• 한국정보전자통신기술학회논문지
• /
• 제7권1호
• /
• pp.55-60
• /
• 2014

본 논문에서는 자연 영상에서 피부색 색상과 채도를 기초로 얼굴영역을 추출하고 얼굴의 특징요소를 추출하는 방법을 제안하였다. 제안한 알고리즘은 조명보정과 얼굴 검출 과정으로 구성되었다. 조명 보정 과정에서는 조명변화에 대한 보정기능을 수행한다. 얼굴 검출 과정은 20개의 피부색 표본 영상에서 색상과 채도를 특징벡터로 사용, 입력영상과의 유클리디안 거리를 구하여 피부색 영역을 추출하였다. 추출된 얼굴 후보영역을 CMY칼라 모델에서 C요소로 눈을 검출하였고, YIQ 칼라 공간에서 Q요소로 입을 검출하였다. 추출된 얼굴 후보영역에서 일반적인 얼굴에 대한 지식을 기반으로 얼굴 영역을 검출하였다. 입력받은 10장의 자연 영상으로 실험한 결과 100%의 얼굴 검출율을 보였다.

• 자원ㆍ환경경제연구
• /
• 제21권2호
• /
• pp.211-235
• /
• 2012

이산화탄소 저감에 대한 사회 경제적 관심이 높아지면서 실제 이산화탄소 배출의 변화요인을 추정하고 그 영향력을 파악하는 분해분석의 중요성이 높아지고 있다. 본 연구는 1980년~2007년의 OECD 27개 국가를 대상으로 방향거리함수를 활용한 이산화탄소 분해산식을 정의하고 각 요인별 변화효과를 추정하였다. 특히 생산 과정에 투입되는 에너지 사용량을 화석연료와 비-화석연료로 구분하고 각 에너지원에 따른 이산화탄소 변화효과를 계측하였다. 상대적으로 급격한 경제성장을 경험한 국가는 표본기간내 생산성 하락과 배출집약도 증가를 동시에 보여주고 있고 규모 확대로 인한 이산화탄소 배출압력을 효과적으로 제어하지 못하고 있었다. 특히 효율 및 기술 변화요인이 이산화탄소의 저감을 유도하지 못하였고 이산화탄소의 배출량 증가에 가장 주요한 요인으로 작용하고 있었다. 또한 적절한 환경규제가 작용하지 않는 상태에서 이루어진 경제성장이 배출집약도의 상승을 유도하고 있고 강력한 환경정책 설정의 필요성을 확인할 수 있었다. 반면 실제 이산화탄소의 감소를 보이는 국가는 생산성의 성장, 에너지 투입의 효율성 상승, 배출집약도의 하락을 동시에 보여주고 있었다. 비-화석연료의 사용비중 증대와 강력한 환경정책 등으로 인한 환경친화적 생산전환을 확인할 수 있었고 대부분이 환경에 대한 사회 경제적 관심이 높은 유럽국가라는 공통점이 있었다. 특히 이들 국가는 기술변화에 의한 저감효과가 강하게 나타났고, 이산화탄소의 배출량 증가속도가 빠른 국가에 비해 그 효과가 확대되는 모습을 보였다.

• Journal of Information Processing Systems
• /
• 제14권6호
• /
• pp.1464-1479
• /
• 2018

In face recognition, sometimes the number of available training samples for single category is insufficient. Therefore, the performances of models trained by convolutional neural network are not ideal. The small sample face recognition algorithm based on novel Siamese network is proposed in this paper, which doesn't need rich samples for training. The algorithm designs and realizes a new Siamese network model, SiameseFacel, which uses pairs of face images as inputs and maps them to target space so that the $L_2$ norm distance in target space can represent the semantic distance in input space. The mapping is represented by the neural network in supervised learning. Moreover, a more lightweight Siamese network model, SiameseFace2, is designed to reduce the network parameters without losing accuracy. We also present a new method to generate training data and expand the number of training samples for single category in AR and labeled faces in the wild (LFW) datasets, which improves the recognition accuracy of the models. Four loss functions are adopted to carry out experiments on AR and LFW datasets. The results show that the contrastive loss function combined with new Siamese network model in this paper can effectively improve the accuracy of face recognition.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 춘계종합학술대회 A
• /
• pp.302-305
• /
• 2008

