• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2298-2315
• /
• 2024

This paper introduces a method aimed at diagnosing the presence or absence of lesions by detecting anomalies in Chest X-ray images. The proposed approach is based on the PatchCore anomaly detection method, which extracts a feature vector containing location information of an image patch from normal image data and calculates the anomaly distance from the normal vector. However, applying PatchCore directly to medical image processing presents challenges due to the possibility of diseases occurring only in specific organs and the presence of image noise unrelated to lesions. In this study, we present an image alignment method that utilizes affine transformation parameter prediction to standardize already captured X-ray images into a specific composition. Additionally, we introduce a region-specific abnormality detection method that requires affine-transformed chest X-ray images. Furthermore, we propose a method to enhance application efficiency and performance through feature map hard masking. The experimental results demonstrate that our proposed approach achieved a maximum AUROC (Area Under the Receiver Operating Characteristic) of 0.774. Compared to a previous study conducted on the same dataset, our method shows a 6.9% higher performance and improved accuracy.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.3
• /
• pp.130-139
• /
• 2024

Artificial intelligence models are being used to detect facility anomalies using physics data such as vibration, current, and temperature for predictive maintenance in the manufacturing industry. Since the types of facility anomalies, such as facility defects and failures, anomaly detection methods using autoencoder-based unsupervised learning models have been mainly applied. Normal or abnormal facility conditions can be effectively classified using the reconstruction error of the autoencoder, but there is a limit to identifying facility anomalies specifically. When facility anomalies such as unbalance, misalignment, and looseness occur, the facility vibration frequency shows a pattern different from the normal state in a specific frequency range. This paper presents an N-segmentation anomaly detection method that performs anomaly detection by dividing the entire vibration frequency range into N regions. Experiments on nine kinds of anomaly data with different frequencies and amplitudes using vibration data from a compressor showed better performance when N-segmentation was applied. The proposed method helps materialize them after detecting facility anomalies.

• Atmosphere
• /
• v.28 no.4
• /
• pp.479-490
• /
• 2018

This study assesses the prediction skill of regional scale model for the mean temperature anomaly over South Korea produced by Pusan National University Coupled General Circulation Model (PNU CGCM)-Weather Research and Forecasting (WRF) chain. The initial and boundary conditions of WRF are derived from PNU CGCM. The hindcast period is 11 years from 2007 to 2017. The model's prediction skill of mean temperature anomaly is evaluated in terms of the temporal correlation coefficient (TCC), root mean square error (RMSE) and skill scores which are Heidke skill score (HSS), hit rate (HR), false alarm rate (FAR). The predictions of WRF and PNU CGCM are overall similar to observation (OBS). However, TCC of WRF with OBS is higher than that of PNU CGCM and the variation of mean temperature is more comparable to OBS than that of PNU CGCM. The prediction skill of WRF is higher in March and April but lower in October to December. HSS is as high as above 0.25 and HR (FAR) is as high (low) as above (below) 0.35 in 2-month lead time. According to the spatial distribution of HSS, predictability is not concentrated in a specific region but homogeneously spread throughout the whole region of South Korea.

• Atmosphere
• /
• v.26 no.1
• /
• pp.35-45
• /
• 2016

Initialization of the global seasonal forecast system is as much important as the quality of the embedded climate model for the climate prediction in sub-seasonal time scale. Recent studies have emphasized the important role of soil moisture initialization, suggesting a significant increase in the prediction skill particularly in the mid-latitude land area where the influence of sea surface temperature in the tropics is less crucial and the potential predictability is supplemented by land-atmosphere interaction. This study developed a new soil moisture initialization method applicable to the KMA operational seasonal forecasting system. The method includes first the long-term integration of the offline land surface model driven by observed atmospheric forcing and precipitation. This soil moisture reanalysis is given for the initial state in the ensemble seasonal forecasts through a simple anomaly initialization technique to avoid the simulation drift caused by the systematic model bias. To evaluate the impact of the soil moisture initialization, two sets of long-term, 10-member ensemble experiment runs have been conducted for 1996~2009. As a result, the soil moisture initialization improves the prediction skill of surface air temperature significantly at the zero to one month forecast lead (up to ~60 days forecast lead), although the skill increase in precipitation is less significant. This study suggests that improvements of the prediction in the sub-seasonal timescale require the improvement in the quality of initial data as well as the adequate treatment of the model systematic bias.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.3
• /
• pp.20-29
• /
• 2021

In the Industry 4.0 era, artificial intelligence has attracted considerable interest for learning mass data to improve the accuracy of forecasting and classification. On the other hand, the current method of detecting anomalies relies on traditional statistical methods for a limited amount of data, making it difficult to detect accurate anomalies. Therefore, this paper proposes an artificial intelligence-based anomaly detection methodology to improve the prediction accuracy and identify new data patterns. In particular, data were collected and analyzed from the point of view that sensor data collected at vehicle idle could be used to detect abnormalities. To this end, a sensor was designed to determine the appropriate time length of the data entered into the forecast model, compare the results of idling data with the overall driving data utilization, and make optimal predictions through a combination of various sensor data. In addition, the predictive accuracy of artificial intelligence techniques was presented by comparing Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) as the predictive methodologies. According to the analysis, using idle data, using 1.5 times of the data for the idling periods, and using CNN over LSTM showed better prediction results.

