• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.100-100
• /
• 2023

단기 강우 예측에는 주로 물리과정 기반 수치예보모델(NWPs, Numerical Prediction Models) 과 레이더 기반 확률론적 방법이 사용되어 왔으며, 최근에는 머신러닝을 이용한 레이더 기반 강우예측 모델이 단기 강우 예측에 뛰어난 성능을 보이는 것을 확인하여 관련 연구가 활발히 진행되고 있다. 하지만 머신러닝 기반 모델은 예측 선행시간 증가 시 성능이 크게 저하되며, 또한 대기의 물리적 과정을 고려하지 않는 Black-box 모델이라는 한계점이 존재한다. 본 연구에서는 이러한 한계를 극복하기 위해 머신러닝 기반 blending 기법을 통해 물리과정 기반 수치예보모델인 Weather Research and Forecasting (WRF)와 최신 머신러닝 기법 (cGAN, conditional Generative Adversarial Network) 기반 모델을 결합한 Hybrid 강우예측모델을 개발하고자 하였다. cGAN 기반 모델 개발을 위해 1시간 단위 1km 공간해상도의 레이더 반사도, WRF 모델로부터 산출된 기상 자료(온도, 풍속 등), 유역관련 정보(DEM, 토지피복 등)를 입력 자료로 사용하여 모델을 학습하였으며, 모델을 통해 물리 정보 및 머신러닝 기반 강우 예측을 생성하였다. 이렇게 생성된cGAN 기반 모델 결과와 WRF 예측 결과를 결합하는 머신러닝 기반 blending 기법을 통해Hybrid 강우예측 결과를 최종적으로 도출하였다. 본 연구에서는 Hybrid 강우예측 모델의 성능을 평가하기 위해 수도권 및 안동댐 유역에서 발생한 호우 사례를 기반으로 최대 선행시간 6시간까지 모델 예측 결과를 분석하였다. 이를 통해 물리과정 기반 모델과 머신러닝 기반 모델을 결합하는 Hybrid 기법을 적용하여 높은 정확도와 신뢰도를 가지는 고해상도 강수 예측 자료를 생성할 수 있음을 확인하였다.

• Korean Journal of Remote Sensing
• /
• v.39 no.4
• /
• pp.459-466
• /
• 2023

In the last decade, artificial intelligence's dramatic advancement with the development of various deep learning techniques has significantly contributed to remote sensing fields and satellite image applications. Among many prominent areas, super-resolution research has seen substantial growth with the release of several benchmark datasets and the rise of generative adversarial network-based studies. However, most previously published remote sensing benchmark datasets represent spatial resolution within approximately 10 meters, imposing limitations when directly applying for super-resolution of small objects with cm unit spatial resolution. Furthermore, if the dataset lacks a global spatial distribution and is specialized in particular land covers, the consequent lack of feature diversity can directly impact the quantitative performance and prevent the formation of robust foundation models. To overcome these issues, this paper proposes a method to generate benchmark datasets by simulating the modulation transfer functions of the sensor. The proposed approach leverages the simulation method with a solid theoretical foundation, notably recognized in image fusion. Additionally, the generated benchmark dataset is applied to state-of-the-art super-resolution base models for quantitative and visual analysis and discusses the shortcomings of the existing datasets. Through these efforts, we anticipate that the proposed benchmark dataset will facilitate various super-resolution research shortly in Korea.

• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.275-284
• /
• 2023

Gene expression data can be utilized in various studies, including the prediction of disease prognosis. However, there are challenges associated with collecting enough data due to cost constraints. In this paper, we propose a gene expression data generation model based on Conditional Variational Autoencoder. Our results demonstrate that the proposed model generates synthetic data with superior quality compared to two other state-of-the-art models for gene expression data generation, namely the Wasserstein Generative Adversarial Network with Gradient Penalty based model and the structured data generation models CTGAN and TVAE.

