• Smart Media Journal
• /
• v.12 no.4
• /
• pp.9-18
• /
• 2023

Steganography is a technology capable of protecting data by hiding the existence of data. Recently, with the development of deep learning technology, deep learning-based steganography are being developed. Deep learning can learn by analyzing high-dimensional features of data, so it can improve the performance and quality of steganography. In this paper, we investigated the research trend of image steganography based on deep learning.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.18-20
• /
• 2020

본 연구에서는 고차원 데이터에 대한 차원축소 및 군집 분석과 같은 비지도 학습 알고리즘에 대해 알아보기 위해서 얼굴 이미지 데이터 셋을 사용한다. 얼굴 데이터 셋에 대하여 주요 비지도 학습 알고리즘을 이용하여 실시간으로 클러스터링하고, 그 성능을 비교한다. 비디오에서 추출된 영상 속의 7명의 인물에 대하여 Scikit-learning 라이브러리에서 제공하는 클러스터링 알고리즘과 더불어 주요 차원축소 알고리즘(Dimension Reduction Algorithm)을 사용하여 총 10개의 알고리즘에 대하여 분석한다. 또한, 클러스터링 성능 검사를 통해 알고리즘의 성능을 비교해보고, 이를 통하여 앞으로의 연구 방향에 대해 고찰한다.

• The KIPS Transactions:PartD
• /
• v.13D no.4 s.107
• /
• pp.455-462
• /
• 2006

Recently, the filtering approach using vector approximation such as VA-file[1] or LPC-file[2] have been proposed to support similarity search in high dimensional data space. This approach filters out many irrelevant vectors by calculating the approximate distance from a query vector using the compact approximations of vectors in database. Accordingly, the total elapsed time for similarity search is reduced because the disk I/O time is eliminated by reading the compact approximations instead of original vectors. However, the search time of the VA-file or LPC-file is not much lessened compared to the brute-force search because it requires a lot of computations for calculating the approximate distance. This paper proposes a new bitmap index structure in order to minimize the calculating time. To improve the calculating speed, a specific value of an object is saved in a bit pattern that shows a spatial position of the feature vector on a data space, and the calculation for a distance between objects is performed by the XOR bit calculation that is much faster than the real vector calculation. According to the experiment, the method that this paper suggests has shortened the total searching time to the extent of about one fourth of the sequential searching time, and to the utmost two times of the existing methods by shortening the great deal of calculating time, although this method has a longer data reading time compared to the existing vector approximation based approach. Consequently, it can be confirmed that we can improve even more the searching performance by shortening the calculating time for filtering of the existing vector approximation methods when the database speed is fast enough.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.44 no.10
• /
• pp.156-161
• /
• 2007

In a TDD system, the length of slots can be unequal, and the number of downlink (DL) and uplink (UL) slots per frame can be different as well. The advantage of using TDD is the capability to accommodate asymmetric high-bit-rate services for the DL and It, which will be one of the prominent features in 4G systems. This paper analyzes the performance of TDD system on mobile channel environments like indoor pedestrian and vehicular channel, and proposes optimum modulation/demodulation method in TDD system. A rectangular QAM (RQAM) used in various communication systems has good BER performance but the much more signal amplitudes also have become one of the barriers to implement receiver. While PSK receiver is implemented easily because it has a constant amplitude, but it's BER performance is worse than RQAM. APSK proposed in this paper integrates merits of RQAM and PSK, and minimizes demerits of then And a simple method is also proposed to demodulate the soft symbol. The results indicate that the proposed APSK has a little worse performance than RQAM but the dynamic range of APSK is about 4 dB, 8 dB better than RQAM at 16-ary, 64-ary modulation/demodulation respectively.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.5
• /
• pp.1079-1094
• /
• 2014

In this paper, we study efficient gene selection methods by using conditional mutual information. We suggest gene selection methods using conditional mutual information based on semiparametric methods utilizing multivariate normal distribution and Edgeworth approximation. We compare our suggested methods with other methods such as mutual information filter, SVM-RFE, Cai et al. (2009)'s gene selection (MIGS-original) in SVM classification. By these experiments, we show that gene selection methods using conditional mutual information based on semiparametric methods have better performance than mutual information filter. Furthermore, we show that they take far less computing time than Cai et al. (2009)'s gene selection but have similar performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.10
• /
• pp.1779-1785
• /
• 2006

