http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
이철학,김상운 명지대학교 산업기술연구소 2006 産業技術硏究所論文集 Vol.25 No.-
Image segmentation is an essential preliminary step in pattern recognition and scene analysis problems. Thresholding is a fundamental approach to segmentation that utilizes a significant degree of pixel popularity or intensity. Otsu's thresholding is one of the most well-known methods proposed in the literature. In the Otsu's method, the normalized histogram was employed as a discrete probability density function and a criterion of maximizing the between-class variance of pixel intensity was utilized to choose a threshold value for segmentation. In this paper, first of all, we simplify the between-class variance by which we can easily compute the criterion. Then, we propose a simple thresholding method for finding multi-level threshold values by extending the Otsu's method. Our experimental results how a possibility that the proposed method could be used efficiently for image segmentation.
이철학,김상운 明知大學校 産業技術硏究所 2008 産業技術硏究所論文集 Vol.27 No.-
Among the facial components, the eyes play an important role in recognizing human faces. This paper proposes an effective method of detecting the eyes from facial images, where the detection is carried out two steps. First of all, eye-windows(EW) containing the eyes and eyebrows are extracted with the vertical/horizontal projections. After then, the positions of the eyes are detected by applying the normalized cross correlation (CC)method to the eye window extracted in the first step. While searching the EWs, we have took smoothing firstly to reduce influence of noise and sharper bounds. From the experimental results tor the well-known face databases, we have confirmed a possibility that the proposed method, which is hierarchically combined with EW and CC in sequence, could be available one for appearance based face retrieval.
문장패턴 변환방법을 이용한 韓-中-日 수화 통신 시스템에 관한 기초연구
이철학,김상운 明知大學校 産業技術硏究所 2003 産業技術硏究所論文集 Vol.22 No.-
Abstract - In this paper, we implemented and investigated a sign-language communication system among Korean, Chinese, and Japanese as an extension of the sign-language communication system between Korean and Japanese. Using the system, the deaf-mute can chat each other with sign-language among the three country having different languages on the Internet. Additionally, a translation of sigh-language enables the people with different languages to communicate each other. The system employed the client-server architecture, where the client inputs the sentence by keyboard, and sends it to the server. Then the server translates it into target sign-language parameters and returns them to the group of clients. Finally, the client re-generates the sigh-language animation with the received parameters. At that time, we have used a translation table of language patterns to translate each other among different language structures.
이철학,김상운 대한전자공학회 2006 電子工學會論文誌-CI (Computer and Information) Vol.43 No.5
Thresholding is a fundamental approach to segmentation that utilizes a significant degree of pixel popularity or intensity. Otsu's thresholding employed the normalized histogram as a discrete probability density function. Also it utilized a criterion that maximizes the between-class variance of pixel intensity to choose a threshold value for segmentation. However, the Otsu's method has a disadvantage of repeatedly searching optimal thresholds for the entire range. In this paper, a simple but fast multi-level thresholding approach is proposed by means of extending the Otsu's method. Rather than invoke the Otsu's method for the entire gray range, we advocate that the gray-level range of an image be first divided into smaller sub-ranges, and that the multi-level thresholds be achieved by iteratively invoking this dividing process. Initially, in the proposed method, the gray range of the object image is divided into 2 classes with a threshold value. Here, the threshold value for segmentation is selected by invoking the Otsu's method for the entire range. Following this, the two classes are divided into 4 classes again by applying the Otsu's method to each of the divided sub-ranges. This process is repeatedly performed until the required number of thresholds is obtained. Our experimental results for three benchmark images and fifty faces show a possibility that the proposed method could be used efficiently for pattern matching and face recognition. 스레쉬홀딩(thresholding)은 영상 화소의 군집이나 강도를 이용하여 영상을 분할하는 기본 기술이다. Otsu의 스레쉬홀딩 방법에서는 정규화 된 히스토그램을 이산 밀도함수로 보아 화소의 클래스 간 분산을 최대화시키는 판별식을 이용한다. 그러나 Otsu의 방법에서는 여러 객체로 이루어진 영상에서 최적의 스레쉬홀드를 찾기 위하여 그레이레벨 전 구간에 대해 모든 가능한 분산 값을 반복적으로 계산해 보아야 하기 때문에 계산 시간이 길게 걸리는 문제가 있다. 본 논문에서는 Otsu의 방법을 개선하여 간단하지만 고속으로 멀티-레벨의 스레쉬홀드 값을 구할 수 있는 방법을 제안한다. 전체 그레이 구간 영역에 대하여 Otsu의 방법을 적용시키기 보다는 먼저 그레이 영역을 작은 부분-구간으로 나눈 다음 Otsu의 방법을 적용시키는 처리를 반복하여 원하는 개수의 스레쉬홀드를 구하는 방법이다. 본 제안 방법에서는 맨 처음 대상 영상의 그레이 구간을 2부류로 나눈다. 이 때, 분할을 위한 스레쉬홀드는 전 구간을 대상으로 Otsu의 방법을 적용하여 구한다. 그 다음에는 전체 구간이 아닌 분할된 부분-구간을 대상으로 Otsu의 방법을 적용하여 두 부류를 4부류로 나눈다. 이와 같은 처리를 원하는 개수의 스레쉬홀드를 얻을 때 까지 반복한다. 세 종류 벤취마크 영상과 50개 얼굴영상에 대해 실험한 결과, 제안 방법은 대상 영상을 특성에 맞게 고속으로 잘 분할하였으며, 패턴 매칭이나 얼굴인식에 이용될 수 있는 가능성을 확인하였다.
이철학(Zhe-Xue Li),김상운(Sang-Woon Kim) 대한전기학회 2006 정보 및 제어 심포지엄 논문집 Vol.2006 No.1
For pattern matching, an object image should be segmented and analyzed for the first time. Thresholding is a fundamental approach to segmentation that utilizes a significant degree of pixel popularity or intensity. Otsu's thresholding is one of the most well-known methods proposed in the literature. However, the method has a disadvantage of repeatedly searching the optimal thresholds for the entire region. To overcome this problem, a number of methods have been proposed. In this paper, we propose a simple and fast thresholding method of finding multi-level threshold values by extending the Otsu's method. Our experimental results for the benchmak images show a possibility that the proposed method could be used efficiently for pattern matching.