• International Journal of Computer Science & Network Security
• /
• v.24 no.1
• /
• pp.119-126
• /
• 2024

Disease caused by the coronavirus (COVID-19) is sweeping the globe. There are numerous methods for identifying this disease using a chest imaging. Computerized Tomography (CT) chest scans are used in this study to detect COVID-19 disease using a pretrain Convolutional Neural Network (CNN) ResNet50. This model is based on image dataset taken from two hospitals and used to identify Covid-19 illnesses. The pre-train CNN (ResNet50) architecture was used for feature extraction, and then fully connected layers were used for classification, yielding 97%, 96%, 96%, 96% for accuracy, precision, recall, and F1-score, respectively. When combining the feature extraction techniques with the Back Propagation Neural Network (BPNN), it produced accuracy, precision, recall, and F1-scores of 92.5%, 83%, 92%, and 87.3%. In our suggested approach, we use a preprocessing phase to improve accuracy. The image was enhanced using the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm, which was followed by cropping the image before feature extraction with ResNet50. Finally, a fully connected layer was added for classification, with results of 99.1%, 98.7%, 99%, 98.8% in terms of accuracy, precision, recall, and F1-score.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.7
• /
• pp.191-198
• /
• 2021

Since the size of training dataset become large and the model is getting deeper to achieve high accuracy in deep learning, the deep neural network training requires a lot of computation and it takes too much time with a single node. Therefore, distributed deep learning is proposed to reduce the training time by distributing computation across multiple nodes. In this study, we propose hybrid allreduce strategy that considers the characteristics of each layer and communication and computational overlapping technique for synchronization of distributed deep learning. Since the convolution layer has fewer parameters than the fully-connected layer as well as it is located at the upper, only short overlapping time is allowed. Thus, butterfly allreduce is used to synchronize the convolution layer. On the other hand, fully-connecter layer is synchronized using ring all-reduce. The empirical experiment results on PyTorch with our proposed scheme shows that the proposed method reduced the training time by up to 33% compared to the baseline PyTorch.

• Steel and Composite Structures
• /
• v.25 no.4
• /
• pp.505-521
• /
• 2017

The steel plate shear wall with beam-only connected infill plate (SSW-BO) is an innovative lateral load resisting system consisting of infill plates connected to surrounding beams and separated from the main columns. In this research, the effects of perforation diameter as well as slenderness ratios of infill plates on the hysteresis behavior of SSW-BO systems were studied experimentally. Experimental testing is performed on eight one-sixth scaled one-story SSW-BO specimens with two plate thicknesses and four different circular opening ratios at the center of the panels under fully reversed cyclic quasi-static loading in compliance with the SAC test protocol. Strength, stiffness, ductility and energy absorption were evaluated based on the hysteresis loops. It is found that the systems exhibited stable hysteretic behavior during testing until significant damage in the connection of infill plates to surrounding beams at large drifts. It is also seen that pinching occurred in the hysteresis loops, since the hinge type connections were used as boundaries at four corners of surrounding frames. The strength and initial stiffness degradation of the perforated specimens containing opening ratio of 0.36 compared to the solid one is in the range of 20% to 30% and 40% to 50%, respectively.

• Ocean Systems Engineering
• /
• v.7 no.3
• /
• pp.195-223
• /
• 2017

This paper presents 3D numerical simulations of a Free Standing Hybrid Riser under Vortex Induced Vibration, with prescribed motion on the top to replace the motion of the buoyancy can. The model is calculated using a fully implicit discretization scheme. The flow field around the riser is computed by solving the Navier-Stokes equations numerically. The fluid domain is discretized using the overset grid approach. Grid points in near-wall regions of riser are of high resolution, while far field flow is in relatively coarse grid. Fluid-structure interaction is accomplished by communication between fluid solver and riser motion solver. Simulation is based on previous experimental data. Two cases are studied with different current speeds, where the motion of the buoyancy can is approximated to a 'banana' shape. A fully three-dimensional CFD approach for VIV simulation for a top side moving Riser has been presented. This paper also presents a simulation of a riser connected to a platform under harmonic regular waves.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.1304-1307
• /
• 2002

In this paper, we describe the structure and the test results of a Bluetooth baseband module we have developed. The module has a distributed buffer, i.e. FIFO, for data stream. Bus interface of the module is designed on the basis of interface of microprocessor widely used and the external interface is designed to consider chips connected directly. Since the module performs as many hardware efficient tasks as possible, processing load of microprocessor is very small. It can also be controlled either by software or by hardware for flexibility. The fully synthesizable baseband module was fabricated in a $0.25\mu\textrm{m}$ CMOS technology occupying $2.79\times2.8{\textrm{mm}^2}$ area. And an FPGA implementation of this module is tested for file and bit-stream transfers between PCs.

