• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.8-8
• /
• 2018

전통적인 빈도해석은 정상성 가정을 기초로 단일 확률분포를 강우 및 홍수량 자료에 적용하는 과정을 통해 확률수문량을 추정하는 것을 목적으로 하고 있다. 그러나 전지구적인 기상학적 변동성 및 기후변화로 기인하는 극치수문량의 발생 빈도 및 양적 크기의 변화는 확률통계학적 관점에서 서로 다른 분포특성을 가지게 된다. 대표적인 기상변동성인 엘니뇨가 발생하는 경우 지역에 따라 홍수 및 가뭄이 발생 발생하게 되며, 이러한 극치수문량은 일반적으로 나타나는 홍수 및 가뭄의 분포특성과는 상이한 경우가 많다. 즉, 2개 이상의 확률분포 특성이 혼재된 혼합분포의 특성을 가지는 경우가 나타내게 되며 이를 고려한 빈도해석 기법의 개발 및 적용이 필요하다. 혼합분포를 활용한 빈도해석에서 가장 중요한 사항 중에 하나는 개별 분포에 적용되는 가중치를 추정하는 것으로서 통계학적 관점에서 자료의 특성에 근거하여 내재되어 있는 은닉상태(latent process)를 추정하는 과정과 유사하다. 이와 더불어 앞서 언급된 기상학적 변동성을 빈도해석에 반영하기 위한 비정상성 해석기법의 개발 및 적용도 필요하다. 본 연구에서는 혼합분포를 활용한 비정상성빈도해석모형을 개발하는데 목적이 있으며 개별매개변수의 동적거동 뿐만 아니라 가중치에 대한 시간적인 종속성도 고려할 수 있는 모형으로 동적모형으로 다양한 실험적 해석이 가능하다. 본 연구에서는 개발된 모형을 기반으로 엘니뇨와 같은 기상변동성에 따른 강우 및 홍수빈도해석 측면에서 은닉상태에 변화, 이로 인한 확률분포의 특성 및 설계수문량의 동적변동성을 평가하고자 한다.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.130-138
• /
• 2021

This paper presents an effective structure by applying various normalization to Convolutional Neural Networks (CNN) for seismic event classification. Normalization techniques can not only improve the learning speed of neural networks, but also show robustness to noise. In this paper, we analyze the effect of input data normalization and hidden layer normalization on the deep learning model for seismic event classification. In addition an effective model is derived through various experiments according to the structure of the applied hidden layer. As a result of various experiments, the model that applied input data normalization and weight normalization to the first hidden layer showed the most stable performance improvement.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.11 no.3
• /
• pp.33-44
• /
• 2001

E-commerce has various advantages such as saving the cost and no constraint on time and space, unlike real-world commerce. On the other hand, e-commerce has some important issues to solve since the commerce is conducted on the cyberspace. The issues are a mutual confidence of parties participating in the commerce as well as a method of payments. On early days, electronic cash was designed using blind digital signature to protect the personal information from being exposed and to provide the perfect anonymity for user. In this paper, a new blind signature scheme based on Feige-Fiat-Shamir digital signature is proposed, which is very efficient compared with the other schemes in terms of the computational complexity. Also, a traceable Electronic Cash System which is based on the proposed blind digital signature is designed, which has a nice feature of identifying the spender in case of using the money illegally.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.2
• /
• pp.313-319
• /
• 1999

We devise the 3 models of Multilayered Recurrent Prediction Neural Network(MLRPNN), which are obtained by modifying the Multilayered Perceptron(MLP) with 4 layers. We experimentally study the speech recognition performance of 3 models by a comparative method, according to the variation of the prediction order, the number of neurons in two hidden layers, initial values of connecting weights and transfer function, respectively. By the experiment, the recognition performance of each MLRPNN is better than that of MLP. At the model that returns the output of the upper hidden layer to the lower hidden layer, the recognition performance shows the best value. All MLRPNNs, which have 10 or 15 neurons in the upper and lower hidden layer and is predicted by 3rd or 4th order, show the improved speech recognition rate. On learning, these MLRPNNs have a better recognition rate when we set the initial weights between -0.5 and 0.5, and use the unipolar sigmoid transfer function in the lower hidden layer.

