국통신학회 2004 韓國通信學會論文誌 Vol.29 No.7A
본 논문에서는 일반적인 비디오 압축에 사용되는 블록 움직임 예측 방법을 위한 육각형 탐색 형태를 이용한 적응형 블록 정합 탐색 알고리즘(Adaptive hexagon based search : AHBS)을 제안한다. 제한하는 알고리즘은 다이아몬드 형태를 이용한 초기 탐색 과정과 제안한 두개의 육각 형태를 이용하여 적응적으로 탐색하는 과정으로 이루어져 있으며, 육각형 탐색 형태의 중앙값에 블록 정합을 위한 목적 함수의 최소간이 발생할 때 그 과정을 종료하는 것으로 구성되어 있다. 또한, 탐색 과정에 있어 적용할 육각형 탐색 형태의 결정은 이전 탐색 과정에서 발생한 최소 값의 위치에 따라 이루어진다. 제안한 알고리즘의 성능측정은 전역탐색 방법을 포함 기존의 다양한 고속 탐색 방법들과 전역 탐색 방법의 결과와의 비교를 통하여 이루어졌다. 기존의 고속 탐색 방법에 비하여 본 논문에서 제안한 방법의 성능이 우수하고 그 수행 속도 또한 개선된 것을 실험 결과 알 수 있다. An adaptive hexagon based search(AHBS) algorithm is proposed in this paper to perform block motion estimation in video coding. The AHBS evaluates the value of a given objective function starting from a diamond-shaped checking block and then continues its process using two hexagon-shaped checking blocks until the minimum value is found at the center of checking blocks. Also, the determination of which checking block is used depends on the position of minimum value occurred in previous searching step. The AHBS is compared with other fast searching algorithms including full search(FS). Experimental results show that the proposed algorithm provides competitive performance with slightly reduced computational complexity.
전병태 한국지식정보기술학회 2017 한국지식정보기술학회 논문지 Vol.12 No.3
Recently, research on the extraction of vehicle information has been actively conducted. Vehicle information has many applications such as unmanned management system, and most of the methods of vehicle information collection are non-contact method. In this paper, we propose a method to extract headlight area information and license plate area information. After the binarization is performed on the whole image, the large area is extracted to the headlight area through the condition check. The license plate area extraction method consists of two stages: a license plate candidate region extraction step and a vehicle candidate region verification step. The license plate candidate region binarizes the boundary detection result and extracts the clustered region as a candidate region. The extracted candidate region of the vehicle is reduced by considering the position of the headlight and the region of the license plate is extracted by using the candidate region verification method in the reduced range. Experimental results show that the detection results of the vehicle headlight area and the license plate area are good. The extracted headlight area can be utilized in the search for illegal headlights in the future.