• 한국소음진동공학회논문집
• /
• 제21권2호
• /
• pp.146-153
• /
• 2011

Model based machine condition diagnosis methods are generally using a normal and many failure models which need sufficient data to train the models. However, data, especially for failure modes of interest, is very hard to get in real applications. So their industrial applications are either severely limited or impossible when the failure models cannot be trained. In this paper, continuous hidden Markov model(CHMM) with only a normal model has been suggested as a very promising machine condition diagnosis method which can be easily used for industrial applications. Generally hidden Markov model also uses many pattern models to recognize specific patterns and the recognition results of CHMM show the likelihood trend of models. By observing this likelihood trend of a normal model, it is possible to detect failures. This method has been successively applied to arc weld defect diagnosis. The result shows CHMM's big potential as a machine condition monitoring method.

• 한국환경과학회지
• /
• 제13권7호
• /
• pp.611-618
• /
• 2004

The Neural Network Models which mathematically interpret human thought processes were applied to resolve the uncertainty of model parameters and to increase the model's output for the streamflow forecast model. In order to test and verify the flood discharge forecast model eight flood events observed at Kumho station located on the midstream of Kumho river were chosen. Six events of them were used as test data and two events for verification. In order to make an analysis the Levengerg-Marquart method was used to estimate the best parameter for the Neural Network model. The structure of the model was composed of five types of models by varying the number of hidden layers and the number of nodes of hidden layers. Moreover, a logarithmic-sigmoid varying function was used in first and second hidden layers, and a linear function was used for the output. As a result of applying Neural Networks models for the five models, the N10-6model was considered suitable when there is one hidden layer, and the Nl0-9-5model when there are two hidden layers. In addition, when all the Neural Network models were reviewed, the Nl0-9-5model, which has two hidden layers, gave the most preferable results in an actual hydro-event.

• 한국지능시스템학회논문지
• /
• 제25권1호
• /
• pp.97-104
• /
• 2015

An 'associative cube', a class of auto-associative memories, is revisited here, in which training data and hidden orthogonal basis functions such as wavelet packets or Fourier kernels, are combined in the weight cube. This weight cube has hidden units in its depth, represented by a three dimensional cubic structure. We develop an unsupervised incremental learning mechanism based upon the adaptive least squares method. Training data are mapped into orthogonal basis vectors in a least-squares sense by updating the weights which minimize an energy function. Therefore, a prescribed orthogonal kernel is incrementally assigned to an incoming data. Next, we show how a decoding procedure finds the closest one with a competitive network in the hidden layer. As noisy test data are applied to an associative cube, the nearest one among the original training data are restored in an optimal sense. The simulation results confirm robustness of associative cubes even if test data are heavily distorted by various types of noise.

• 한국정보전자통신기술학회논문지
• /
• 제17권1호
• /
• pp.17-24
• /
• 2024

본 논문에서는 4중 암호화 기법을 사용하여 영상 픽셀에 기밀 데이터를 은닉하는 보안이 강력한 기법을 제안한다. 제안된 기법에서는 영상의 윤곽선이 존재하는 경계면과 픽셀값의 변화가 거의 없는 평탄면을 조사한다. 영상의 경계면에서는 경계면의 특성을 보존하기 위해 경계면에 위치하는 픽셀의 LSB(Least Significant Bit)에 다중으로 암호화된 기밀 데이터 1비트를 또다시 공간적으로 암호화하여 기밀 데이터를 은닉한다. 영상의 경계면이 아니고 픽셀값의 변화가 적은 평탄면에 존재하는 픽셀들에서는 다중으로 암호화된 기밀 데이터 2비트를 위치기반 암호화 기법과 공간적 암호화 기법을 사용하여 픽셀의 하위 2비트에 은닉한다. 제안 기법을 적용하여 기밀 데이터를 은닉하는 경우 스테고 이미지의 화질이 최대 49.64dB이고, 기존 LSB 방식에 비해 은닉되는 기밀 데이터의 양이 최대 92.2% 증가하고, 암호화키가 없으면 스테고 이미지에 은닉된 암호화된 기밀 데이터를 추출할 수 없으며 추출한다 해도 해독할 수 없어 스테고 이미지에 은닉된 기밀 데이터의 보안은 매우 강력하게 유지된다. 제안된 기법은 가역 데이터 은닉 기법이 사용되지 않아도 되는 웹툰과 같은 일반적인 상업적 이미지에 저작권 정보를 숨기는 데 효과적으로 사용될 수 있다.

