• Imaging Science in Dentistry
• /
• v.52 no.3
• /
• pp.239-244
• /
• 2022

Purpose: Despite the proliferation of numerous morphometric and anthropometric methods for sex identification based on linear, angular, and regional measurements of various parts of the body, these methods are subject to error due to the observer's knowledge and expertise. This study aimed to explore the possibility of automated sex determination using convolutional neural networks(CNNs) based on lateral cephalometric radiographs. Materials and Methods: Lateral cephalometric radiographs of 1,476 Iranian subjects (794 women and 682 men) from 18 to 49 years of age were included. Lateral cephalometric radiographs were considered as a network input and output layer including 2 classes(male and female). Eighty percent of the data was used as a training set and the rest as a test set. Hyperparameter tuning of each network was done after preprocessing and data augmentation steps. The predictive performance of different architectures (DenseNet, ResNet, and VGG) was evaluated based on their accuracy in test sets. Results: The CNN based on the DenseNet121 architecture, with an overall accuracy of 90%, had the best predictive power in sex determination. The prediction accuracy of this model was almost equal for men and women. Furthermore, with all architectures, the use of transfer learning improved predictive performance. Conclusion: The results confirmed that a CNN could predict a person's sex with high accuracy. This prediction was independent of human bias because feature extraction was done automatically. However, for more accurate sex determination on a wider scale, further studies with larger sample sizes are desirable.

• The Journal of Korea Robotics Society
• /
• v.17 no.4
• /
• pp.463-476
• /
• 2022

In this paper, we propose a table tennis posture classification system using a cooperative robot to develop a table tennis robot that can be trained like a real game. The most ideal table tennis robot would be a robot with a high joint driving speed and a high degree of freedom. Therefore, in this paper, we intend to use a cooperative robot with sufficient degrees of freedom to develop a robot that can be trained like a real game. However, cooperative robots have the disadvantage of slow joint driving speed. These shortcomings are expected to be overcome through quick recognition. Therefore, in this paper, we try to quickly classify the opponent's posture to overcome the slow joint driving speed. To this end, learning about dynamic postures was conducted using image data as input, and finally, three classification models were created and comparative experiments and evaluations were performed on the designated dynamic postures. In conclusion, comparative experimental data demonstrate the highest classification accuracy and fastest classification speed in classification models using MLP (Multi-Layer Perceptron), and thus demonstrate the validity of the proposed algorithm.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.1
• /
• pp.71-76
• /
• 2023

This paper studied a method of attenuating nonlinear noise using a Wiener filter of a neural network structure in an acoustic noise attenuator. This system improves nonlinear noise attenuation performance with a deep learning algorithm using a neural network Wiener filter instead of using a conventional adaptive filter. A voice is estimated from a single input voice signal containing nonlinear noise using a 128-neuron, 8-neuron hidden layer and an error back propagation algorithm. In this study, a simulation program using the Keras library was written and a simulation was performed to verify the attenuation performance for nonlinear noise. As a result of the simulation, it can be seen that the noise attenuation performance of this system is significantly improved when the FNN filter is used instead of the Wiener filter even when nonlinear noise is included. This is because the complex structure of the FNN filter expresses any type of nonlinear characteristics well.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.27 no.5
• /
• pp.193-202
• /
• 2023

In this paper, a dynamic centrifuge model test was conducted on a 24.8-meter-deep excavation consisting of a 20 m sand layer and 4.8 m bedrock, classified as S3 by Korean seismic design code KDS 17 10 00. A braced excavation wall supports the hole. From the results, the mechanism of seismically induced earth pressure was investigated, and their distribution and loading points were analyzed. During earthquake loadings, active seismic earth pressure decreases from the at-rest earth pressure since the backfill laterally expands at the movement of the wall toward the active direction. Yet, the passive seismic earth pressure increases from the at-rest earth pressure since the backfill pushes to the wall and laterally compresses at it, moving toward a passive direction and returning to the initial position. The seismic earth pressure distribution shows a half-diamond distribution in the dense sand and a uniform distribution in loose sand. The loading point of dynamic thrust corresponding with seismic earth pressure is at the center of the soil backfill. The dynamic thrust increased differently depending on the backfill's relative density and input motion type. Still, in general, the dynamic thrust increased rapidly when the maximum horizontal displacement of the wall exceeded 0.05 H%.