• Journal of Korean Society for Quality Management
• /
• v.51 no.1
• /
• pp.119-129
• /
• 2023

Purpose: This study proposes a classification model for implementing condition-based maintenance (CBM) by monitoring the real-time status of a machine using acceleration sensor data collected from a vehicle. Methods: The classification model's performance was improved by applying Fourier transform to convert the acceleration sensor data from the time domain to the frequency domain. Additionally, the Generative Adversarial Network (GAN) algorithm was used to augment images and further enhance the classification model's performance. Results: Experimental results demonstrate that the GAN algorithm can effectively serve as an image augmentation technique to enhance the performance of the classification model. Consequently, the proposed approach yielded a significant improvement in the classification model's accuracy. Conclusion: While this study focused on the effectiveness of the GAN algorithm as an image augmentation method, further research is necessary to compare its performance with other image augmentation techniques. Additionally, it is essential to consider the potential for performance degradation due to class imbalance and conduct follow-up studies to address this issue.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.43.1-43.1
• /
• 2019

In this study, we apply a deep learning model to denoising solar magnetograms. For this, we design a model based on conditional generative adversarial network, which is one of the deep learning algorithms, for the image-to-image translation from a single magnetogram to a denoised magnetogram. For the single magnetogram, we use SDO/HMI line-of-sight magnetograms at the center of solar disk. For the denoised magnetogram, we make 21-frame-stacked magnetograms at the center of solar disk considering solar rotation. We train a model using 7004 paris of the single and denoised magnetograms from 2013 January to 2013 October and test the model using 1432 pairs from 2013 November to 2013 December. Our results from this study are as follows. First, our model successfully denoise SDO/HMI magnetograms and the denoised magnetograms from our model are similar to the stacked magnetograms. Second, the average pixel-to-pixel correlation coefficient value between denoised magnetograms from our model and stacked magnetogrmas is larger than 0.93. Third, the average noise level of denoised magnetograms from our model is greatly reduced from 10.29 G to 3.89 G, and it is consistent with or smaller than that of stacked magnetograms 4.11 G. Our results can be applied to many scientific field in which the integration of many frames are used to improve the signal-to-noise ratio.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.484-486
• /
• 2022

대체 불가능한 토큰을 의미하는 NFT 는 블록 체인 기반 기술 중 하나로 소유권과 거래 이력이 블록 체인에 기록이 된다는 장점을 지녀 미술 작품 거래에도 활발히 활용되고 있다. 하지만 현재 NFT 작품 거래 시장은 주식거래와 같이 회전율과 거래량을 중점으로 투기성이 짙으며, 작품의 섬네일 이미지만으로 거래 의사를 판단하는 경우가 많다는 문제점이 대두되고 있다. 따라서 작품의 다양한 특성에 대한 고려가 함께 이루어질 필요가 있다. 본 논문은 작품의 다양한 변수를 수집하여 최적의 변수 조합을 찾아내고, 이미지 전이 학습과 생성 알고리즘을 활용해 다양한 변수가 가격에 미치는 영향을 알아보고자 했다. 연구 결과 이미지 만을 활용한 가격 예측은 정확도가 높지 않다는 사실을 확인하였고 작품이 가진 여러 특성을 변수로 한 정형 데이터의 가격 예측 정확도가 더 높은 것을 알 수 있었다. 또한 생성 알고리즘을 통해 새로운 특성 조합의 작품들을 만들었고 가격을 수치 예측해보았다. 이를 실제 작가의 작품의 가격과 가격의 변동 추이를 대조해 작가의 작품 판매 가격 평균치 이상인 작품의 특성을 확인할 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.522-524
• /
• 2023

위치 기반 서비스(LBS)의 확산으로 다양한 분야에서 활용할 수 있는 많은 양의 경로 데이터가 생성되고 있다. 하지만 공격자가 경로 데이터를 통해 잠재적으로 사용자의 개인정보를 유추할 수 있다는 문제점이 존재한다. 따라서 경로 데이터의 프라이버시를 보존하며 유용성을 유지할 수 있는 GAN(Generative Adversarial Network)을 사용한 많은 연구가 진행되고 있다. 그러나 GAN은 생성된 결과물을 제어하지 못한다는 한계점을 가지고 있다. 본 논문에서는 ACGAN(Auxiliary classifier GAN)을 통해 생성된 결과물을 제어함으로써 경로 데이터의 민감한 정점을 숨기고, Attention mechanism을 결합하여 높은 유용성과 익명성을 제공하는 합성 경로 생성 모델인 TAP-GAN(Trajectory attention and protection-GAN)을 제안한다. 또한 모델의 성능을 입증하기 위해 유용성 및 익명성 실험을 진행하고, 선행 연구 모델과의 비교를 통해 TAP-GAN이 경로 데이터의 유용성을 보장하면서 사용자의 프라이버시를 효과적으로 보호할 수 있음을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.7
• /
• pp.2304-2320
• /
• 2021