• Journal of Intelligence and Information Systems
• /
• v.10 no.2
• /
• pp.53-63
• /
• 2004

Recurrent-Cascade-Correlation(RCC) is a supervised teaming algorithm that automatically determines the size and topology of the network. RCC adds new hidden neurons one by one and creates a multi-layer structure in which each hidden layer has only one neuron. By second order RCC, new hidden neurons are added to only one hidden layer. These created neurons are not connected to each other. We present a generalization of the RCC Architecture by combining the standard RCC Architecture and the second order RCC Architecture. Whenever a hidden neuron has to be added, the new RCC teaming algorithm automatically determines whether the network topology grows vertically or horizontally. This new algorithm using sigmoid, tanh and new activation functions was tested with the morse-benchmark-problem. Therefore we recognized that the number of hidden neurons was decreased by the experiments of the RCC network generalization which used the activation functions.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.5C
• /
• pp.683-691
• /
• 2004

To transmit MPEG-2 video on an errorneous channel, a number of error control techniques are needed. Especially, error concealment techniques which can be implemented on receivers independent of transmitters are essential to obtain good video quality. In this paper, prediction of motion vector and an adaptive boundary matching algorithm are presented for temporal error concealment. Before the complex BMA, we perform error concealment by a motion vector prediction using neighboring motion vectors. If the candidate of error concealment is not satisfied, search range and reliable boundary pixels are selected by the temporal activity or motion vectors and a damaged macroblock is concealed by applying an adaptive BMA. This error concealment technique reduces the complexity and maintains a PSNR gain of 0.3∼0.7㏈ compared to conventional BMA.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.1
• /
• pp.56-63
• /
• 2009

Nowdays, Most of Internet Contents delivered as a compressed file. It gives many advantages like deduction of communication bandwidth and transmission time etc. In case of JPEG Compression, Quantization is the most important procedure which accomplish the compression. In general signal processing, Quantization is the process which converts continuous analog signal to discrete digital signal. As you known already, Quantization over JPEG compression is to reduce magnitude of pixel value in spatial domain or coefficient in frequency domain. A lot of Data Hiding algorithms also developed to applicable for those compressed files. In this paper, we are going to unveil the influence of quantization table which used in the process of JPEG compression. Even thought most of algorithm modify frequency coefficients with considering image quality, they are ignoring the influence of quantization factor corresponding with the modified frequency coefficient. If existing algorithm adapt this result, they can easily evaluate their performances.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.232-237
• /
• 2013

The role of the window length of time-frequency transformation is important for the audio data hiding methods utilizing phase information. In this paper, the experiments for our audio data hiding method were conducted in order to evaluate the audio quality and robustness against reverberant environment. The experimental results showed the tendency that the worse audio quality but better robustness were obtained when the lengthy window was applied. The important reason for quality degradation was pre-echo which flatters the percussive sound. The results also indicated that the wireless communication theory related to the length of time-frequency transform can be applied in the field of audio data hiding and acoustic data transmission.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.1
• /
• pp.51-57
• /
• 2010

In this paper, It propose methods compressing BTC image utilizing data hiding technique. BTC is used to compress general digital image into binary image and applied into application such as printer. Additional information, transferred with binary image, is as big as the size of binary image. Therefore, we wish to reduce the total transmission bandwidth by decreasing the additional information with sustaining the small image degradation. Because typical BTC image doesn't have enough space for data hiding, we adopt Adaptive AMBTC (Absolute Moment BTC) algorithm to produce the binary image, and calculate virtual histogram from created binary image and modify this histogram for reducing the additional information. The proposed algorithm can reduce about 6-11 % of the image file size, compared with the existing BTC algorithm, without making perceptible image degradation.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.2
• /
• pp.32-40
• /
• 2011

Reversible data hiding, which can recover the original image without any distortion after the extraction of the hidden data, has drawn considerable attention in recent years. However, underflow and overflow problems have occurred occasionally in the embedded image. To overcome these problems, we propose a new reversible data hiding algorithm which embeds a compressed location map used to identify these underflow and overflow points. In addition, the proposed algorithm allows for multilevel data hiding to increase the hiding capacity. The simulation results demonstrate that the proposed algorithm generates good performances in the PSNR, the embedding capacity, and the size of side information.