• The Bulletin of The Korean Astronomical Society
• /
• v.43 no.1
• /
• pp.38.4-39
• /
• 2018

We analyze the Planck 2015 cosmic microwave background temperature fluctuation data to find any anomaly in the angular power spectra measured for partial regions on the sky. For disks with radius of $20^{\circ}$, $45^{\circ}$ and $90^{\circ}$, which are densely overlapping on the sky, we estimate the power excess and its statistical significance relative to the LambdaCDM expectation for some chosen ranges of angular scales. We also investigate the dipolar asymmetry using the power excess maps obtained for some chosen angular scales, and confirm the previously announced consistent dipole directions. The average dipole amplitude and the inner products of dipoles have been measured from the power excess maps at different angular scales. We conclude that although dipole directions are consistent the measured amplitudes are not statistically significant compared to the LambdaCDM model prediction.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.6 no.2
• /
• pp.199-207
• /
• 2010

Recently, large amount of information in IDS(Intrusion Detection System) can be un manageable and also be mixed with false prediction error. In this paper, we propose a data mining methodology for IDS, which contains uncertainty based on training process and post-processing analysis additionally. Our system is trained to classify the existing attack for misuse detection, to detect the new attack pattern for anomaly detection, and to define border patter between attack and normal pattern. In experimental results show that our approach improve the performance against existing attacks and new attacks, from 0.62 to 0.84 about 35%.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.04b
• /
• pp.927-930
• /
• 2002

본 논문에서는 중요 프로세스(privileged process)의 시스템 호출 순서(system call sequence)를 이용한 침입탐지 시스템을 제안한다. 기존 연구의 정상행위 기반 침입탐지 시스템은 정상행위를 모델링하여 시스템을 구성하고, 이와 비교를 통해 프로세스의 이상(anomaly) 여부를 결정한다. 이러한 방법은 모델링되지 않은 미지의 행위에 대한 적절한 판단을 행할 수 없으므로, 높은 오류율(false-positive/negative)을 보인다. 본 논문에서는 현재까지 알려진 공격에서 공통적으로 나타나는 윈도우들을 수집하여 침입예상윈도우를 구축하고, 이를 기존의 침입탐지 시스템에 부가적으로 사용하여 효과적으로 오류율(false-positive/negative)을 낮출 수 있음을 보인다. 실험 결과 제안된 방법을 통한 침입탐지는 기존의 방법에 비해 공격 탐지율은 증가하고 정상행위에 대한 오류율은 감소하였다.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1049-1058
• /
• 2022

HAR(Human Action Recognition) such as anomaly and object detection has become a trend in research field(s) that focus on utilizing Artificial Intelligence (AI) methods to analyze patterns of human action in crime-ridden area(s), media services, and industrial facilities. Especially, in real-time system(s) using video streaming data, HAR has become a more important AI-based research field in application development and many different research fields using HAR have currently been developed and improved. In this paper, we propose and analyze a deep-learning-based HAR that provides more efficient scheme(s) using an intelligent AI models, such system can be applied to media services using RGB video streaming data usage without feature extraction pre-processing. For the method, we adopt Slowfast based on the Deep Neural Network(DNN) model under an open dataset(HMDB-51 or UCF101) for improvement in prediction accuracy.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.404-407
• /
• 2021

석유개발 현장에서 시추장비의 고장으로 인한 장비교체 및 시추시간 증가는 막대한 비용소모를 발생시킨다. 본 논문은 딥러닝 기반의 시추장비 중 드릴비트의 동력을 구동시키는 디젤엔진의 고장 요소를 분류하고 이 요소에 따른 고장여부를 판별하는 딥러닝 기반의 이상 예측 및 진단 모델을 개발하였다. 또한 제안한 모델의 우수성을 확인하기 위해 로지스틱 회귀분석 분류모델과의 예측성능 비교분석도 수행하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.122-123
• /
• 2022

최근 들어 인공지능에 대한 발달과 많은 매체들로 인해 사람들의 관심이 증가하고 있다. 또한 GPS 나 Beacon 과 같이 위치 측위 기술이 증가함에 따라 실외 측위 기술이 많이 발달되었고, 실내에서도 사용자의 정확한 위치를 측정할 수 있는 기술들이 발달되고 있다. 본 논문에서는 RNN 알고리즘을 이용하여 비콘을 통해 수집된 사용자의 반복적이고 순차적인 위치정보, 타임스탬프 데이터를 학습시키고 ECG 를 결합하여 사용자 인증을 하여 사용자의 시간별 위치 예측과 이상 징후 탐지 시스템을 제안하고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.498-500
• /
• 2022