Intelligently detecting anomalies in health sensor data streams (e.g., Electrocardiogram, ECG) can improve the development of E-health industry. The physiological signals of patients are collected through sensors. Timely diagnosis and treatment save medical resources, promote physical health, and reduce complications. However, it is difficult to automatically classify the ECG data, as the features of ECGs are difficult to extract. And the volume of labeled ECG data is limited, which affects the classification performance. In this paper, we propose a Generative Adversarial Network (GAN)-based deep learning framework (called CAB) for heart arrhythmia classification. CAB focuses on improving the detection accuracy based on a small number of labeled samples. It is trained based on the class-imbalance ECG data. Augmenting ECG data by a GAN model eliminates the impact of data scarcity. After data augmentation, CAB classifies the ECG data by using a Bidirectional Long Short Term Memory Recurrent Neural Network (Bi-LSTM). Experiment results show a better performance of CAB compared with state-of-the-art methods. The overall classification accuracy of CAB is 99.71%. The F1-scores of classifying Normal beats (N), Supraventricular ectopic beats (S), Ventricular ectopic beats (V), Fusion beats (F) and Unclassifiable beats (Q) heartbeats are 99.86%, 97.66%, 99.05%, 98.57% and 99.88%, respectively. Unclassifiable beats (Q) heartbeats are 99.86%, 97.66%, 99.05%, 98.57% and 99.88%, respectively.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.69-76
• /
• 2020

This study was conducted with the support of the Information and Communication Technology Promotion Center, funded by the government (Ministry of Science and ICT) in 2019. Highly pathogenic avian influenza (HPAI) is an acute infectious disease of birds caused by highly pathogenic avian influenza virus infection, causing serious damage to poultry such as chickens and ducks. High pathogenic avian influenza (HPAI) is caused by focusing on winter rather than year-round, and sometimes does not occur at all during a certain period of time. Due to these characteristics of HPAI, there is a problem that does not accumulate enough actual data. In this paper study, GAN network was utilized to generate actual similar data containing missing values and the process is introduced. The results of this study can be used to measure risk by generating realistic simulation data for certain times when HPAI did not occur.

• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.687-701
• /
• 2022

Inevitable response loss under complex operational conditions significantly affects the integrity and quality of measured data, leading the structural health monitoring (SHM) ineffective. To remedy the impact of data loss, a common way is to transfer the recorded response of available measure point to where the data loss occurred by establishing the response mapping from measured data. However, the current research has yet addressed the structural condition changes afterward and response mapping learning from a small sample. So, this paper proposes a novel data driven structural response reconstruction method based on a sophisticated designed generating adversarial network (UAGAN). Advanced deep learning techniques including U-shaped dense blocks, self-attention and a customized loss function are specialized and embedded in UAGAN to improve the universal and representative features extraction and generalized responses mapping establishment. In numerical validation, UAGAN efficiently and accurately captures the distinguished features of structural response from only 40 training samples of the intact structure. Besides, the established response mapping is universal, which effectively reconstructs responses of the structure suffered up to 10% random stiffness reduction or structural damage. In the experimental validation, UAGAN is trained with ambient response and applied to reconstruct response measured under earthquake. The reconstruction losses of response in the time and frequency domains reached 16% and 17%, that is better than the previous research, demonstrating the leading performance of the sophisticated designed network. In addition, the identified modal parameters from reconstructed and the corresponding true responses are highly consistent indicates that the proposed UAGAN is very potential to be applied to practical civil engineering.