• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.907-923
• /
• 2020

For the severe haze situation in the Beijing-Tianjin-Hebei region, conventional fine particulate matter (PM2.5) concentration prediction methods based on pollutant data face problems such as incomplete data, which may lead to poor prediction performance. Therefore, this paper proposes a method of predicting the PM2.5 concentration based on image analysis technology that combines image data, which can reflect the original weather conditions, with currently popular machine learning methods. First, based on local parameter estimation, autoregressive (AR) model analysis and local estimation of the increase in image blur, we extract features from the weather images using an approach inspired by free energy and a no-reference robust metric model. Next, we compare the coefficient energy and contrast difference of each pixel in the AR model and then use the percentages to calculate the image sharpness to derive the overall mass fraction. Furthermore, the results are compared. The relationship between residual value and PM2.5 concentration is fitted by generalized Gauss distribution (GGD) model. Finally, nonlinear mapping is performed via the wavelet neural network (WNN) method to obtain the PM2.5 concentration. Experimental results obtained on real data show that the proposed method offers an improved prediction accuracy and lower root mean square error (RMSE).

• The Journal of the Korea institute of electronic communication sciences
• /
• v.10 no.4
• /
• pp.521-526
• /
• 2015

Research on how to use the biological characteristics of human to confirm the identity of the individual is being actively conducted. Electrocardiogram(: ECG) based biometric system is difficult to counterfeit and does not cause skin irritation on the subject. It can be easily combined with conventional biometrics such as fingerprint and face recognition to give multimodal biometric systems. In this thesis, biometric identification method analysing ECG waveform characteristics from Discrete Wavelet Transform(DWT) coefficients is suggested. Feature selection is performed on the 9 coefficients of DWT using the correlation analysis. The verification is achieved by using the error back propagation neural networks. Using the proposed approach on 24 subjects of MIT-BIH QT Database, 98.88% verification rate has been obtained.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.11
• /
• pp.2409-2414
• /
• 2002

The surfaces of prototypes become rough due to the stair-stepping which is the inevitable phenomenon in the Rapid Prototypes are not used only for the verification of feature. The grinding, coating, or the composition of them is a main operation in post-processing in which lots of costs and long build time are needed. The solution is proposed to increase the efficiency of rapid prototyping by minimizing or removing the composition of them is a main operation in post-processing in which lots of costs and long build time are needed. the solution is proposed to increase the efficiency of rapid prototyping by minimizing or removing the regions for post-processing. the factors to cause the surface roughness and their effects are analyzed through the experiments. Software modules are developed to predict the surface roughness of each face in the prototyping with the result. An experimental compensation method is developed to apply the modules to various RP equipments, materials and build styles. The build direction is searched with use of genetic algorithm to maximize the total areas of the surface of which roughness is better than the user-defined value.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.881-884
• /
• 2017

좋은 특징을 도출할 수 있는 신경망은 곧 대상을 잘 이해하고 있는 신경망을 의미한다. 그러나 얼굴과 같이 유사한 이미지를 분류하기 위해서는 신경망이 좀 더 구분되는 특징을 도출해야한다. 본 논문에서는 얼굴과 같이 유사도한 이미지를 분류하기 위해 오차함수에 중심확장(Center Expansion)이라는 오차를 추가한다. 중심확장은 도출된 특징이 밀집되면 클래스를 분류하는 매니폴드를 구하기 어려워져 분류 성능이 하락되는 문제를 해결하기 위해 제안한 것으로 특징이 밀집될 가능성이 높은 부분에 특징이 도출되지 않도록 강제하는 방식이다. 학습 시 활용하는 오차는 일반적으로 분류 문제를 위해 사용되는 softmax cross-entropy 오차와 각 클래스의 분산을 줄이는 오차 그리고 제안한 중심확장 오차를 조합해 구할 것이다. 본 논문에서는 제안한 중심확장 오차를 조합한 모델과 조합되지 않은 모델이 결과적으로 특징 도출과 분류에 어떠한 영향을 주었는지 알아볼 것이다. 중심확장을 조합해 학습한 모델이 어떤 영향을 주었는지 알기 위해 본 논문에서는 Labeled Faces in the Wild를 활용해 분류 실험을 진행할 것이다. Labeled Faces in the Wild을 활용해 실험한 결과 중심확장을 활용한 모델과 활용하지 않은 모델간의 성능을 차이를 확인할 수 있었다.

• Journal of KIISE:Software and Applications
• /
• v.31 no.6
• /
• pp.799-807
• /
• 2004

In this paper, we deal with an emotion recognition method using facial images and speech signal. Six basic human emotions including happiness, sadness, anger, surprise, fear and dislike are investigated. Emotion recognition using the facial expression is performed by using a multi-resolution analysis based on the discrete wavelet transform. And then, the feature vectors are extracted from the linear discriminant analysis method. On the other hand, the emotion recognition from speech signal method has a structure of performing the recognition algorithm independently for each wavelet subband and then the final recognition is obtained from a multi-decision making scheme.

• Proceedings of the Korea Contents Association Conference
• /
• 2009.05a
• /
• pp.1020-1025
• /
• 2009

The five elements theory that people in the world objective theory is earned by a long-term observation. The key to the world and everything in the neck. The core content of this theroy is summed up in the world all things wood, fire, earth, gold, water-five material form. In the theory, the chracter's appearance, feature the five elements face is content. And the chracter's appearance, personality chracterisics and features of chracters and abstract nature of the mutual response plan which was formed way. Based on the number of world-famous cartoon chracter design analysis, special features and art theroy point the five elements the nature of man, who looks and chracter design by comparing the response of theroy and research and has the five elements. The analysis of theory and research the five elements art and people who find the nature of the systematic correlation to the future of animation, chracter design is applied.