• Journal of Communications and Networks
• /
• 제7권3호
• /
• pp.313-324
• /
• 2005

The success of the IEEE 802.11 standard has prompted research into efficiency of the different medium access methods and their support for different traffic types. A modified version of the point coordination function (PCF) called modified PCF has been introduced as a way to improve the efficiency over the standard method. It has been shown through a simulation study and a mathematical analysis that channel utilization can be much improved compared to the standard, in case there is no so-called hidden station problem. However, under the hidden station problem, the efficiency of the modified PCF would obviously decrease. In this paper, some enhancements of the modified PCF are introduced. Firstly, we propose a retransmission process to allow frames involved in collisions to be retransmitted. Then, we propose a collision resolution mechanism to reduce the frame collision probability due to the hidden station problem. In addition, we propose a priority scheme to support prioritization for different traffic types such as interactive voice and video, and real-time data traffic in the modified PCF. To prevent the starvation of one low priority traffic, minimum transmission period is also guaranteed to each traffic type via an admission control algorithm. We study the performance of the modified PCF under the hidden station problem and the performance of the modified PCF with priority scheme through simulations. To illustrate the efficiency of the priority scheme, we therefore compare its simulation results with those of some standardized protocols: The distributed coordination function (DCF), the enhanced distributed channel access (EDCA), the PCF, and our previously proposed protocol: The modified PCF without priority scheme. The simulation results show that the increment of delay in the network due to the hidden station problem can be reduced using the proposed collision resolution mechanism. In addition, in a given scenario the modified PCF with priority scheme can provide better quality of service (QoS) support to different traffic types and also support a higher number of data stations than the previous proposals.

• 융합신호처리학회논문지
• /
• 제20권4호
• /
• pp.205-211
• /
• 2019

스테그아날리시스(steganalysis)는 스테가노그래피(steganography)에 의해 숨겨진 데이터를 감지하고 복구하기 위한 기법이다. 스테그아날리시스 방법은 데이터 삽입 시 발생하는 시각적, 통계적 변화를 분석하여 숨겨진 데이터를 찾는다. 숨겨진 데이터를 복원하기 위해서는 어떤 스테가노그래피 방법에 의해 데이터가 숨겨졌는지를 알아야 한다. 그러므로 본 논문은 다층 분류를 통해 입력 영상에 적용된 스테가노그래피 방법을 식별하는 계층적 CNN 구조를 제안한다. 이를 위해 4개의 기본 CNN을 각각 입력 영상에 스테가노그래피 방법이 적용되었는지 여부나 서로 다른 두 스테가노그래피 방법 중에 어떤 방법이 적용되었는지를 이진 판별하도록 학습시켰으며, 학습된 CNN을 계층적으로 연결하였다. 실험 결과를 통해 제안된 계층적 CNN 구조는 4개의 서로 다른 스테가노그래피 방법인 LSB(Least Significant Bit Substitution), PVD(Pixel Value Difference), WOW(Wavelet Obtained Weights), UNIWARD(Universal Wavelet Relative Distortion)을 79%의 정확도로 식별할 수 있음을 확인하였다.

• 한국콘텐츠학회논문지
• /
• 제8권5호
• /
• pp.44-51
• /
• 2008

시계열 데이터에서 패턴을 찾고 검색하는 문제는 여러 분야에서 오랫동안 관심을 가지고 연구되어 왔다. 본 논문은 시간의 흐름에 따라 값의 변화를 나타내는 시계열 형태의 주식 데이터에 적용할 수 있는 새로운 패턴 매칭 방법을 제안한다. 우선, 의미를 기반으로 패턴을 정의하고 정의된 패턴에 일치하는 데이터들을 추출하여 학습모델을 작성한다. 그리고 새로운 질의 시퀀스가 어떤 종류의 패턴과 일치하는가는 각 학습 모델과의 유사도를 측정하여 결정하게 된다. 학습 모델은 시계열을 잘 설명하는 것으로 알려진 은닉 마코프 모델을 사용하여 작성하였다. 실험 결과 은닉 마코프 모델의 특성을 사용하여 생성된 각 학습 모델은 주어진 의미를 잘 나타내는 패턴을 생성하였으며, 새로운 시퀀스가 주어졌을 때 일치하는 패턴에 따라서 시퀀스가 가진 의미를 파악할 수 있었다.

