• The Journal of the Acoustical Society of Korea
• /
• v.38 no.6
• /
• pp.670-677
• /
• 2019

As the presence of background noise in acoustic signal degrades the performance of speech or acoustic event recognition, it is still challenging to extract noise-robust acoustic features from noisy signal. In this paper, we propose a combined structure of Wasserstein Generative Adversarial Network (WGAN) and MultiTask AutoEncoder (MTAE) as deep learning architecture that integrates the strength of MTAE and WGAN respectively such that it estimates not only noise but also speech features from noisy acoustic source. The proposed MTAE-WGAN structure is used to estimate speech signal and the residual noise by employing a gradient penalty and a weight initialization method for Leaky Rectified Linear Unit (LReLU) and Parametric ReLU (PReLU). The proposed MTAE-WGAN structure with the adopted gradient penalty loss function enhances the speech features and subsequently achieve substantial Phoneme Error Rate (PER) improvements over the stand-alone Deep Denoising Autoencoder (DDAE), MTAE, Redundant Convolutional Encoder-Decoder (R-CED) and Recurrent MTAE (RMTAE) models for robust speech recognition.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.6
• /
• pp.257-264
• /
• 2019

Recently, deep neural network based approaches have shown a good performance for various fields of natural language processing. A huge amount of training data is essential for building a deep neural network model. However, collecting a large size of training data is a costly and time-consuming job. A data augmentation is one of the solutions to this problem. The data augmentation of text data is more difficult than that of image data because texts consist of tokens with discrete values. Generative adversarial networks (GANs) are widely used for image generation. In this work, we generate sentimental texts by using one of the GANs, CS-GAN model that has a discriminator as well as a classifier. We evaluate the usefulness of generated sentimental texts according to various measurements. CS-GAN model not only can generate texts with more diversity but also can improve the performance of its classifier.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.2
• /
• pp.291-298
• /
• 2022

A generative adversarial network (GAN) is a network in which two internal neural networks (generative network and discriminant network) learn while competing with each other. The generator creates an image close to reality, and the delimiter is programmed to better discriminate the image of the constructor. This technology is being used in various ways to create, transform, and restore the entire image X into another image Y. This paper describes a method that can be forged into another object naturally, after extracting only a partial image from the original image. First, a new image is created through the previously trained DCGAN model, after extracting only a partial image from the original image. The original image goes through a process of naturally combining with, after re-styling it to match the texture and size of the original image using the overall style transfer technique. Through this study, the user can naturally add/transform the desired object image to a specific part of the original image, so it can be used as another field of application for creating fake images.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1588-1594
• /
• 2020

In this paper, we discuss a data balancing method for learning a neural network that generates digital holograms using a deep neural network (DNN). Deep neural networks are based on deep learning (DL) technology and use a generative adversarial network (GAN) series. The fringe pattern, which is the basic unit of a hologram to be created through a deep neural network, has very different data types depending on the hologram plane and the position of the object. However, because the criteria for classifying the data are not clear, an imbalance in the training data may occur. The imbalance of learning data acts as a factor of instability in learning. Therefore, it presents a method for classifying and balancing data for which the classification criteria are not clear. And it shows that learning is stabilized through this.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.942-945
• /
• 2019

현재 딥러닝은 컴퓨터 분야에서 이미지 처리 방법으로 활용도가 높아지면서 딥러닝 모델 개발 연구가 활발히 진행되고 있다. 딥러닝 모델 중에서 이미지 생성모델은 대표적으로 GAN(Generative Adversarial Network, 생성적 적대 신경망) 모델을 활용하고 있다. GAN은 생성기 네트워크와 판별기 네트워크를 이용하여 진짜 같은 이미지를 생성한다. 생성된 이미지는 실제 이미지와의 오차를 최소화해야 하며 이때 사용하는 함수를 손실함수라고 한다. GAN에서 손실함수는 이미지를 생성하는 학습이 불안정하여 이미지 품질이 떨어지는 문제가 있다. 개선된 GAN 관련 연구가 진행되고 있지만 완전한 문제 해결에는 부족하다. 본 논문은 7개의 GAN 모델에서 사용하는 손실함수를 분류하고 특징을 분석한다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.1
• /
• pp.21-29
• /
• 2024

