• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.448-448
• /
• 2021

최근 급격한 기후변화와 도시화 및 산업화로 인한 지류하천에서의 수량과 수질의 변동은 생물 다양성 감소와 수생태계 건강성 저하에 큰 영향을 미치고 있다. 효율적인 수생태 관리를 위해서는 지속적인 유량, 수질, 그리고 수생태 모니터링을 통한 데이터 축적과 더불어 면밀한 상관 분석을 통해 수생태계 건강성의 악화 원인을 규명해야 할 필요가 있다. 그러나 수많은 지류하천을 대상으로 한 지속적인 모니터링은 현실적으로 어려움이 있으며, 수생태계의 특성 상 단일 영향 인자만으로 수생태계의 건강성 변화와의 관계를 정확히 파악하는데 한계가 있다. 따라서 지류하천에서의 유량 및 수질의 시공간적인 변동성과 다양한 영향 인자를 고려하여 수생태계의 건강성을 효율적으로 예측할 수 있는 기술이 필요하다. 이에 본 연구에서는 경험적 데이터 기반의 머신러닝 모델 구축을 통해 미계측 하천에서의 수생태계 건강성 지수(BMI, TDI, FAI)의 등급(A to E)을 예측하고자 하였다. 머신러닝 모델은 학습 데이터셋의 양과 질에 따라 성능이 크게 달라질 수 있으며, 학습 데이터셋의 분포가 불균형적일 경우 과적합 또는 과소적합 문제가 발생할 수 있다. 이를 보완하고자 본 연구에서는 실제 측정망 데이터셋을 바탕으로 생성적 적대 신경망 GAN(Generative Adversarial Network) 알고리즘을 통해 머신러닝 모델 학습에 필요한 추가 데이터셋(유량, 수질, 기상, 수생태 등급)을 확보하였다. 머신러닝 모델의 성능은 5차 교차검증 과정을 통해 평가하였으며, GAN 데이터셋의 정확도는 실제 측정망 데이터셋의 정규분포와의 비교 분석을 통해 평가하였다. 최종적으로 SWAT(Soil and Water Assessment Tool) 모형을 통해 예측 된 미계측 하천에서의 데이터셋을 머신러닝 모델의 검증 자료로 사용하여 수생태계 건강성 등급 예측 정확도를 평가하였다. 본 연구에서의 GAN에 의해 강화된 머신러닝 모델은 수질 및 수생태 관리가 필요한 우심 지류하천 선정과 구조적/비구조적 최적관리기법에 따른 수생태계 건강성 개선 효과를 평가하는데 활용될 수 있을 것이다. 또한 이를 통해 예측된 미계측 하천에서의 수생태계 건강성 등급 자료는 수량-수질-수생태를 유기적으로 연계한 통합 물관리 정책을 수립하는데 기초자료로 활용될 수 있을 것이라 사료된다.

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.766-769
• /
• 2021

In order to perform tasks such as design, control, optimization, and prediction of flight vehicle trajectories based on machine learning techniques including deep learning, a certain amount of flight vehicle trajectory data is required. However, there are cases in which it is difficult to secure more than a certain amount of flight vehicle trajectory data for various reasons. In such cases, synthetic data generation could be one way to make machine learning possible. In this paper, to explore this possibility, we generated and evaluated synthetic flight vehicle trajectory data using time-series generative adversarial neural network. In addition, various ablation studies (comparative experiments) were performed to explore the possibility of using synthetic data in the aircraft trajectory prediction task. The experimental results presented in this paper are expected to be of practical help to researchers who want to conduct research on the possibility of using synthetic data in the generation of synthetic flight vehicle trajectory data and the work related to flight vehicle trajectories.

