• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅳ
• /
• pp.2084-2087
• /
• 2003

본 논문에서는 MPEG 압축 도메인 상에서 카메라 움직임 정보를 추출하는 효과적인 방법을 제안한다. 카메라 움직임 정보는 동영상에서의 주요 장면과 프레임간의 관계를 기술할 수 있는 실마리를 제공한다. 본 논문에서는 MPEG Video의 모션벡터를 이용하여 카메라 움직임 정보를 추출한다. 카메라 움직임에 따라 모션벡터는 특징적으로 분포하는 특성이 있다. 본 논문에서는 이러한 특징들을 이용하여 MPEG 모션벡터의 방향성과 크기를 이용하여 각 모션벡터끼리 교차점과 평행성분을 구한다. 그리고 이것을 이용하여 같은 교차점과 평행성분끼리 모션벡터 Clustering 을 수행한다 본 논문에서는 클러스터링 된 모션벡터를 Fuzzy inference rule을 이용하여 카메라 움직임이 Zoom, Pan, Tilt 인지 여부를 판단한다. 실험은 전통적인 방법 중에 하나인 Affine Model 방법과 비교하며 본 논문의 방법이 어느정도 우수함을 입증한다.

• 대한수학회보
• /
• 제57권1호
• /
• pp.139-165
• /
• 2020

Mirror symmetry for del Pezzo surfaces was studied in [3] where they suggested that the mirror should take the form of a Landau-Ginzburg model with a particular type of elliptic fibration. This argument came from symplectic considerations of the derived categories involved. This problem was then considered again but from an algebro-geometric perspective by Gross, Hacking and Keel in [8]. Their construction allows one to construct a formal mirror family to a pair (S, D) where S is a smooth rational projective surface and D a certain type of Weil divisor supporting an ample or anti-ample class. In the case where the self intersection matrix for D is not negative semi-definite it was shown in [8] that this family may be lifted to an algebraic family over an affine base. In this paper we perform this construction for all smooth del Pezzo surfaces of degree at least two and obtain explicit equations for the mirror families and present the mirror to dP₂ as a double cover of ℙ².

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.54.4-54
• /
• 2001

This paper proposes an adaptive control scheme for an autopilot design of Relaxed-Static-Stability(RSS) Missiles with saturating actuator. The feedback linearization controller eliminates nonlinear terms in RSS missile dynamics and makes the entire system linear. But modeling errors, disturbances and the nonlinear mismatch due to input constraints exert a bad influence on the performance of the feedback linearization controller Thus, first, we derive a parametric affine uncertainty model with modeling errors and disturbances. Then an adaptive control law with anti-windup scheme is developed, where the bounds of uncertainties are estimated with adaptive laws. The proposed adaptive controller can remove the bad effects of uncertainties, of disturbances, and of saturating actuator ...

• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1439-1447
• /
• 2018

Palm image acquisition without contact has advantages in user convenience and hygienic issues, but such images generally display more image variations than those acquired employing a contact plate or pegs. Therefore, it is necessary to develop a palmprint identification method which is robust to affine variations. This study proposes a deep learning approach which can effectively identify contactless palmprints. In general, it is very difficult to collect enough volume of palmprint images for training a deep convolutional neural network(DCNN). So we adopted an approach to use a pretrained DCNN. We designed two new DCNNs based on the VGGNet. One combines the VGGNet with SVM. The other add a shallow network on the middle-level of the VGGNet. The experimental results with two public palmprint databases show that the proposed method performs well not only contact-based palmprints but also contactless palmprints.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 춘계학술발표대회
• /
• pp.497-500
• /
• 2022

High-quality images and videos are being generated as technologies for deep learning-based image style translation and conversion of static images into dynamic images have developed. However, it takes a lot of time and resources to manually transform images, as well as professional knowledge due to the difficulty of natural image transformation. Therefore, in this paper, we study natural style mixing through a style conversion network using GAN and natural dynamic image generation using the First Order Motion Model network (FOMM).

• 대한전기학회논문지:시스템및제어부문D
• /
• 제51권5호
• /
• pp.212-219
• /
• 2002

In this paper, the epipolar geometry, genera17y used as a pin-hole camera model, is newly adapted to our proposed method that enables the affine reconstruction of the 3D object from two projected views. The proposed method models the projective projection of inherent X-ray imaging system, obviates the need to attach artifirially constructed material on the body, and requires none of the prior-knowledge regarding to intrinsic and extrinsic parameters of two X-ray imaging systems. The optimum numerical solution is obtained by applying the least mean square estimator to corresponding points on two projected X-ray planes. The performance of this proposed method is Quantitatively analyzed using computer synthesized model of Cochlear implantation electrodes. In simulated experiments, the propnsed method is insensitive to the added random noise, the scaling factor change, the center point change, and rotational angular change between two projection planes, as well as enables the stable 3D reconstruction in least square sense even in worst testing cases.