With the development of ICT and network, security management of IT infrastructure that has grown in size is becoming very difficult. Many companies and public institutions are having difficulty managing system and network security. In addition, as the complexity of hardware and software grows, it is becoming almost impossible for a person to manage all security. Therefore, AI is essential for network security management. However, since it is very dangerous to operate an attack model in a real network environment, cybersecurity emulation research was conducted through reinforcement learning by implementing a real-life network environment. To this end, this study applied reinforcement learning to the network environment, and as the learning progressed, the agent accurately identified the vulnerability of the network. When a network vulnerability is detected through AI, automated customized response becomes possible.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.662-665
• /
• 2011

The cognitive radio (CR) network has been studied in the form of open source by vast number of communities, and the potential expectation is very high since the CR is based on reprogrammable platform. However, this characteristics of open-source software take high risk as well. As the peer-to-peer software has been abused, so high is the chance that the CR network can be abused public wide. Consequently, the benefit from the study of next-generation wireless network can be at risk because of the negative impact of violation of communication law or abusing the CR. In this contribution, we analyze the issues and the problems of the CR and discuss an efficient measure against security attacks.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04a
• /
• pp.394-396
• /
• 2002

'분산 모의훈련 전장환경'은 실제 또는 가상전쟁, 전투 등 군사 상황을 묘사하기 위해, 둘 또는 그 이상의 적대 세력간에 발생하는 군사적 상황을 게임규칙, 자료 및 절차를 사용하여 모의하는 군사작전 게임(war game)을 의미한다. 이는 컴퓨터와 네트워크를 이용한 '합성 가상 전장환경'에서 구현되는데, 특성상 다수의 게이머와 게이머, 게이머와 객체, 객체들간에 네트워크를 통한 빈번한 메시지 송.수신을 통한 정보를 교환하고 실시간으로 상호동작 한다. 즉, '합성 가상 전장환경'의 구현을 위해서는 신뢰성과 실시간성을 보장하는 다자간 통신의 보장이 핵심을 이룬다. 그러나, 일반 인터넷상의 브로드캐스팅 기술을 적용할 경우, 가상전장환경에서 폭주하는 통신량을 실시간으로 처리하기엔 어려운 문제점이 있다. 따라서 '합성 가상전장가상환경'의 보다 효과적인 구현을 위해서는 고신뢰 실시간 다중 그룹통신 기술의 개발이 요구된다. 본 논문은 그룹통신 개념에 기반하여 가상 합성 전장 환경에 적용할 수 있는 효율적인 전송 서비스를 위한 네 가지 전송 알고리즘을 제시하며, API 설계를 통해 서비스를 구현한다.

• Journal of Broadcast Engineering
• /
• v.25 no.5
• /
• pp.672-684
• /
• 2020

Restoring a low resolution and motion blurred light field has become essential due to the growing works on parallax-based image processing. These tasks are known as light-field enhancement process. Unfortunately, only a few state-of-the-art methods are introduced to solve the multiple problems jointly. In this work, we design a framework that jointly solves light field spatial super-resolution and motion deblurring tasks. Particularly, we generate a straight-forward neural network that is trained under low-resolution and 6-degree-of-freedom (6-DOF) motion-blurred light field dataset. Furthermore, we propose the strategy of local region optimization on the adversarial network to boost the performance. We evaluate our method through both quantitative and qualitative measurements and exhibit superior performance compared to the state-of-the-art methods.

• Smart Media Journal
• /
• v.8 no.4
• /
• pp.33-37
• /
• 2019

Because the ECG signals are time-series data acquired as time elapses, it is important to obtain comparative data the same in size as the enrolled data every time. This paper suggests a network model of GAN (Generative Adversarial Networks) based on an auxiliary classifier to generate synthetic ECG signals which may address the different data size issues. The Cosine similarity and Cross-correlation are used to examine the similarity of synthetic ECG signals. The analysis shows that the Average Cosine similarity was 0.991 and the Average Euclidean distance similarity based on cross-correlation was 0.25: such results indicate that data size difference issue can be resolved while the generated synthetic ECG signals, similar to real ECG signals, can create synthetic data even when the registered data are not the same as the comparative data in size.