• 정보보호학회논문지
• /
• 제29권2호
• /
• pp.347-363
• /
• 2019

사람의 DNA가 변하지 않는 것과 같이 사이버상의 악성코드도 변하지 않는 고유의 행위 특징을 갖고 있다. APT(Advanced Persistent Threat) 공격에 대한 방어수단을 사전에 확보하기 위해서는 악성코드의 악성 행위 특징을 추출해야 한다. 이를 위해서는 먼저 악성코드 간의 유사도를 계산하여 유사한 악성코드끼리 분류할 수 있어야 한다. 본 논문에서는 Windows OS 상에서 동작하는 악성코드 간의 유사도 계산 방법으로 'TF-IDF 코사인 유사도', 'Nilsimsa 유사도', '악성코드 기능 유사도', 'Jaccard 유사도'를 사용해 악성코드의 유형을 예측해보고, 그 결과를 보인다. 실험결과, 유사도 계산 방식마다 악성코드 유형에 따라 예측률의 차이가 매우 컸음을 발견할 수 있었다. 모든 결과에 월등한 정확도를 보인 유사도는 존재하지 않았으나, 본 실험결과를 이용하여 특정 패밀리의 악성코드를 분류할 때 어떤 유사도 계산 방식을 활용하는 것이 상대적으로 유리할지를 결정할 때 도움이 될 것으로 판단된다.

• 한국컴퓨터정보학회논문지
• /
• 제14권8호
• /
• pp.25-31
• /
• 2009

RGB 모델을 통한 정적인 추론 규칙을 적용한 기존의 색채 정보 인식 방법은 RGB 모델이 가지는 인간 시각과의 괴리감과 특정한 환경에서만 적용할 수 있는 문제점이 있다. 본 논문에서는 HSI 모델을 적용하여 색채에 대한 인간 인식 과정과 유사한 형태의 추론 방식과, 사용자에 의해서 추론 규칙을 추가, 수정, 삭제 할 수 있는 방법을 제안한다. 본 논문에서는 각각의 H, S, I 소속 구간에 대하여 H는 Sine, Cosine 함수를 사용하여 소속 구간을 설계하며, S, I는 삼각형 타입의 소속 함수로 설계한다. 설계된 각각의 소속 구간에 대하여 소속 구간 병합을 적용하여 소속도를 계산하고, 계산된 결과들은 미리 제시된 추론 규칙에 적용하여 색채를 추론한다. 제안된 두가지 방법을 적용하여 실험한 결과, 기존의 방법보다 제안된 방법이 비교적 직관적이며 효율적인 형태로 결론을 도출할 수 있음을 확인하였다.

• 한국멀티미디어학회논문지
• /
• 제21권8호
• /
• pp.952-959
• /
• 2018

Clustering is one of the most fundamental algorithms in machine learning. The performance of clustering is affected by the distribution of data, and when there are more data or more dimensions, the performance is degraded. For this reason, we use a stacked auto encoder, one of the deep learning algorithms, to reduce the dimension of data which generate a feature vector that best represents the input data. We use k-means, which is a famous algorithm, as a clustering. Sine the feature vector which reduced dimensions are also multi dimensional, we use the Euclidean distance as well as the cosine similarity to increase the performance which calculating the similarity between the center of the cluster and the data as a vector. A deep clustering networks combining a stacked auto encoder and k-means re-trains the networks when the k-means result changes. When re-training the networks, the loss function of the stacked auto encoder and the loss function of the k-means are combined to improve the performance and the stability of the network. Experiments of benchmark image ad document dataset empirically validated the power of the proposed algorithm.

• 한국환경과학회지
• /
• 제4권5호
• /
• pp.41-41
• /
• 1995

In order to discuss the differences among the SMP(Surface Moisture Availability Parameter), by previous researchers on the basis of their own theoretical and empirical background, we assessed the SMP according to the soil types and volumetric soil water contents. The results are as follows. There are differences among all the five SMAPs. There''s a tendency that the larger grain size, the higher value of parameters. And they divided into two groups for their value: one group has parameters with exponential function and the other with cosine and linear function. The maximum difference between the two groups appears when the volumetric soil water contents are 0.07$m^3m^{-3}$ for sand, 0.l1$m^3m^{-3}$ for loam, 0.12 for clay, and 0.13$m^3m^{-3}$ for silt loam. So, these differences must be considered when we estimate the surface evaporation rate. From field data, the paddy field soil around Junam reservoir is classified as a silt has high wetness, 0.56. So, the parameter obtained from the field measurement is much higher than that of Clapp and Hornberger(1978)''s Table. This study treated the SMP for a certain point of time in winter season. But if we measured the soil water contents continuously, we could obtain better time-dependent parameter.

• 한국환경과학회지
• /
• 제4권5호
• /
• pp.427-435
• /
• 1995

In order to discuss the differences among the SMP(Surface Moisture Availability Parameter), by previous researchers on the basis of their own theoretical and empirical background, we assessed the SMP according to the soil types and volumetric soil water contents. The results are as follows. There are differences among all the five SMAPs. There's a tendency that the larger grain size, the higher value of parameters. And they divided into two groups for their value: one group has parameters with exponential function and the other with cosine and linear function. The maximum difference between the two groups appears when the volumetric soil water contents are 0.07m3m-3 for sand, 0.l1m3m-3 for loam, 0.12 for clay, and 0.13m3m-3 for silt loam. So, these differences must be considered when we estimate the surface evaporation rate. From field data, the paddy field soil around Junam reservoir is classified as a silt has high wetness, 0.56. So, the parameter obtained from the field measurement is much higher than that of Clapp and Hornberger(1978)'s Table. This study treated the SMP for a certain point of time in winter season. But if we measured the soil water contents continuously, we could obtain better time-dependent parameter. Key words : SMAP(Surface Moisture Availability Parameter), Paddy field, Volumetric soil water content, Evaporation, Capillary potential.