This paper presents a method of virtual robot arm control by EMG pattern recognition using the proposed hybrid system. The proposed hybrid system is composed of the LVQ and the SOFM, and the SOFM is used for the preprocessing of the LVQ. The SOFM converts the high dimensional EMG signals to 2-dimensional data. The EMG measurement system uses three surface electrodes to acquire the EMG signal from operator. Six hand gestures can be classified sufficiently by the proposed hybrid system. Experimental results are presented that show the effectiveness of the virtual robot arm control by the proposed hybrid system based classifier for the recognition of hand gestures from EMG signal patterns.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.4
• /
• pp.881-891
• /
• 2014

Graphical models can be used as an intuitive tool for modeling a complex stochastic system with a large number of variables related each other because the conditional independence between random variables can be visualized as a network. Graphical least absolute shrinkage and selection operator (LASSO) is considered to be effective in avoiding overfitting in the estimation of Gaussian graphical models for high dimensional data. In this paper, we consider the model selection problem in graphical LASSO. Particularly, we compare various model selection criteria via simulations and analyze a real financial data set.

• The Korean Journal of Applied Statistics
• /
• v.36 no.6
• /
• pp.529-545
• /
• 2023

This paper introduces the Banded-VHAR model suitable for high-dimensional long-memory time series with band structure. The Banded-VHAR model has nonignorable correlations only with adjacent dimensions due to data features, for example, geographical information. Row-wise estimation method is adapted for fast computation. Also, two estimation methods, namely BIC and ratio methods, are proposed to estimate the width of band. We demonstrate asymptotic consistency of our proposed estimation methods through simulation study. Real data applications to pm2.5 and apartment trading volume substantiate that our Banded-VHAR model outperforms traditional sparse VHAR model in forecasting and easy to interpret model coefficients.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.387-389
• /
• 2012

초분광영상은 사람이 볼 있는 가시광선 영역부터 자외선 파장 대역까지 수십에서 수천 개의 데이터를 가지고 있는 고차원 데이터이다. 그렇기 때문에 초분광영상을 이용한 연구에는 많은 저장 공간과 고사양의 성능을 필요로 한다. 따라서 초분광영상의 차원을 감소시켜 데이터용량을 줄이고, 처리속도를 향상시키기 위한 연구들이 이루어지고 있다. 기존에 자주 사용되던 방법인 PCA와 ICA는 차원축소를 위하여 고유벡터를 계산하고 이를 이용하여 축을 변경하여 차원축소를 한다. 하지만 초분광영상에서는 이러한 방법으로 차원을 축소할 시 정확도가 감소한다. 따라서 본 논문에서는 특징 밴드를 추출하고 이를 이용하여 차원축소를 하는 SPVD 알고리즘을 제안한다. SPVD(Spectral pair vector decomposition) 알고리즘은 d개의 그룹으로 나누고 각 그룹들의 양벡터 각과 음벡터 각을 계산한 후 이를 이용하여 차원축소를 한다. 실험 결과 PCA는 61차원에서 70.05%, ICA는 71차원에서 63.03% 정확도를 보이는데 비해 SPVD 알고리즘은 3차원에서 83% 정확도를 보였다.

• The KIPS Transactions:PartB
• /
• v.8B no.4
• /
• pp.327-334
• /
• 2001

본 논문에서는 기울기하강과 동적터널링이 조합된 학습알고리즘의 다층신경망을 이용한 고신회성의 회귀분석 모델을 제안하였다. 기울기하강은 빠른 수렴속도의 최적화가 가능하도록 하기 위함이고, 동적터널링은 국소최적해를 만났을 때 이를 벗어난 새로운 연결가중치를 설정하여 전역최적해로 수렴되도록 하기 위함이다. 또한 대용량의 입력 데이터를 통계적으로 독립인 특징들의 집합으로 변환시키는 주요성분분석 기법의 속성을 살려 학습데이터의 차원을 감소시킴으로서 고차원의 학습데이터에 따른 회귀분석 모델의 제약도 동시에 해결하였다. 제안된 기법의 신경망을 3개의 독립변수 패턴을 가진 암모니아 제조공정문제와 10개의 독립변수 패턴을 가진 자동차 연비문제에 각각 적용하여 시뮬레이션한 결과, 기존의 역전과 알고리즘의 신경망이나 주요성분분석에 의한 차원을 감소시키지 않은 학습패턴을 이용한 신경망보다 각각 더욱 우수한 학습성능과 회귀성능이 있음을 확인할 수 있었다. 또한 학습패턴의 영평균 정규화로 회귀용 신경망의 성능을 더욱 더 개선하였다.