• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.468-475
• /
• 2020

A good acoustic word embedding that can well express the characteristics of word plays an important role in wake-up word detection (WWD). However, the representation ability of acoustic word embedding may be weakened due to various types of environmental noise occurred in the place where WWD works, causing performance degradation. In this paper, we proposed triplet loss based Domain Adversarial Training (tDAT) mitigating environmental factors that can affect acoustic word embedding. Through experiments in noisy environments, we verified that the proposed method effectively improves the conventional DAT approach, and checked its scalability by combining with other method proposed for robust WWD.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1511-1514
• /
• 2002

This paper presents content adaptive image watermark embedding using stochastic visual model based on multiwavelet transform. To embedding watermark, the original image is decomposed into 4 levels using a discrete multiwavelet transform, then a watermark is embedded into the JND(just noticeable differences) of the image each subband. The perceptual model is applied with a stochastic approach fer watermark embedding. This is based on the computation of a NVF(noise visibility function) that have local image properties. The perceptual model with content adaptive watermarking algorithm embed at the texture and edge region for more strongly embedded watermark by the JND. This method uses stationary Generalized Gaussian model characteristic because watermark has noise properties. The experiment results of simulation of the proposed watermark embedding method using stochastic visual model based on multiwavelet transform techniques was found to be excellent invisibility and robustness.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.5
• /
• pp.1027-1037
• /
• 2019

Conventional cyber-attack detection solutions are generally based on signature-based or malicious behavior analysis so that have had difficulty in detecting unknown method-based attacks. Since the various information occurring all the time reflects the state of the system, by modeling it in a steady state and detecting an abnormal state, an unknown attack can be detected. Since a variety of system information occurs in a string form, word embedding, ie, techniques for converting strings into vectors preserving their order and semantics, can be used for modeling and detection. Novelty Detection, which is a technique for detecting a small number of abnormal data in a plurality of normal data, can be performed in order to detect an abnormal condition. This paper proposes a method to detect system anomaly by cyber attack using embedding and novelty detection.

• Proceedings of the IEEK Conference
• /
• 2002.07a
• /
• pp.172-175
• /
• 2002

This paper presents perceptual model with a stochastic rnultiresolution characteristic that can be applied with watermark embedding in the biorthogonal wavelet domain. The perceptual model with adaptive watermarking algorithm embed at the texture and edge region for more strongly embedded watermark by the SSQ(successive subband quantization). The watermark embedding is based on the computation of a NVF(noise visibility function) that have local image properties. This method uses non-stationary Gaussian model stationary Generalized Gaussian model because watermark has noise properties. In order to determine the optimal NVF, we consider the watermark as noise. The particularities of embedding in the stationary GG model use shape parameter and variance of each subband regions in multiresolution. To estimate the shape parameter, we use a moment matching method. Non-stationary Gaussian model use the local mean and variance of each subband. The experiment results of simulation were found to be excellent invisibility and robustness. Experiments of such distortion are executed by Stirmark benchmark test.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.1
• /
• pp.382-403
• /
• 2020

Steganography is the art of concealing the existence of a secret data in a non-secret digital carrier called cover media. While the image of steganography methods is extensively researched, studies on other cover files remain limited. Videos are promising research items for steganography primitives. This study presents an improved approach to video steganography. The improvement is achieved by allowing senders and receivers exchanging secret data without embedding the hidden data in the cover file as in traditional steganography methods. The method is based mainly on searching for exact matches between the secret text and the video frames RGB channel pixel values. Accordingly, a random key-dependent data is generated, and Elliptic Curve Public Key Cryptography is used. The proposed method has an unlimited embedding capacity. The results show that the improved method is secure against traditional steganography attacks since the cover file has no embedded data. Compared to other existing Steganography video systems, the proposed system shows that the method proposed is unlimited in its embedding capacity, system invisibility, and robustness. The system achieves high precision for data recovery in the receiver. The performance of the proposed method is found to be acceptable across different sizes of video files.

• Journal of Korea Multimedia Society
• /
• v.17 no.1
• /
• pp.34-42
• /
• 2014

In this paper an embedding algorithm based on commute time is implemented by organizing patches according to the graph-based metric, and its performance is analyzed by comparing with the results of principal component analysis embedding. It is usual that the dimensionality reduction be done within some acceptable approximation error. However this paper shows the proposed manifold embedding method generates the intrinsic geometry corresponding to the signal despite severe approximation error, so that it can be applied to the areas such as pattern classification or machine learning.

• Journal of Convergence for Information Technology
• /
• v.11 no.7
• /
• pp.31-38
• /
• 2021

In this paper, in order to obtain the optimization of the RNN model used for sentiment analysis, the correlation of each model was studied by observing the trend of loss and accuracy according to hyperparameter tuning. As a research method, after configuring the hidden layer with LSTM and the embedding layer that are most optimized to process sequential data, the loss and accuracy of each model were measured by tuning the unit, batch-size, and embedding size of the LSTM. As a result of the measurement, the loss was 41.9% and the accuracy was 11.4%, and the trend of the optimization model showed a consistently stable graph, confirming that the tuning of the hyperparameter had a profound effect on the model. In addition, it was confirmed that the decision of the embedding size among the three hyperparameters had the greatest influence on the model. In the future, this research will be continued, and research on an algorithm that allows the model to directly find the optimal hyperparameter will continue.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39B no.8
• /
• pp.491-498
• /
• 2014

Due to the ossification of current Internet, it is hard to accommodate new service requirements. One of the solutions to this problem is network virtualization. In this paper, we propose a heuristic virtual network embedding method for network virtualization. The proposed method checks whether the candidate substrate nodes in the substrate network have the possibility of satisfying virtual link requirements. It gives priority to the virtual nodes and the substrate nodes, and embeds the node with higher priority first. Also, the proposed method tries to cluster the mapped substrate nodes if possible. We evaluate the performance of the proposed method in terms of time complexity and virtual network acceptance rate.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.2
• /
• pp.162-170
• /
• 2015

Commute time embedding involves computing the spectral decomposition of the graph Laplacian. It requires the computational burden proportional to $o(n^3)$, not suitable for large scale dataset. Many methods have been proposed to accelerate the computational time, which usually employ the Nystr${\ddot{o}}$m methods to approximate the spectral decomposition of the reduced graph Laplacian. They suffer from the lost of information by dint of sampling process. This paper proposes to reduce the errors by approximating the spectral decomposition of the graph Laplacian using that of the affinity matrix. However, this can not be applied as the data size increases, because it also requires spectral decomposition. Another method called approximate commute time embedding is implemented, which does not require spectral decomposition. The performance of the proposed algorithms is analyzed by computing the commute time on the patch graph.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.11
• /
• pp.1233-1240
• /
• 2014

For the last three decades after the advent of the Toffoli gate in 1980, while many reversible circuit syntheses have been presented reversible embedding methods onto suitable reversible functions, only a few proposed direct irreversible-to-reversible mapping methods without reversible embedding. In this paper we present two effective policies to reduce the gate cost and complexity for the existing direct reversible mapping methods without reversible embedding. In order to develop new cost reduction policies we consider the cost influence of Toffoli module according to NOT gate arrangement in classical circuits. From this we deduced an inverse proportional property between inverting input numbers of classical AND/OR gates and reversible Toffoli module cost based on a fact - the inverting inputs of classical AND(OR) gates increase(decrease) the Toffoli module cost. We confirm the applications of the inverting input rearrangement and maximum fan-out policies preceding direct reversible mapping will be effective method to improve the reversible Toffoli module cost and complexity with the parallel using of the fan-out and supercell ones.