• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.77-84
• /
• 2022

In this paper, we propose a novel unsupervised feature selection method. Conventional unsupervised feature selection method defines virtual label and uses a regression analysis that projects the given data to this label. However, since virtual labels are generated from data, they can be formed similarly in the space. Thus, in the conventional method, the features can be selected in only restricted space. To solve this problem, in this paper, features are selected using orthogonal projections and low-rank approximations. To solve this problem, in this paper, a virtual label is projected to orthogonal space and the given data set is also projected to this space. Through this process, effective features can be selected. In addition, projection matrix is restricted low-rank to allow more effective features to be selected in low-dimensional space. To achieve these objectives, a cost function is designed and an efficient optimization method is proposed. Experimental results for six data sets demonstrate that the proposed method outperforms existing conventional unsupervised feature selection methods in most cases.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.11
• /
• pp.621-628
• /
• 2003

This paper addresses a method of enabling objects in an image to have apparent 3D motion. Many researchers have solved this issue by reconstructing 3D model from several images using image-based modeling techniques, or building a cube-modeled scene from camera calibration using vanishing points. This paper, however, presents the possibility of image-based motion without exact 3D information of scene geometry and camera calibration. The proposed system considers the image plane as a projective plane with respect to a view point and models a 2D frame of a projected 3D object using only lines and points. And a modeled frame refers to its vanishing points as local coordinates when it is transformed.

• Journal of Korean Association for Spatial Structures
• /
• v.4 no.2 s.12
• /
• pp.73-80
• /
• 2004

In this paper, a study of damage detection for 2-Dimensional Truss Structures using the parametric projection filter theorr is presented. Many researchers are interested in inverse problem and one of solution procedures for inverse problems that are very effective is the approach using the filtering algorithm in conjunction with numerical solution methods. In filtering algorithm, the Kalman filtering algorithm is well known and have been applied to many kind of inverse problems. In this paper, the Parametric projection filtering in conjunction with structural analysis is applied to the identification of damages in 2-D truss structures. The natural frequency and modes of damaged truss model are adopted as the measurement data. The effectiveness of proposed method is verified through the numerical examples.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.3
• /
• pp.547-554
• /
• 2014

This paper discusses a method of estimating variance components for random effects model. Henderson's method I and III are discussed for the esimation of variance components. This paper shows how to use projections instead of using Henderson's methods for the calculation of sums of squares which are quadratic forms in the observations. It also discusses that eigenvalues can be used for getting the expectations of sums of squares in place of using the method of Hartley's synthesis. It shows the suggested method is much more effective than those methods.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.107-107
• /
• 2016

표면영상유속계는 영상분석을 이용하여 홍수시 하천 수표면 유속을 측정하는 비접촉식 유속측정장치이다. 때문에 안전하고 편하게 홍수시 유속을 측정할 수 있으나, 이를 위해서는 영상과 실세계와의 좌표변환을 위한 참조점 측량이 반드시 필요하였다. 좌표 변환에 8-변수 사영변환을 이용할 경우 최소한 4개의 참조점이 필요하다. 그러나 홍수시 참조점 측량은 불가능에 가까우며, 홍수전후의 측량도 매우 번거로운 일이다. 본 연구에서는 참조점을 이용하지 않고, 좌표변환 관계를 구성할 수 있는 카메라 모형(camera model)을 구성하였다. 여기서 카메라 모형은 실세계 좌표(world coordinates)를 영상좌표(image coordinates)로 변환해 주는 관계를 말한다. 이 카메라 모형에 필요한 외부 변수는 하천수표면과 카메라와의 높이 및 카메라의 두 가지 경사각뿐이다. 여기에 일반적인 카메라 보정에 이용하는 방법으로 구한 카메라 내부 변수를 결합하면 된다. 이 모형은 표면영상유속계 장비로 스마트폰을 이용할 때 간편하게 적용할 수 있으며, 별도의 경사계만 부착하면 일반적인 CCTV나 캠코더를 이용할 때도 적용할 수 있다. 이 카메라 모형과 종래 사용하던 8-변수 사영변환에 의한 좌표변환관계를 비교한 결과 상호간 오차가 거의 없이 적용할 수 있었다. 또한, 스마트폰 표면영상유속계와 열영상 표면영상유속계에 적용한 결과 종전보다 훨씬 적용이 간편하며, 정확도 또한 거의 차이가 없이 적용할 수 있었다.

