• Nuclear Engineering and Technology
• /
• v.55 no.1
• /
• pp.285-294
• /
• 2023

Autonomous radiation source detection has long been studied for radiation emergencies. Compared to conventional data-driven or path planning methods, deep reinforcement learning shows a strong capacity in source detection while still lacking the generalized ability to the geometry in unknown environments. In this work, the detection task is decomposed into two subtasks: exploration and localization. A hierarchical control policy (HC) is proposed to perform the subtasks at different stages. The low-level controller learns how to execute the individual subtasks by deep reinforcement learning, and the high-level controller determines which subtasks should be executed at the current stage. In experimental tests under different geometrical conditions, HC achieves the best performance among the autonomous decision policies. The robustness and generalized ability of the hierarchy have been demonstrated.

• Journal of Drive and Control
• /
• v.20 no.4
• /
• pp.54-63
• /
• 2023

High and intermittent loading cycles induce fatigue damage to transmission components, resulting in premature gearbox failure. To identify gearbox defects, numerous vibration-based diagnostics techniques, using several artificial intelligence (AI) algorithms, have recently been presented. In this paper, an optimized deep belief network (DBN) model for gearbox problem diagnosis was designed based on time-frequency visual pattern identification. To optimize the hyperparameters of the model, a particle swarm optimization (PSO) approach was integrated into the DBN. The proposed model was tested on two gearbox datasets: a wind turbine gearbox and an experimental gearbox. The optimized DBN model demonstrated strong and robust performance in classification accuracy. In addition, the accuracy of the generated datasets was compared using traditional ML and DL algorithms. Furthermore, the proposed model was evaluated on different partitions of the dataset. The results showed that, even with a small amount of sample data, the optimized DBN model achieved high accuracy in diagnosis.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.3
• /
• pp.387-400
• /
• 2021

The increase in number of streaming platforms and contents thereof, owing to an advancement of cloud environment, has triggered the rapid proliferation of illegally replicated contents as well as legal contents. This necessitates the development of technology capable of discriminating the copyright infringement of various contents. The Korea Copyright Protection Agency operates a video content demonstration system using AI, but it has limitations on distortions such as resolution changes. In this paper, we propose the powerful mechanism using skeleton, which is resistant against distorted video contents and capable of discriminating copyright infringement of platforms streaming illegal video contents. The proposed mechanism exploits the calculation of Hamming distance to the original video by converting collected data into binary ones for the efficient calculation. As a result of the experiment, the proposed mechanism have demonstrated the discrimination of illegally replicated video contents with an accuracy of 94.79% and average magnitude of 215KB.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.8
• /
• pp.1121-1129
• /
• 2022

Object recognition is carried out using RGB images in various object recognition studies. However, RGB images in dark illumination environments or environments where target objects are occluded other objects cause poor object recognition performance. On the other hand, IR images provide strong object recognition performance in these environments because it detects infrared waves rather than visible illumination. In this paper, we propose an RGB-IR fusion model, high-frequency interchange network (HINet), which improves object recognition performance by combining only the strengths of RGB-IR image pairs. HINet connected two object detection models using a mutual high-frequency transfer (MHT) to interchange advantages between RGB-IR images. MHT converts each pair of RGB-IR images into a discrete cosine transform (DCT) spectrum domain to extract high-frequency information. The extracted high-frequency information is transmitted to each other's networks and utilized to improve object recognition performance. Experimental results show the superiority of the proposed network and present performance improvement of the multispectral object recognition task.

• Journal of Internet of Things and Convergence
• /
• v.8 no.1
• /
• pp.79-85
• /
• 2022

Internet of Things (IoT) connects devices with various platforms, computing power, and functions. Due to the diversity of networks and the ubiquity of IoT devices, demands for security and privacy are increasing. Therefore, cryptographic mechanisms must be strong enough to meet these increased requirements, while at the same time effective enough to be implemented in devices with long-range specifications. In this paper, we present the performance and memory limitations of modern cryptographic primitives and schemes for different types of devices that can be used in IoT. In addition, detailed performance evaluation of the performance of the most commonly used encryption algorithms in low-spec devices frequently used in IoT networks is performed. To provide data protection, the binary ring uses encryption asymmetric fully homomorphic encryption and symmetric encryption AES 128-bit. As a result of the experiment, it can be seen that the IoT device had sufficient performance to implement a symmetric encryption, but the performance deteriorated in the asymmetric encryption implementation.