• 한국지능시스템학회논문지
• /
• 제18권2호
• /
• pp.243-250
• /
• 2008

본 논문에서는 은닉 마코프 모델을 이용하여 논문 모집 공고에서 정보를 추출하는 시스템을 제안한다. 논문 모집 공고는 완전히 정형화된 형식을 가지지는 않지만, 내용의 출현 순서에 따른 흐름이 어느 정도 존재한다. 따라서 순차적인 데이터를 해석하는데 강점을 지닌 은닉 마코프 모델을 논문 모집 공고를 분석하는데 사용한다. 하지만, 논문 모집 공고를 은닉 마코프 모델로 직관적으로 모델링하면 정보 경계가 정확히 인식되지 않는 문제가 발생한다. 본 논문에서는 이 문제를 해결하기 위해 2-단계의 은닉 마코프 모델을 사용한다. 즉, 첫 번째 단계에서, 문서를 구로 모델링한 P-HMM(Phrase hidden Markov model)이 지역적으로 문서를 인식한다. 그리고 두 번째 단계에서 D-HMM(Document hidden Markov model)은 문서가 가진 전체적인 구조와 정보의 흐름을 파악한다. 웹에서 수집된 400개의 논문 모집 공고에 대한 실험 결과, F-measure 성능이 0.49를 보인다. 이는 직관적인 은닉 마코프 모델보다 F-measure로 0.15 정도 향상된 결과이다.

• 한국정밀공학회지
• /
• 제28권7호
• /
• pp.814-820
• /
• 2011

It takes long time in numerical simulation because structural design for tire requires the nonlinear material property. Neural networks has been widely studied to engineering design to reduce numerical computation time. The numbers of hidden layer, hidden layer neuron and training data have been considered as the structural design variables of neural networks. In application of neural networks to optimize design, there are a few studies about arrangement method of input layer neurons. To investigate the effect of input layer neuron arrangement on neural networks, the variables of tire contour design and tension in bead area were assigned to inputs and output for neural networks respectively. Design variables arrangement in input layer were determined by main effect analysis. The number of hidden layer, the number of hidden layer neuron and the number of training data and so on have been considered as the structural design variables of neural networks. In application to optimization design problem of neural networks, there are few studies about arrangement method of input layer neurons. To investigate the effect of arrangement of input neurons on neural network learning tire contour design parameters and tension in bead area were assigned to neural input and output respectively. Design variables arrangement in input layer was determined by main effect analysis.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2018년도 학술발표회
• /
• pp.499-499
• /
• 2018

Hydro-meteorological extremes are trivial in these days. Therefore, it is important to identify extreme hydrological events in advance to mitigate the damage due to the extreme events. In this context, exploring temporal distribution of sub-daily extreme rainfall at multiple rain gauges would informative to identify different states to describe severity of the disaster. This study proposehidden Markov chain model (HMM) based rainfall analysis tool to understand the temporal sub-daily rainfall patterns over South Korea. Hourly and daily rainfall data between 1961 and 2017 for 92 stations were used for the study. HMM was applied to daily rainfall series to identify an observed hidden state associated with rainfall frequency and intensity, and further utilized the estimated hidden states to derive a temporal distribution of daily extreme rainfall. Transition between states over time was clearly identified, because HMM obviously identifies the temporal dependence in the daily rainfall states. The proposed HMM was very useful tool to derive the temporal attributes of the daily rainfall in South Korea. Further, daily rainfall series were disaggregated into sub-daily rainfall sequences based on the temporal distribution of hourly rainfall data.