• Convergence Security Journal
• /
• v.21 no.2
• /
• pp.57-66
• /
• 2021

Deep neural networks (DNNs) provide excellent performance for image, speech, and pattern recognition. However, DNNs sometimes misrecognize certain adversarial examples. An adversarial example is a sample that adds optimized noise to the original data, which makes the DNN erroneously misclassified, although there is nothing wrong with the human eye. Therefore studies on defense against adversarial example attacks are required. In this paper, we have experimentally analyzed the success rate of detection for adversarial examples by adjusting various parameters. The performance of the ensemble defense method was analyzed using fast gradient sign method, DeepFool method, Carlini & Wanger method, which are adversarial example attack methods. Moreover, we used MNIST as experimental data and Tensorflow as a machine learning library. As an experimental method, we carried out performance analysis based on three adversarial example attack methods, threshold, number of models, and random noise. As a result, when there were 7 models and a threshold of 1, the detection rate for adversarial example is 98.3%, and the accuracy of 99.2% of the original sample is maintained.

• Journal of Internet Computing and Services
• /
• v.23 no.6
• /
• pp.39-48
• /
• 2022

Cyberattacks around the world continue to increase, and their damage extends beyond government facilities and affects civilians. These issues emphasized the importance of developing a system that can identify and detect cyber anomalies early. As above, in order to effectively identify cyber anomalies, several studies have been conducted to learn BGP (Border Gateway Protocol) data through a machine learning model and identify them as anomalies. However, BGP data is unbalanced data in which abnormal data is less than normal data. This causes the model to have a learning biased result, reducing the reliability of the result. In addition, there is a limit in that security personnel cannot recognize the cyber situation as a typical result of machine learning in an actual cyber situation. Therefore, in this paper, we investigate BGP (Border Gateway Protocol) that keeps network records around the world and solve the problem of unbalanced data by using SMOTE. After that, assuming a cyber range situation, an autoencoder classifies cyber anomalies and visualizes the classified data. By learning the pattern of normal data, the performance of classifying abnormal data with 92.4% accuracy was derived, and the auxiliary index also showed 90% performance, ensuring reliability of the results. In addition, it is expected to be able to effectively defend against cyber attacks because it is possible to effectively recognize the situation by visualizing the congested cyber space.

• Convergence Security Journal
• /
• v.23 no.4
• /
• pp.61-70
• /
• 2023

Clustering is an unsupervised learning method that involves grouping data based on features such as distance metrics, using data without known labels or ground truth values. This method has the advantage of being applicable to various types of data, including images, text, and audio, without the need for labeling. Traditional clustering techniques involve applying dimensionality reduction methods or extracting specific features to perform clustering. However, with the advancement of deep learning models, research on deep clustering techniques using techniques such as autoencoders and generative adversarial networks, which represent input data as latent vectors, has emerged. In this study, we propose a deep clustering technique based on deep learning. In this approach, we use an autoencoder to transform the input data into latent vectors, and then construct a vector space according to the cluster structure and perform k-means clustering. We conducted experiments using the MNIST and Fashion-MNIST datasets in the PyTorch machine learning library as the experimental environment. The model used is a convolutional neural network-based autoencoder model. The experimental results show an accuracy of 89.42% for MNIST and 56.64% for Fashion-MNIST when k is set to 10.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.495-497
• /
• 2022

본 데이터 GAN 기법 데이터 증강기법을 적용하여 가속도 데이터를 증강하는 방법에 대해 연구한다. 가속도 데이터는 사람의 활동패턴을 인지하는데 있어 가장 기본적인 데이터로 활용된다. 가속도 데이터를 증강한 뒤, 활동패턴을 인지하는 머신러닝 모델 훈련에 사용한 결과 생성한 데이터가 육안으로 확인하였을 때 실제 데이터와 유사한 패턴을 형성하였고, 실제 활동패턴인지 모델 훈련에 사용한 결과 정확도(Accuracy)는 기존 데이터로만 훈련한 경우 74%인데 비해 증강된 데이터를 혼합하여 훈련하였을 때 약 88%로 개선된 것을 확인하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.700-705
• /
• 2020

Generative adversarial networks are methods of generating images by opposing two neural networks. When generating the image, randomly generated noise is rearranged to generate the image. The image generated by this method is not generated well depending on the noise, and it is difficult to generate a proper image when the number of pixels of the image is small In addition, the speed and size of data accumulation in data classification increases, and there are many difficulties in labeling them. In this paper, to solve this problem, we propose a technique to generate noise based on random noise using real data. Since the proposed system generates an image based on the existing image, it is confirmed that it is possible to generate a more natural image, and if it is used for learning, it shows a higher hit rate than the existing method using the hostile neural network respectively.