공공 안전을 위한 영상 감시 시스템이 증가함에 따라 CCTV 관제사가 관제해야 할 영상의 수가 증가하고 있다. 점점 증가하는 관제 영상 수로 인해 CCTV 관제사는 수많은 영상 사이에서 발생하는 살인, 강도, 폭력 등 위급한 이상 상황을 놓치는 문제가 발생할 수 있다. 이러한 문제를 해결하기 위해 최근에는 영상에서 발생하는 이상 상황을 자동으로 탐지하고 CCTV 관제사에게 알려 관제 효율을 향상시키는 연구가 진행되고 있다. 본 논문은 영상에서 발생하는 이상 상황을 자동으로 탐지하기 위해 예측 기반 이상 탐지 방법에 다중 프레임 예측 에러를 활용해서 영상 이상 탐지 정확도를 향상시키는 방법을 제안한다. 결과적으로 제안한 방법을 사용함으로써 프레임 레벨 AUC가 Ped2 데이터 셋에서 92.70%에서 94.56%, Avenue 데이터셋에서 87.37%에서 89.17%로 상승하였다.

• Journal of the Korean Geographical Society
• /
• v.45 no.3
• /
• pp.319-341
• /
• 2010

The purposes of this study are to classify heavy snowfall types in the Republic of Korea based on fresh snowfall data and atmospheric circulation data during the last 36(1973/74-2008/09) snow seasons and to identify typical surface synoptic climate patterns that characterize each heavy snowfall type. Four synoptic climate categories and seventeen regional heavy snowfall types are classified based on sea level pressure/surface wind vector patterns in East Asia and frequent spatial clustering patterns of heavy snowfall in the Republic of Korea, respectively. Composite analyses of multiple surface synoptic weather charts demonstrate that the locations and intensity of pressure/wind vector mean and anomaly cores in East Asia differentiate each regional heavy snowfall type in Korea. These differences in synoptic climatic fields are primarily associated with the surge of the Siberian high pressure system and the appearance of low pressure systems over the Korean Peninsula. In terms of hemispheric atmospheric circulation, synoptic climatic patterns in the negative mode of winter Arctic Oscillation (AO) are also associated with frequent heavy snowfall in the Republic of Korea at seasonal scales. These results from long-term synoptic climatic data could contribute to improvement of short-range or seasonal prediction of regional heavy snowfall.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.3989-4006
• /
• 2020

The intelligent agriculture monitoring is based on the perception and analysis of environmental data, which enables the monitoring of the production environment and the control of environmental regulation equipment. As the scale of the application continues to expand, a large amount of data will be generated from the perception layer and uploaded to the cloud service, which will bring challenges of insufficient bandwidth and processing capacity. A fog-based offline and real-time hybrid data analysis architecture was proposed in this paper, which combines offline and real-time analysis to enable real-time data processing on resource-constrained IoT devices. Furthermore, we propose a data process-ing algorithm based on the incremental principal component analysis, which can achieve data dimensionality reduction and update of principal components. We also introduce the concept of Squared Prediction Error (SPE) value and realize the abnormal detection of data through the combination of SPE value and data fusion algorithm. To ensure the accuracy and effectiveness of the algorithm, we design a regular-SPE hybrid model update strategy, which enables the principal component to be updated on demand when data anomalies are found. In addition, this strategy can significantly reduce resource consumption growth due to the data analysis architectures. Practical datasets-based simulations have confirmed that the proposed algorithm can perform data fusion and exception processing in real-time on resource-constrained devices; Our model update strategy can reduce the overall system resource consumption while ensuring the accuracy of the algorithm.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.979-987
• /
• 2021

Multiple Sclerosis (MS) can be early diagnosed by detecting lesions in brain magnetic resonance images (MRI). Unsupervised anomaly detection methods based on autoencoder have been recently proposed for automated detection of MS lesions. However, these autoencoder-based methods were developed only for 2D images (e.g. 2D cross-sectional slices) of MRI, so do not utilize the full 3D information of MRI. In this paper, therefore, we propose a novel 3D autoencoder-based framework for detection of the lesion volume of MS in MRI. We first define a 3D convolutional neural network (CNN) for full MRI volumes, and build each encoder and decoder layer of the 3D autoencoder based on 3D CNN. We also add a skip connection between the encoder and decoder layer for effective data reconstruction. In the experimental results, we compare the 3D autoencoder-based method with the 2D autoencoder models using the training datasets of 80 healthy subjects from the Human Connectome Project (HCP) and the testing datasets of 25 MS patients from the Longitudinal multiple sclerosis lesion segmentation challenge, and show that the proposed method achieves superior performance in prediction of MS lesion by up to 15%.