• Nuclear Engineering and Technology
• /
• v.53 no.12
• /
• pp.4072-4079
• /
• 2021

This article presents a study on the state-of-the-art methods for automated radioactive material detection and identification, using gamma-ray spectra and modern machine learning methods. The recent developments inspired this in deep learning algorithms, and the proposed method provided better performance than the current state-of-the-art models. Machine learning models such as: fully connected, recurrent, convolutional, and gradient boosted decision trees, are applied under a wide variety of testing conditions, and their advantage and disadvantage are discussed. Furthermore, a hybrid model is developed by combining the fully-connected and convolutional neural network, which shows the best performance among the different machine learning models. These improvements are represented by the model's test performance metric (i.e., F1 score) of 93.33% with an improvement of 2%-12% than the state-of-the-art model at various conditions. The experimental results show that fusion of classical neural networks and modern deep learning architecture is a suitable choice for interpreting gamma spectra data where real-time and remote detection is necessary.

• Smart Structures and Systems
• /
• v.23 no.6
• /
• pp.669-682
• /
• 2019

Vibration mitigation of cables or hangers is one of the crucial problems for cable supported bridges. Previous research focused on the behaviors of cable with dampers or crossties, which could help engineering community apply these mitigation devices more efficiently. However, less studies are available for hybrid applied cross-ties and dampers, especially lack of both analytical and experimental verifications. This paper studied damping and frequency of two parallel identical cables with a connection cross-tie and an attached damper. The characteristic equation of system was derived based on transfer matrix method. The complex characteristic equation was numerically solved to find the solutions. Effects of non-dimensional spring stiffness and location on the maximum cable damping, the corresponding optimum damper constant and the corresponding frequency of lower vibration mode were further addressed. System with twin small-scale cables with a cross-link and a viscous damper were tested. The damping and frequency from the test were very close to the analytical ones. The two branches of solutions: in-phase modes and the out-of-phase modes, were identified; and the two branches of solutions were different for damping and frequency behaviors.

• ETRI Journal
• /
• v.43 no.1
• /
• pp.40-52
• /
• 2021

DNS (Domain Name System) tunnels almost obscure the true network activities of users, which makes it challenging for the gateway or censorship equipment to identify malicious or unpermitted network behaviors. An efficient way to address this problem is to conduct a temporal-spatial analysis on the tunnel traffic. Nevertheless, current studies on this topic limit the DNS tunnel to those with a single protocol, whereas more than one protocol may be used simultaneously. In this paper, we concentrate on the refined identification of two protocols mixed in a DNS tunnel. A feature set is first derived from DNS query and response flows, which is incorporated with deep neural networks to construct a regression model. We benchmark the proposed method with captured DNS tunnel traffic, the experimental results show that the proposed scheme can achieve identification accuracy of more than 90%. To the best of our knowledge, the proposed scheme is the first to estimate the ratios of two mixed protocols in DNS tunnels.

• Journal of Information Technology Applications and Management
• /
• v.29 no.3
• /
• pp.43-55
• /
• 2022

In this study, with the goal of developing a deep learning-based product recommendation model for effective matching of influencers and products, a deep learning model with a collaborative filtering model combined with generalized matrix decomposition(GMF), a collaborative filtering model based on multi-layer perceptron (MLP), and neural collaborative filtering and generalized matrix Factorization (NeuMF), a hybrid model combining GMP and MLP was developed and tested. In particular, we utilize one-class problem free boosting (OCF-B) method to solve the one-class problem that occurs when training is performed only on positive cases using implicit feedback in the deep learning-based collaborative filtering recommendation model. In relation to model selection based on overall experimental results, the MLP model showed highest performance with weighted average precision, weighted average recall, and f1 score were 0.85 in the model (n=3,000, term=15). This study is meaningful in practice as it attempted to commercialize a deep learning-based recommendation system where influencer's promotion data is being accumulated, pactical personalized recommendation service is not yet commercially applied yet.

• International Journal of High-Rise Buildings
• /
• v.11 no.4
• /
• pp.287-300
• /
• 2022

Unconventional structures are getting more popular in recent days. Large-span roofs are used for many structures, such as airports, stadiums, and conventional halls. Identifying the pressure distribution and wind load acting on those structures is essential. This paper offers a collaborative study of computational fluid dynamics (CFD) simulations and wind tunnel tests for assessing wind pressure distribution for a building with a combined slender curved roof. The hybrid turbulence model, Improved Delayed Detached Eddy Simulation (IDDES), simulates the open terrain turbulent flow field. The wind-induced local pressure coefficients on complex roof structures and the turbulent flow field around the structure were thus calculated based upon open terrain wind flow simulated with the FLUENT software. Local pressure measurements were investigated in a boundary layer wind tunnel simultaneous to the simulation to determine the pressure coefficient distributions. The results predicted by CFD were found to be consistent with the wind tunnel test results. The comparative study validated that the recommended IDDES model and the vortex method associated with CFD simulation are suitable tools for structural engineers to evaluate wind effects on long-span complex roofs and plan irregular buildings during the design stage.