• 제어로봇시스템학회논문지
• /
• 제3권3호
• /
• pp.272-279
• /
• 1997

In this paper, a fuzzy generalized predictive control (FGPC) for non-linear plants is proposed. In the proposed method, the receding horizon control is applied to the control part, while fuzzy systems are used for the predictor part. It is suggested that the fuzzy predictor is time-varying affine with respect to input variables for easy computation of control inputs. Since the receding horizon control can be obtained only with a predictor instead of a plant model, the fuzzy predictor is obtained directly from input-output data without identifying a plant model. A parameter estimation algorithm is used for identifying the fuzzy predictor. The control inputs of the FGPC are computed by minimizing a receding horizon cost function with predicted plant outputs. The proposed controller has a similar architecture to the generalized predictive control (GPC) except for the predictor synthesis method, and thus may possess inherent good properties of the GPC. Computer simulations show that the performance of the FGPC is satisfactory.

• Tribology and Lubricants
• /
• 제30권6호
• /
• pp.378-383
• /
• 2014

This study predicts the of the hysteretic friction of a rubber block sliding on an SMA asphalt road. The friction of filled rubber on a rough surface is primarily determined by two elements:the viscoelasticity of the rubber and the multi-scale perspective asperities of the road. The surface asperities of the substrate exert osillating forces on the rubber surface leading to energy dissipation via the internal friction of the rubber when rubber slides on a hard and rough substrate. This study defines the power spectra at different length scales by using a high-resolution surface profilometer, and uses rubber and road surface samples to conduct friction tests. I consider in detail the case when the substrate surface has a self affine fractal structure. The theory developed by Persson is applied to describe these tests through comparison with the hysteretic friction coefficient relevant to the energy dissipation of the viscoelastic rubber attributable to cyclic deformation. The results showed differences in the absolute values of predicted and measured friction, but with high correlation between these values. Hence, the friction prediction model is an appropriate tool for separating the effects of each factor. Therefore, this model will contribute to clearer understanding of the fundamental principles of rubber friction.

• International Journal of Computer Science & Network Security
• /
• 제24권2호
• /
• pp.89-94
• /
• 2024

Indian subcontinent is a birthplace of multilingual people where documents such as job application form, passport, number plate identification, and so forth is composed of text contents written in different languages/scripts. These scripts may be in the form of different indic numerals in a single document page. Due to this reason, building a generic recognizer that is capable of recognizing handwritten indic digits written by diverse writers is needed. Also, a lot of work has been done for various non-Indic numerals particularly, in case of Roman, but, in case of Indic digits, the research is limited. Moreover, most of the research focuses with only on MNIST datasets or with only single datasets, either because of time restraints or because the model is tailored to a specific task. In this work, a hybrid model is proposed to recognize all available indic handwritten digit images using the existing benchmark datasets. The proposed method bridges the automatically learnt features of Capsule Network with hand crafted Bag of Feature (BoF) extraction method. Along the way, we analyze (1) the successes (2) explore whether this method will perform well on more difficult conditions i.e. noise, color, affine transformations, intra-class variation, natural scenes. Experimental results show that the hybrid method gives better accuracy in comparison with Capsule Network.

• 융합신호처리학회논문지
• /
• 제2권4호
• /
• pp.16-21
• /
• 2001

본 논문에서는 2매의 2차원 얼굴영상으로부터 이들의 합성하여 3차원 얼굴의 가상형상을 복원한다. 여기서 2매의 2차원 얼굴영상은 정면과 측면 영상을 사용한다. 임의의 일반 얼굴에 대한 기준모델을 정하고 이 모델에서 얼굴형상의 특징을 표현하는 귀 , 눈 코 및 입 부분에 집중적으로 특징 점을 규정한다. 그 이외에 이마 및 턱부분에도 특징 점을 지정하여 그 위치 좌표를 저장해 둔다. 그 후 정면 영상의 좌 우측에 측면영상을 대칭적으로 접속하고 영상의 기하 변환방법을 적용하여 점차적으로 합성한다. 합성을 효과적으로 처리하기 위해 측면 영상을 정면 영상에 정합될 수 있도록 기하변환을 이용한다. 이 때 나타나는 합성부분에 색상 및 명도의 차를 제거하기 위해 스므딩필터(3$\times$3 화소의 마스크)를 적용하여 자연스런 3차원 가상얼굴을 구현하게 된다. 그 결과 불특정 얼굴형상도 3차원으로 구현할수 있음을 확인하였다.