• Electronics and Telecommunications Trends
• /
• v.36 no.1
• /
• pp.22-31
• /
• 2021

In this thesis, we examine the development status of autonomous mobility services using various artificial intelligence algorithms and propose a solution by combining edge and cloud computing to overcome technical difficulties. A fully autonomous vehicle with enhanced safety and ethics can be implemented using the proposed solution. In addition, for the future of 2035, we present a new concept that enables two- and three-dimensional movement via cooperation between ecofriendly, low-noise, and modular fully autonomous vehicles. The zero-error autonomous driving system will safely and conveniently transport people, goods, and services without time and space constraints and contribute to the autonomous mobility services that are free from movement in connection with various mobility.

• Journal of the Korean Geotechnical Society
• /
• v.38 no.3
• /
• pp.35-42
• /
• 2022

The N-value from the Standard Penetration Test (SPT), which is one of the representative in-situ test, is an important index that provides basic geological information and the depth of the bearing layer for the design of geotechnical structures. In the aspect of time and cost-effectiveness, there is a need to carry out a representative sampling test. However, the various variability and uncertainty are existing in the soil layer, so it is difficult to grasp the characteristics of the entire field from the limited test results. Thus the spatial interpolation techniques such as Kriging and IDW (inverse distance weighted) have been used for predicting unknown point from existing data. Recently, in order to increase the accuracy of interpolation results, studies that combine the geotechnics and deep learning method have been conducted. In this study, based on the SPT results of about 22,000 holes of ground survey, a comparative study was conducted to predict the depth of the bearing layer using deep learning methods and IDW. The average error among the prediction results of the bearing layer of each analysis model was 3.01 m for IDW, 3.22 m and 2.46 m for fully connected network and PointNet, respectively. The standard deviation was 3.99 for IDW, 3.95 and 3.54 for fully connected network and PointNet. As a result, the point net deep learing algorithm showed improved results compared to IDW and other deep learning method.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.4 no.5
• /
• pp.413-422
• /
• 1999

Auto-connected multipulse(12/24pulse) rectifier schemes are cost effective methods for reducing line current hamonics in PWM drive systems. Employing these schemes to enhance utility power quality requires careful attention to several design considerations In particular, excursion of dc-link voltage at no load, effect of pre-existing voltage distortion, impedance mismatches, unequal diode drops on rectifier current sharing and performance, are fully analyzed, Several corrective measures to improve the performance of 12/24­pulse rectifier systems are also discussed. Finally, experimental results on a 460V, 60Hz 400kVA commercial ASD, retrofitted with 12/24pulse rectifier systems are discussed in detail.

• Proceedings of the KIEE Conference
• /
• 1991.11a
• /
• pp.189-192
• /
• 1991

This paper presents a modular structure design of analog Hopfield neural network. Each multiplier consists of four MOS transistors which are connected to an op-amp at the front end of a neuron. A pair of MOS transistor is used in order to maintain linear operation of the synapse and can produce positive or negative synaptic weight. This architecture can be expandable to any size neural network by forming tree structure. By altering the connections, other nework paradigms can also be implemented using this basic modules. The stength of this approach is the expandability and the general applicability. The layout design of a four-neuron fully connected feedback neural network is presented and is simulated using SPICE. The network shows correct retrival of distorted patterns.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.20 no.2
• /
• pp.152-159
• /
• 2015

In a weak grid condition, the precise grid impedance estimation is essential to guaranteeing the high performance current control and power transfer for a grid-connected inverter. This study proposes a precise estimation method for grid impedance by PQ variations by employing the variation method of reference currents. The operation principle of grid impedance estimation is fully presented, and the negative impact of the phase locked loop is analyzed. Estimation error by a synchronization angle in the park's transformation using the phase locked loop is derived. As a result, the variation method of reference currents for accurate estimation is introduced. The validation of the proposed method is verified through several simulation results and experiments based on a 2-kW voltage source inverter prototype.