• 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 Convergence for Information Technology
• /
• v.12 no.4
• /
• pp.260-269
• /
• 2022

In this paper, We propose an anomaly detection model using deep neural network to automate the identification of outliers of the national air pollution measurement network data that is previously performed by experts. We generated training data by analyzing missing values and outliers of weather data provided by the Institute of Environmental Research and based on the BeatGAN model of the unsupervised learning method, we propose a new model by changing the kernel structure, adding the convolutional filter layer and the transposed convolutional filter layer to improve anomaly detection performance. In addition, by utilizing the generative features of the proposed model to implement and apply a retraining algorithm that generates new data and uses it for training, it was confirmed that the proposed model had the highest performance compared to the original BeatGAN models and other unsupervised learning model like Iforest and One Class SVM. Through this study, it was possible to suggest a method to improve the anomaly detection performance of proposed model while avoiding overfitting without additional cost in situations where training data are insufficient due to various factors such as sensor abnormalities and inspections in actual industrial sites.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.340-340
• /
• 2020

본 연구는 산지-도심-해안으로 연결된 해안도시의 지역적인 특성으로 인해 해일과 같은 조위 상승 원인이 침수 및 홍수 위험을 증가시키는 복합재난에 대응하기 위하여 실시간 침수 위험정보 확인 및 대응, 대시민 상황전파 등이 가능하도록 IoT 융합 침수 대응시스템을 개발하는 것이다. 또한, 침수 및 홍수방어벽 등의 구조적대책을 검토하여 실시간 IoT 융합 침수 대응시스템과 연계운영함으로써 해안도시의 시설 및 재산을 보호하고 시민의 안전을 확보하는데 기여할 수 있는 실증연구를 수행하는 것을 목표로 한다. 침수 대응을 위한 구조적 대책은 양산형 IoT 침수 및 하천수위 감지 장비, 센서 연동 침수 및 홍수방어벽, 복합재난알림 디스플레이 등의 개발이며, 비구조적 대책은 강우-하천수위-조위 실측자료기반의 머신러닝 침수위험 분석 모듈 개발, 수치모의에 의한 예상 침수심 및 침수위험지도 작성, 재난상황 전파 알림서비스 등이 있으며 개발된 시제품과 기술을 테스트베드에 적용하고 부산광역시에서 운영하는 재난상황관리시스템인 스마트빅보드(SBB, Smart Big Board)에 연계 및 탑재함으로써 향상된 침수대응 및 대시민 서비스가 가능한 시스템을 구축 및 실증하고자 한다. 센서, 통신, 자료연계 및 분석에 대한 IoT 융합 기술을 침수 대응 기술개발에 적극 활용하여 실측자료와 수치모의 분석 결과가 연계된 침수위험정보를 생산하고 실시간 대응이 가능한 효율적이고 실질적으로 피해를 저감 할 수 있는 기술을 개발하고 적용성을 입증한다면 국내외 해안도시에 확대 보급이 가능해져 침수 피해로부터 효과적으로 국민의 재산과 생명 보호할 수 있다. 또한, IoT 기술을 연계한 재난 모니터링 시스템의 확산으로 복합재난 대응 연구에 기여할 수 있고 상대적으로 실측자료 수집, 연구에 소외되어 온 소하천 등에도 적용이 가능하여 재난으로 인한 사회적 비용을 경감시킬 수 있을 것으로 기대된다.