For an complete image composition to be stitched on several mosaic images, tracing displacement of direction and distance between successive images are important parameters. The input image is modeled by using a general second order two-dimensional Taylor-series and then converting it to a $3{\times}3$ correlation block and storing the data. A moving factor and coordinate is calculated by comparing the continuous correlation blocks. The experimentation result has a success rate of 85% for moving path tracing as continuous images are moved to 10% of image central position.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.495-500
• /
• 1998

This paper introduces our three approaches to reduce the burden of human interactive EC operators: (1) improvement of the interface of presenting individuals, (2)improvement of the interface of inputting fitness values, and (3) fast EC convergence. We propose methods to display individuals in order of predicted fitness values by neural networks or Euclidean distance measure for (1), to input quantized fitness values for (2), and to make a new elite by approximating the EC search space with a quadratic function for (3). They are evaluated through simulations and subjective testes, and their effects have shown.

• 한국지능시스템학회논문지
• /
• 제8권4호
• /
• pp.1-7
• /
• 1998

This paper describes a meshless of element-free method based on fuzzy knowledge processing. To efficiently simulate complicated physical phenomena with dynmics and non-linear ploblem using computational mechanics, special method is required such as parallel processing or adaptive analysis techniques. However, the conventional finite element method is too complicated to be employed in the above cases. In order to reduce the above complexity of the conventional finite element analysis systms, the so called meshles finite elements as an input information have been stuided. Node is generated if its distance form existing node points is similar to the node spacing fuction at the point. The node spacing function is well controlled by the fuzzy knowledge processing Practical performances of the present system are demonstrated through several three-dimensional(3D) problems.

• 대한전기학회논문지:전력기술부문A
• /
• 제48권9호
• /
• pp.1055-1063
• /
• 1999

This paper present a conceptual possibilistic approach using fuzzy set theory to manage the uncertainties in the given reliability input date of the practical power system. In this paper, an algorithm is introduced to calculate the possibilstic reliability indices according to the degree of uncertainty in the given data. The probability distribution function can be transformed into an appropriate possibilstic representation using the probability-Possibility Consistency principle(PPCP) algorithm. In this the algorithm, the transformation is performation by making a compromise between the transformation consistency and the human updating experience. Fuzzy classifcation theory is applied to reduced the number of load data. The fuzzy classification method determines the closeness of load data points by assigning them to various clusters and then determening the distance between the clusters. The IEEE-RTS with 32-generating units is used to demonstrate the capability of the proposed algorithm.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2002년도 춘계학술대회 논문집 디스플레이 광소자 분야
• /
• pp.97-100
• /
• 2002

We analyzed the excitation coefficients and the extinction ratio as a function of waveguide width, center to center distance, and angle of input waveguide for a directional coupler. The variation of thc difference between excitation coefficients is similar to that of the extinction ratio. Also, the smaller the difference between those is, the better the extinction ratio is. This concept will have application to devices using optical coupling.

• 반도체디스플레이기술학회지
• /
• 제3권2호
• /
• pp.1-4
• /
• 2004

An experimental study was conducted on the optimal configuration and size of ADFL(Aerodynamic Focusing Lens) which used in modified ISPM(In-Situ Particle Monitoring). The particle counting efficiency has been known as a function of distance and size of ADFL, thus we varied these parameters to find out the optimum values. From a result of experiment, it was found that two lenses and 6mm space between them showed a maximum particle measuring efficiency. To apply this modified ISPM to semiconductor manufacturing field, we need more experiment about the pressure change, flow rate, and input particle size.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(5)
• /
• pp.131-134
• /
• 2001

This paper proposes a stable control rule for nonlinear unicycle-type mobile robot. The control method uses a local error coordinate system, velocity and distance constants $\kappa$$\_$x/, $\kappa$$\_$y/, and he. Stability of control rule is proved Liapunov function. System input to the mobile robot is reference posture ($\chi$$\_$r/, y$\_$r/, $\theta$$\_$r/)/sup/ $\tau$/ and reference e velocity (ν$\_$r/,$\omega$$\_$r/)$\^$$\tau$/. System output of the mobi-le robot is velocity of driving wheels. We introduce limit velocity for preventing high initial speed. From simulation results, we can see the proposed control rule is stable.