• Atmosphere
• /
• v.28 no.2
• /
• pp.123-139
• /
• 2018

The prediction skills of stratospheric sudden warming (SSW) events and its impacts on the tropospheric prediction skills in global seasonal forecasting system version 5 (GloSea5), an operating subseasonal-to-seasonal (S2S) model in Korea Meteorological Administration, are examined. The model successfully predicted SSW events with the maximum lead time of 11.8 and 13.2 days in terms of anomaly correlation coefficient (ACC) and mean squared skill score (MSSS), respectively. The prediction skills are mainly determined by phase error of zonal wave-number 1 with a minor contribution of zonal wavenumber 2 error. It is also found that an enhanced prediction of SSW events tends to increase the tropospheric prediction skills. This result suggests that well-resolved stratospheric processes in GloSea5 can improve S2S prediction in the troposphere.

• Nuclear Engineering and Technology
• /
• v.56 no.4
• /
• pp.1501-1512
• /
• 2024

Plasma disruption in tokamak experiments is a challenging issue that causes damage to the device. Reliable prediction methods are needed, but the lack of full understanding of plasma disruption limits the effectiveness of physics-driven methods. Data-driven methods based on supervised learning are commonly used, and they rely on labelled training data. However, manual labelling of disruption precursors is a time-consuming and challenging task, as some precursors are difficult to accurately identify. The mainstream labelling methods assume that the precursor onset occurs at a fixed time before disruption, which leads to mislabeled samples and suboptimal prediction performance. In this paper, we present disruption prediction methods based on anomaly detection to address these issues, demonstrating good prediction performance on J-TEXT and EAST. By evaluating precursor onset times using different anomaly detection algorithms, it is found that labelling methods can be improved since the onset times of different shots are not necessarily the same. The study optimizes precursor labelling using the onset times inferred by the anomaly detection predictor and test the optimized labels on supervised learning disruption predictors. The results on J-TEXT and EAST show that the models trained on the optimized labels outperform those trained on fixed onset time labels.

• Journal of the Korean earth science society
• /
• v.39 no.3
• /
• pp.211-227
• /
• 2018

This study evaluates the prediction skills of stratospheric polar vortex intensification events (VIEs) in Global Seasonal Forecasting System (GloSea5) model, an operational subseasonal-to-seasonal (S2S) prediction model of Korea Meteorological Administration (KMA). The results show that the prediction limits of VIEs, diagnosed with anomaly correlation coefficient (ACC) and mean squared skill score (MSSS), are 13.6 days and 18.5 days, respectively. These prediction limits are mainly determined by the eddy error, especially the large-scale eddy phase error from the eddies with the zonal wavenumber 1. This might imply that better prediction skills for VIEs can be obtained by improving the model performance in simulating the phase of planetary scale eddy. The stratospheric prediction skills, on the other hand, tend to not affect the tropospheric prediction skills in the analyzed cases. This result may indicate that stratosphere-troposphere dynamic coupling associated with VIEs might not be well predicted by GloSea5 model. However, it is possible that the coupling process, even if well predicted by the model, cannot be recognized by monotonic analyses, because intrinsic modes in the troposphere often have larger variability compared to the stratospheric impact.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.3
• /
• pp.1119-1143
• /
• 2014

Lightweight trust mechanism with lightweight cryptographic primitives has emerged as an important mechanism in resource constraint wireless sensor based mobile devices. In this work, outlier detection in lightweight Mobile Ad-hoc NETworks (MANETs) is extended to create the space of reliable trust cycle with anomaly detection mechanism and minimum energy losses [1]. Further, system is tested against outliers through detection ratios and anomaly scores before incorporating virtual programmable nodes to increase the efficiency. Security in proposed system is verified through ProVerif automated toolkit and mathematical analysis shows that it is strong against bad mouthing and on-off attacks. Performance of proposed technique is analyzed over different MANET routing protocols with variations in number of nodes and it is observed that system provide good amount of throughput with maximum of 20% increase in delay on increase of maximum of 100 nodes. System is reflecting good amount of scalability, optimization of resources and security. Lightweight modeling and policy analysis with lightweight cryptographic primitives shows that the intruders can be detection in few milliseconds without any conflicts in access rights.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.1
• /
• pp.79.3-80
• /
• 2016

We have investigated the correlation analysis between global temperature anomaly and two main factors: geomagnetic activity (aa index) of Earth external factor and CO2 of Earth internal factor. For this, we used NOAA Global Surface Temperature anomaly (Ta) data from 1868 to 2015. The aa index indicates the geomagnetic activity measured at two anti-podal subauroral stations (Canberra Australia and Hartland England) and the CO2 data come from historical ice core records and NOAA/ESRL data. From the comparison between (Ta) and aa index, we found several interesting things, First, the linear correlation coefficient between two parameters increases until 1985 and then decreases rapidly. Second, the scattered plot between two parameters shows a boundary of the correlation tendency (positive and negative correlation) near 1985. A partial correlation of (Ta) and two main factors (aa index, CO2) also shows that the geomagnetic effect (aa index) is dominant until about 1985 and the CO2 effect becomes much more important after then. These results indicate that the CO2 effect become very an important factor since at least 1985. For a further analysis, we simply assume that Ta = Ta(aa)+Ta(CO2) and made a linear regression between (Ta) and aa index from 1868 to 2015. A linear model is then made from the linear regression between energy consumption (a proxy of CO2 effect) and Ta-Ta(aa) since 1985. Our results will be discussed in view of the prediction of global warming.