• Convergence Security Journal
• /
• v.23 no.4
• /
• pp.81-92
• /
• 2023

In recent years, the ROK military is promoting Defense Innovation 4.0 with the goal of fostering strong military based on science and technology equipped with artificial intelligence(AI) to prepare for the upcoming population cliff. In particular, at the present time of increased threats of North Korea, the South Korean military is seeking to deal with a decrease in military service resources through the introduction of a Scientific Guard System using advanced technology. TICN which is a core basic communication system to ensure the integrated combat capability of the ROK military is, however, limited to use as a based network for the emerging Scientific Guard System due to the narrow transmission bandwidth with widely spread poor reception area. To deal with this problem, this paper proposes TVWS-based Scientific Guard Systems with TVWS-based wireless network construction technology that has been available for free in Korea since 2017. The TVWS-based Scientific Guard System proposed in this paper, when compared to the existing wired network-based Scientific Guard Systems, has various advantages in terms of minimizing operational gaps, reducing construction costs, and flexibility in installation and operation.

• International Journal of Computer Science & Network Security
• /
• v.24 no.9
• /
• pp.30-40
• /
• 2024

Building an accurate 3-D spatial road network model has become an active area of research now-a-days that profess to be a new paradigm in developing Smart roads and intelligent transportation system (ITS) which will help the public and private road impresario for better road mobility and eco-routing so that better road traffic, less carbon emission and road safety may be ensured. Dealing with such a large scale 3-D road network data poses challenges in getting accurate elevation information of a road network to better estimate the CO₂ emission and accurate routing for the vehicles in Internet of Vehicle (IoV) scenario. Clustering and regression techniques are found suitable in discovering the missing elevation information in 3-D spatial road network dataset for some points in the road network which is envisaged of helping the public a better eco-routing experience. Further, recently Explainable Artificial Intelligence (xAI) draws attention of the researchers to better interprete, transparent and comprehensible, thus enabling to design efficient choice based models choices depending upon users requirements. The 3-D road network dataset, comprising of spatial attributes (longitude, latitude, altitude) of North Jutland, Denmark, collected from publicly available UCI repositories is preprocessed through feature engineering and scaling to ensure optimal accuracy for clustering and regression tasks. K-Means clustering and regression using Support Vector Machine (SVM) with radial basis function (RBF) kernel are employed for 3-D road network analysis. Silhouette scores and number of clusters are chosen for measuring cluster quality whereas error metric such as MAE ( Mean Absolute Error) and RMSE (Root Mean Square Error) are considered for evaluating the regression method. To have better interpretability of the Clustering and regression models, SHAP (Shapley Additive Explanations), a powerful xAI technique is employed in this research. From extensive experiments , it is observed that SHAP analysis validated the importance of latitude and altitude in predicting longitude, particularly in the four-cluster setup, providing critical insights into model behavior and feature contributions SHAP analysis validated the importance of latitude and altitude in predicting longitude, particularly in the four-cluster setup, providing critical insights into model behavior and feature contributions with an accuracy of 97.22% and strong performance metrics across all classes having MAE of 0.0346, and MSE of 0.0018. On the other hand, the ten-cluster setup, while faster in SHAP analysis, presented challenges in interpretability due to increased clustering complexity. Hence, K-Means clustering with K=4 and SVM hybrid models demonstrated superior performance and interpretability, highlighting the importance of careful cluster selection to balance model complexity and predictive accuracy.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 1994.11a
• /
• pp.1-1
• /
• 1994