• Korean Journal of Logic
• /
• v.20 no.3
• /
• pp.335-366
• /
• 2017

Sorensen proved that 'vague' is vague through a sorites paradox that he presented. I will show that his sorites paradox satisfies the conditions for a sorites paradox and does not satisfy those conditions. So we will face a new paradox which exemplifies inconsistency of conditions for a sorites paradox. But those conditions include a principle which says when two cases are close enough, there will be no change in truth value. It represents an essential feature of vagueness. Therefore the paradox I will present in this article is not only shows inconsistency of conditions for a sorites paradox but also shows inconsistency of vagueness itself.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.5
• /
• pp.393-398
• /
• 2007

In this paper, we propose the development of Artificial Hippocampus Algorithm(AHA) which remodels a principle of brain of hippocampus. Hippocampus takes charge auto-associative memory and controlling functions of long-term or short-term memory strengthening. We organize auto-associative memory based 4 steps system (EC, DG CA3, and CA1) and improve speed of teaming by addition of modulator to long-term memory teaming. In hippocampus system, according to the 3 steps order, information applies statistical deviation on Dentate Gyrus region and is labeled to responsive pattern by adjustment of a good impression. In CA3 region, pattern is reorganized by auto-associative memory. In CA1 region, convergence of connection weight which is used long-term memory is learned fast a by neural network which is applied modulator. To measure performance of Artificial Hippocampus Algorithm, PCA(Principal Component Analysis) and LDA(Linear Discriminants Analysis) are applied to face images which are classified by pose, expression and picture quality. Next, we calculate feature vectors and learn by AHA. Finally, we confirm cognitive rate. The results of experiments, we can compare a proposed method of other methods, and we can confirm that the proposed method is superior to the existing method.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.1012-1018
• /
• 2007

본 논문에서는 비전 기반 3차원 얼굴 모델의 자동 표정 생성 시스템을 제안한다. 기존의 3차원 얼굴 애니메이션에 관한 연구는 얼굴의 움직임을 나타내는 모션 추정을 배제한 얼굴 표정 생성에 초점을 맞추고 있으며 얼굴 모션 추정과 표정 제어에 관한 연구는 독립적으로 이루어지고 있다. 제안하는 얼굴 모델의 표정 생성 시스템은 크게 얼굴 검출, 얼굴 모션 추정, 표정 제어로 구성되어 있다. 얼굴 검출 방법으로는 얼굴 후보 영역 검출과 얼굴 영역 검출 과정으로 구성된다. HT 컬러 모델을 이용하며 얼굴의 후보 영역을 검출하며 얼굴 후보 영역으로부터 PCA 변환과 템플릿 매칭을 통해 얼굴 영역을 검출하게 된다. 검출된 얼굴 영역으로부터 얼굴 모션 추정과 얼굴 표정 제어를 수행한다. 3차원 실린더 모델의 투영과 LK 알고리즘을 이용하여 얼굴의 모션을 추정하며 추정된 결과를 3차원 얼굴 모델에 적용한다. 또한 영상 보정을 통해 강인한 모션 추정을 할 수 있다. 얼굴 모델의 표정을 생성하기 위해 특징점 기반의 얼굴 모델 표정 생성 방법을 적용하며 12개의 얼굴 특징점으로부터 얼굴 모델의 표정을 생성한다. 얼굴의 구조적 정보와 템플릿 매칭을 이용하여 눈썹, 눈, 입 주위의 얼굴 특징점을 검출하며 LK 알고리즘을 이용하여 특징점을 추적(Tracking)한다. 추적된 특징점의 위치는 얼굴의 모션 정보와 표정 정보의 조합으로 이루어져있기 때문에 기하학적 변환을 이용하여 얼굴의 방향이 정면이었을 경우의 특징점의 변위인 애니메이션 매개변수를 획득한다. 애니메이션 매개변수로부터 얼굴 모델의 제어점을 이동시키며 주위의 정점들은 RBF 보간법을 통해 변형한다. 변형된 얼굴 모델로부터 얼굴 표정을 생성하며 모션 추정 결과를 모델에 적용함으로써 얼굴 모션 정보가 결합된 3차원 얼굴 모델의 표정을 생성한다.

• Structural Engineering and Mechanics
• /
• v.57 no.6
• /
• pp.1051-1064
• /
• 2016

Link-to-column connections in Eccentrically Braced Frames (EBFs) have critical role in their safety and seismic performance. Accordingly, in this study, contribution of supplemental stiffeners on hysteretic behavior of the link-to-column connection is investigated. Considered stiffeners are placed on both sides and parallel to the link web between the column face and the first stiffener of the link. Hysteretic behaviors of the link beams with supplemental stiffeners are numerically investigated using a pre-validated numerical model in ANSYS. It turned out that supplemental stiffeners can change energy dissipation mechanism of intermediate links from shear-flexure to shear. Both rectangular and trapezoidal supplemental stiffeners are studied. Moreover, optimal placement of the supplemental stiffeners is also investigated. Obtained results indicate a discrepancy of less than 9% in maximum link shear of the numerical and experimental specimens. This indicates that the numerical results are in good agreement with those obtained from the test. Trapezoidal supplemental stiffeners improve rotational capacity of the link. Moreover, use of two supplemental stiffeners at both ends of the link can more effectively improve hysteretic behavior of intermediate links. Supplemental stiffeners would also alleviate the imposed demands on the connections. This latter feature is more pronounced in the case of two supplemental stiffeners at both ends of the link.