• Journal of Sensor Science and Technology
• /
• v.32 no.2
• /
• pp.124-130
• /
• 2023

Implementing flexible production systems in domestic and foreign automotive design parts assembly has increased demand for automation and power reduction. Consequently, transition to a hybrid production method is observed where multiple vehicles are assembled in a single assembly line. Multiple robots, 3D vision sensors, mounting positions, and correction software have complex configurations in the automotive glass mounting system. Hence, automation is required owing to significant difficulty in the assembly process of automobile parts. This study presents a primer lighting and inspection algorithm that is robust to the assembly environment of real automotive design parts using high power 'ㄷ'-shaped LED inclined lighting. Furthermore, a 2D camera was developed in the primer coating inspection system-the core technology of the glass mounting system. A primer application demo line applicable to the actual automobile production line was established using the proposed high power lighting and algorithm. Furthermore, application inspection performance was verified using this demo system. Experimental results verified that the performance of the proposed system exceeded the level required to satisfy the automobile requirements.

• Bulletin of the Korean Chemical Society
• /
• v.33 no.10
• /
• pp.3321-3330
• /
• 2012

The complexation characteristics of mixed [2 + 2'] calix[4]aryl derivatives (3 and 4) with alkali metal cations were investigated by the mPW1PW91 (hybrid HF-DF) calculation method. The total electronic and Gibbs free energies of the various complexes (cone, partial-cone, 1,2-alternate, and 1,3-alternate) of sodium and potassium cations with 3 and 4 were analyzed and compared. The structures of the endo- or exo-complexes of the alkali metal cation with the host 3 were optimized using the mPW1PW91/6-31G(d) method, followed by mPW1PW91/6-311+G(d) calculations. The structures of the endo- or exo-complexes of the alkali metal cation with the host 4 were optimized using the mPW1PW91/6-31G(d,p) method. The mPW1PW91 calculated relative energies of the various conformations of the free hosts (3 and 4) suggest that the cone conformers of 3 and 4 are the most stable. The mPW1PW91calculations also suggest that the complexation efficiencies of the sodium ion with hosts 3 and 4 are about 24 and 27 kcal/mol better than those of the potassium ion, respectively. These trends are in good agreement with the experimental results. The exo-complexation efficiencies of the sodium ion toward the conformers of hosts 3 and 4 are roughly 14 and 17 kcal/mol better than those for the endo-$Na^+$-complexes of 3 and 4, respectively. The exo-complexation of the cone isomer of 3 with cation could be confirmed by the differences of the diagnostic C=O bands in the free host and its complex's IR spectra.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.18 no.3
• /
• pp.217-222
• /
• 1998

This experiment was carried out to investigate effects of herbicide application method on Calysteria jcqoonica control and corn yield at the forage experimental field, grassland and forage crops division, National Livestock Research Institute, RDA, Suwon from 1996 to 1997. The experiment was arranged in a randomized block design with three replication. The herbicide application consisted of control, 2~3 leaf stage(Dicamba), 5~6 leaf stage(Dicamba) and soil treatment (Pendimethalin). The hybrid of corn used in this experiment was P 3352. The results obtained are summarized as follow; 1. The plant height was the lowest at control and the highest at Dicamba treatment in 5~6 leaf stage, but no significant difference was found among herbicide application method at ear height. Tasseling and silking date were delayed 2~3 day at control. The length and weight of ear were highest at Dicamba treatment in 2~3 leaf stage. 2. The herbicide injury of corn was very slightly after 10-20 day, but corn was recovered soon. The weed control value was 94.2~67.5% at herbicide application plot. 3. The fresh and dry matter yield of Dicamba treatment in 2-3 leaf stage was highest as 42,878 and 16,033kg/ha, respectively. The application of Dicamba increased the DM yield of corn by 13~30%. 4. Herbicide application time had little effect on the ADF, NDF and CP contents, but the forage quality of ear was higher than that of stover. The result of this experiment indicate that the application of Dicamba in 2~6 leaf stage of corn will remove almost all of Calysreria jqonica and increase DM yield of com.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37A no.11
• /
• pp.918-926
• /
• 2012

In 3D video transmission, a depth map being used for depth image based rendering (DIBR) is generally compressed by reducing resolution for coding efficiency. Errors in resolution reduction are recovered by an appropriate up-sampling method after decoding. However, most previous works only focus on up-sampling techniques to reduce errors. In this paper, we propose a novel down-sampling technique of depth map that applies different down-sampling rates on moving objects and background in order to enhance human perceptual quality. Experimental results demonstrate that the proposed scheme provides both higher visual quality and peak signal-to-noise ratio (PSNR). Also, our method is compatible with other up-sampling techniques.

• Korea-Australia Rheology Journal
• /
• v.14 no.3
• /
• pp.107-114
• /
• 2002

Residual stress distribution in injection molded short fiber composites is determined by using the layer-removal method. Polystyrene is mixed with carbon fibers of 3% volume fraction (4.5% weight fraction) in an extruder and the tensile specimen is injection-molded. The layer-removal process, in which removing successive thin uniform layers of the material from the surface of the specimen by a milling machine, is employed and the resulting curvature is acquired by means of an image processing. The isotropic elastic analysis proposed by Treuting and Read which assumes a constant Yaung’s modulus in the thickness direction is one of the most frequently used methods to determine residual stresses. However, injection molded short fiber composites experience complex fiber orientation during molding and variation of Yaung’s modulus distribution occurs in the specimen. In this study, variation of Yaung’s modulus with respect to the thickness direction is considered for calculation of the residual stresses as proposed by White and the result is compared with that by assuming constant modulus. Residual stress distribution obtained from this study shows a typical stress profile of injection-molded products as reported in many literatures. Young’s modulus distribution is predicted by using numerical methods instead of experimental results. For the numerical analysis of injection molding process, a hybrid FEM/FDM method is used in order to predict velocity, temperature field, fiber orientation, and resulting mechanical properties of the specimen at the end of molding.