http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
김경민,이병진,류경,박귀태,Kim, Kyoung-Min,Lee, Byung-Jin,Lyou, Kyoung,Park, Gwi-Tae 제어로봇시스템학회 1997 제어·로봇·시스템학회 논문지 Vol.3 No.5
This paper presents the automatic recognition algorithm of the license number in on vehicle image. The proposed algorithm uses the correlation coefficient and Hough transform to detect license plate. The m/n ratio reduction is performed to save time and memory. By the correlation coefficient between the standard pattern and the target pattern, licence plate area is roughly extracted. On the extracted local area, preprocessing and binarization is performed. The Hough transform is applied to find the extract outline of the plate. If the detection fails, a smaller or a larger standard pattern is used to compute the correlation coefficient. Through this process, the license plate of different size can be extracted. Two algorithms to each separate number are proposed. One segments each number with projection-histogram, and the other segments each number with the label. After each character is separated, it is recognized by the neural network. This research overlomes the problems in conventional methods, such as the time requirement or failure in extraction of outlines which are due to the processing of the entire image, and by processing in real time, the practical application is possible.
김경민,류경 여수대학교 1998 論文集 Vol.13 No.2
In a still mill, steel slabs often have surface defects which have to be detected and classified. So, the incentive to use the effective surface defect inspection algorithm and the classification algorithm arises. In case of surface defect images with small defects, it is difficult that detect precisely the shape of defects by one-stage preprocessing, since images contain noises which are caused by the illumination effect or other. One of the factors making classification is the number of classes to be distinguished. This is due not only to the obvious effect that with increasing number of classes the potential grows for conflicts between the classes but also to the increasing complexity of the optimum discriminant functions prob(k/v). One way to escape this consequence of a growing number of classes is to organize the classification system as a hierarchy of classifiers. So, to investigate the performance of a hierarchical classifier, we show that a hierarchical classifier has better classification rate than one-step classifier, which classify classes with one classifier, using Polynomial Regression Classifier among many classifiers.