• Korean Journal of Heritage: History & Science
• /
• v.38
• /
• pp.179-222
• /
• 2005

Studies prosecuted on relics in those 17 bronze workshops that have been thus far excavated show that these workshops may be roughly classified into a royal workshop, a state-operated workshop and a private workshop depending upon by whom they were operated. Workshops in the Gyeongju area developed from a small royal handicraft manufacturing to a large state-operated handicraft manufacturing scale, and then later on gradually changed to a private handicraft manufacturing industry. The royal bronze workshops were operated in a small scale, as shown from the relics excavated at Wolseong(月城), Imhaejeonji(Anapji:雁鴨池) and their neighborhood places around Hwangnam_dong(皇南洞). The state-operated bronze workshops are concentrated upon one point around Dongcheon-dong(東川洞), Gyeongju city. On the other hand, in the state-operated workshop stage, a broad street was built by a workshop, which is presumed to aim to thoroughly transport materials needed for the workshop. And the point that wastes from bronze workshops were used for road repairs indicates that road repair works were carried at the bronze workshops near the road. The private workshop as a new type of workshop was operated by the aristocracy. For that purpose, craftsmen belonging to state-operated workshops or individual artisans were absorbed into the aristocracy-operated workshops. These types of workshops were pervaded throughout the city. When private workshops came to emerge in the houses of the aristocracy, the operating subjects of workshops began to change from state-operated to private workshops. Temple workshops were located at a Buddhist temple within the Court and directly produced things needed for the court, including bronze foundries. As aforementioned, through the presence of bronze workshops operated in the Silla Court, we can identify the relationships between their technical level and trading areas and among their origin, supply and demand sources, along with phases of social life in those days.

• Bulletin of the Korean Mathematical Society
• /
• v.7 no.1
• /
• pp.27-34
• /
• 1970

• Proceedings of the Korean Society for History of Mathematics Conference
• /
• 1998.11a
• /
• pp.11.1-11.1
• /
• 1998

• Journal of the Korean Society of Clothing and Textiles
• /
• v.2 no.2
• /
• pp.269-276
• /
• 1978

This thesis studied the costumes of Gwan-Rye(관예). the lowest ranking officials of Yi-Dynasty. It researched the written materials and compared it with relics of Gwan-Rye costumes. The gists of the result are; 1) The Koryo tradition of Gwan-Rye costumes continued until the early days of Yi-Dynasty. 2) The Gwan-Rye costumes can be classified into two styles. One is Sa-Ryung(사영) style, which consisted of Dan-Ryung(단령) robe and Jo-Geon(조시) as head gear. Na-Jang(라장) style. the another, consisted of Dan-Ryung. Ban-Bi-Ui(반비의) and Jo-Geon. 3) These styles changed around the days of Yeon-San(연산), the 10th King of the Dynasty. The Na-Jang of later days wore Cheop-Ri(첩리) robe instead of Dan-Ryung. And the Sa-Ryung costume was devided into three different styles. They are (1) Gat(립)- Cheop-Ri (2) Bung-Geo-Ji(단립)-Chang-Ui(창의) (3) Gat- Kwoe-Ja(쾌자) styles.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.4
• /
• pp.57-57
• /
• 2001

본 논문은 영상 시퀸스로부터 이미지 모자이킹의 구성을 위한 효율적인 알고리즘을 기술한다. 영상의 기하학적인 특징을 이용하거나 비선형 방정식을 풀었던 기존의 알고리즘과는 달리, 제안한 알고리즘은 4개의 유사특징점을 이용해 영상간 사영 변환식의 8개 파라미터를 직접 계산한다. 본 논문에서 정의된 유사특징점은 영상의 그레이레벨의 분산을 기반으로 하고, 두 영상의 중첩 영역에서만 결정된다. 또한 선택된 4개의 유사특징점에 대한 대응점 검출을 위해 카메라 이동 및 조명 변화에 의한 영상의 변화를 고려한 블록 정합 알고리즘을 적용한다. 제안된 알고리즘은 다양한 영상에 적용하여 그 성능을 평가하였다. 모의 실험 결과는 제안된 알고리즘이 기존의 알고리즘에 비해 계산량을 감소시키면서, 정확한 사영 변환식을 유도하여 모자이킹 영상을 구성하는 것을 보여주고 있다.