;The electronic state calculations for various types of ceramic materials have beell performed by the use of $DV-X\;{\alpha}$ cluster method. The molecular orbital levels and wave functions for model clusters have been computed to study the electronic properties ami chemical bonding of the ceramics. For ${\beta}-sialon(Si_{6-z}Al_zO_zN_{8-z})$ which is a high temperature structural material based on ${\beta}-Si_3N_4$, we have made model cluster calculations to estimate the strength of chemical bonding between atoms by the Mulliken population analysis. It is found that the covalent bonding between Si and N atoms is very strong in pure ${\beta}-Si_3N_4$, but the covalency around solute atom is considerably weakened when Si atom is substituted by AI. This tendency is enhanced by an additional substitution of oxygen atom for N. The result calculated can well explain the experimental data of changes in mechanical properties such as the reductions of Young's modulus and Vickers hardness with increment of z-value in ${\beta}-sialon$. Various model clusters for transition metal oxides which show many interesting physical and chemical properties have also been calculated. High-valent perovskite-type iron oxides EMFe0_3E(M=Ca and Sr) possess very interesting magnetic and chemical properties. In these oxides, iron exists as $Fe^{4+}$ state, but the experimental measurement of Mossba~er effect suggests that disproportionation $2Fe^{4+}=Fe^{3+}+Fe^{5+}$ takes place for $CaFe0_3$ at low temperatures. The model cluster calculations for these compounds indicated the existence of considerably strong covalent bonding of Fe-O. The calculations of hyperfine interaction at iron neucleus show very good agreement with the experimental Mossbauer measurements. The result calculated also implies that the disproportionation reaction is strongly possible by assuming the quenching of breathing phonon mode at low temperatures.tures.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.538-541
• /
• 2005

Tidal flats develop along the south coast ofthe Korean peninsula. These areas are famous for sea farming. Specially, strong and coherent radar backscattering signals are observed over oyster sea farms that consist of artificial structures. Tide height in oyster farm is possible to measure by using interferometric phase and intensity of SAR data. It is assumed that the radar signals from oyster farm could be considered as double-bouncing returns by vertical and horizontal bars. But, detailed backscattering mechanism and polarimetric characteristics in oyster farm had not been well studied. We could not demonstrate whether the assumption is correct or not and exactly understand what the properties of back scattering were in oyster farm without full polarimetric data. The results of AIRSAR L-band POLSAR data, experiments in laboratory and JERS-l images are discussed. We carried out an experiment simulating a target structure using vector network analyser (Y.N.A.) in an anechoic chamber at Niigata University. Radar returns from vertical poles are stronger than those from horizontal poles by 10.5 dB. Single bounce components were as strong as double bounce components and more sensitive to antenna look direction. Double bounce components show quasi-linear relation with height of vertical poles. As black absorber replaced AI-plate in bottom surface, double bounce in vertical pole decreased. It is observed that not all oyster farms are characterized by double bounced scattering in AIRSAR data. The image intensity of the double bounce dominant oyster farm was investigated with respect to that of oyster farm dominated by single bounce in JERS-l SAR data. The image intensity model results in a correlation coefficient (R2 ) of 0.78 in double bounce dominant area while that of 0.54 in single bouncing dominant area. This shows that double bounce dominant area should be selected for water height measurement using In8AR technique.

• Journal of Fashion Business
• /
• v.19 no.1
• /
• pp.95-105
• /
• 2015

This study investigated the color culture in the Japanese Medieval Age. The Japanese Medieval Age included the Kamakura period (1180-1333) and Muromachi period (1336-1573), and the leading group transitioned from the Kuge families to the Buke families. The taboos about colors from ancient times became nominal, and forbidden colors, such as purple, celadon, and red, became the colors of the samurai, leading to beautiful soldier gears that were unparalleled in history. In the Kamakura period, colors that conveyed a strong impression were created and preferred with the combination of a samurai's reasonable spirit and zen thoughts. The period was also called "the era of hari", and cross dyeing based on basic colors such as suou (red), ai (blue), and kuchinasi (yellow) was popular. In both the Kamakura and Muromachi periods, conspicuous and strong colors were sought for costumes, and embroidery was used with gold leaf, silver leaf, gold threads, silver threads, and background color. The colors of costume preferred by Buke men in the period included green, blue, and brown. In the characteristics of the kosode, the sugan and hitadare were used for men's formal dress, while kosode was used for the grooming of the working class. In these periods, additionally, the working class began to be socially engaged in actively wearing the one-layer kosode, which became popular, and the characteristics of the Japanese Medieval Age, during which functionality and practicality was valued, were also reflected in the dressing.