• Geomechanics and Engineering
• /
• 제16권2호
• /
• pp.141-150
• /
• 2018

In this current work a quasi 3D "trigonometric shear deformation theory" is proposed and discussed for the dynamic of thick orthotropic plates. Contrary to the classical "higher order shear deformation theories" (HSDT) and the "first shear deformation theory" (FSDT), the constructed theory utilizes a new displacement field which includes "undetermined integral terms" and presents only three "variables". In this model the axial displacement utilizes sinusoidal mathematical function in terms of z coordinate to introduce the shear strain impact. The cosine mathematical function in terms of z coordinate is employed in vertical displacement to introduce the impact of transverse "normal deformation". The motion equations of the model are found via the concept of virtual work. Numerical results found for frequency of "flexural mode", mode of shear and mode of thickness stretch impact of dynamic of simply supported "orthotropic" structures are compared and verified with those of other HSDTs and method of elasticity wherever considered.

• 한국콘텐츠학회논문지
• /
• 제13권4호
• /
• pp.44-52
• /
• 2013

SIFT(Scale Invariant Feature Transform) 알고리즘은 현재 비디오 감시카메라, 자율 주행시스템 등과 같은 영상 시스템에서 많이 사용되고 있다. SIFT 알고리즘에서 연산량과 연산시간이 가장 많이 필요한 부분이 descriptor의 sin/cos 함수를 연산하는 부분이다. 그러므로 본 논문에서는 SIFT 알고리즘에 사용되는 descriptor를 위한 sin/cos 함수를 하드웨어로 구현하였다. Verilog-HDL 언어를 사용하여 FPGA로 구현하고 그 성능을 분석한다. Xilinx Spartan 2E(XC2S200E-PQ208-6) 를 사용하여 구현하였을때, 149 Slices에 233 LUTs가 소모되었으며, 최대 주파수는 60.01MHz로 동작하였다. 또한 descriptor에 적용하여 소프트웨어와 비교 하였을 때 40배 정도의 빠른 성능 향상을 얻었다.

• ETRI Journal
• /
• 제27권4호
• /
• pp.341-354
• /
• 2005

The adaptation of video according to the heterogeneous and dynamic resource constraints on networks and devices, as well as on user preferences, is a promising approach for universal access and consumption of video content. For optimal adaptation that satisfies the constraints while maximizing the utility that results from the adapted video, it is necessary to devise a systematic way of selecting an appropriate adaptation operation among multiple feasible choices. This paper presents a general conceptual framework that allows the formulation of various adaptations as constrained optimization problems by modeling the relations among feasible adaptation operations, constraints, and utilities. In particular, we present the feasibility of the framework by applying it to a use case of rate adaptation of MPEG-4 video with an explicit modeling of adaptation employing a combination of frame dropping and discrete cosine transform coefficient dropping, constraint, utility, and their mapping relations. Furthermore, we provide a description tool that describes the adaptation-constraint-utility relations as a functional form referred to as a utility function, which has been accepted as a part of the terminal and network quality of service tool in MPEG-21 Digital Item Adaptation (DIA).

• Geomechanics and Engineering
• /
• 제11권5호
• /
• pp.671-690
• /
• 2016

This work presents a static flexure analysis of laminated composite plates by utilizing a higher order shear deformation theory in which the stretching effect is incorporated. The axial displacement field utilizes sinusoidal function in terms of thickness coordinate to consider the transverse shear deformation influence. The cosine function in thickness coordinate is employed in transverse displacement to introduce the influence of transverse normal strain. The highlight of the present method is that, in addition to incorporating the thickness stretching effect (${\varepsilon}_z{\neq}0$), the displacement field is constructed with only 5 unknowns, as against 6 or more in other higher order shear and normal deformation theory. Governing equations of the present theory are determined by employing the principle of virtual work. The closed-form solutions of simply supported cross-ply and angle-ply laminated composite plates have been obtained using Navier solution. The numerical results of present method are compared with those of the classical plate theory (CPT), first order shear deformation theory (FSDT), higher order shear deformation theory (HSDT) of Reddy, higher order shear and normal deformation theory (HSNDT) and exact three dimensional elasticity theory wherever applicable. The results predicted by present theory are in good agreement with those of higher order shear deformation theory and the elasticity theory. It can be concluded that the proposed method is accurate and simple in solving the static bending response of laminated composite plates.

• Structural Engineering and Mechanics
• /
• 제51권5호
• /
• pp.867-891
• /
• 2014

An equivalent single layer trigonometric shear deformation theory taking into account transverse shear deformation effect as well as transverse normal strain effect is presented for static flexure of cross-ply laminated composite and sandwich plates. The inplane displacement field uses sinusoidal function in terms of thickness coordinate to include the transverse shear deformation effect. The cosine function in thickness coordinate is used in transverse displacement to include the effect of transverse normal strain. The kinematics of the present theory is much richer than those of the other higher order shear deformation theories, because if the trigonometric term (involving thickness coordinate z) is expanded in power series, the kinematics of higher order theories (which are usually obtained by power series in thickness coordinate z) are implicitly taken into account to good deal of extent. Governing equations and boundary conditions of the theory are obtained using the principle of virtual work. The closed-form solutions of simply supported cross-ply laminated composite and sandwich plates have been obtained. The results of present theory are compared with those of the classical plate theory (CPT), first order shear deformation theory (FSDT), higher order shear deformation theory (HSDT) of Reddy and exact three dimensional elasticity theory wherever applicable. The results predicted by present theory are in good agreement with those of higher order shear deformation theory and the elasticity theory.