• Journal of the Korea Society for Simulation
• /
• v.19 no.4
• /
• pp.43-50
• /
• 2010

Recently, the Internet telephony has been used rather than the traditional telephony by many Internet users, with low cost. Session Initiation Protocol(SIP) is the standard of application layer protocol for establishment and disconnection of the session for Internet telephony. SIP mainly runs over the UDP for transport. So in case of the loss of the INVITE request message, the message is retransmitted by an appropriate timer for reliable transmission of the UDP message. Though the retransmission is useful for ensuring the reliability of SIP messages sent by the users, it may cause the overload traffic in the SIP proxy server. The overload in SIP proxy servers results in the loss of many input messages. This paper presents a non-linear overload control algorithm to resolve the overload condition of the server. we simulate our proposed algorithm using the network simulator ns-2. The simulation results show that the throughput of the server with the proposed algorithm have been improved about 12% compared to the existing linear control algorithm.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.59-64
• /
• 2023

Fire and disaster broadcasting systems are divided into analog, digital, and network-based digital public address systems, and important specifications in network-based digital public address systems are low-latency audio, high sampling rate, and multi-channel input and output. In the past, it has been widely used to the AoE method for distinguishing based on the MAC address of the data link layer. However, this method has a problem of increasing complexity and cost. This proposal is an AoIP/UDP method, which allows communication to be easily distinguished by IP address without the need for a separate redundant network, so that the network can be freely used and configured, and cost can be reduced by reducing complexity. After implementing the AoIP/UDP method, the experimental results showed that the cost was improved with the equivalent performance with 2.66ms latency.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.60.2-60.2
• /
• 2021

In this study, we determinate the differential emission measure(DEM) of solar corona using three SDO/AIA EUV channel images and three AI-generated ones. To generate the AI-generated images, we apply a deep learning model based on multi-layer perceptrons by assuming that all pixels in solar EUV images are independent of one another. For the input data, we use three SDO/AIA EUV channels (171, 193, and 211). For the target data, we use other three SDO/AIA EUV channels (94, 131, and 335). We train the model using 358 pairs of SDO/AIA EUV images at every 00:00 UT in 2011. We use SDO/AIA pixels within 1.2 solar radii to consider not only the solar disk but also above the limb. We apply our model to several brightening patches and loops in SDO/AIA images for the determination of DEMs. Our main results from this study are as follows. First, our model successfully generates three solar EUV channel images using the other three channel images. Second, the noises in the AI-generated EUV channel images are greatly reduced compared to the original target ones. Third, the estimated DEMs using three SDO/AIA images and three AI-generated ones are similar to those using three SDO/AIA images and three stacked (50 frames) ones. These results imply that our deep learning model is able to analyze temperature response functions of SDO/AIA channel images, showing a sufficient possibility that AI-generated data can be used for multi-wavelength studies of various scientific fields. SDO: Solar Dynamics Observatory AIA: Atmospheric Imaging Assembly EUV: Extreme Ultra Violet DEM: Diffrential Emission Measure

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.131-131
• /
• 2021

Deep learning models, especially those based on long short-term memory (LSTM), have presented their superiority in addressing time series data issues recently. This study aims to comprehensively evaluate the performance of deep learning models that belong to the supervised learning category in streamflow prediction. Therefore, six deep learning models-standard LSTM, standard gated recurrent unit (GRU), stacked LSTM, bidirectional LSTM (BiLSTM), feed-forward neural network (FFNN), and convolutional neural network (CNN) models-were of interest in this study. The Red River system, one of the largest river basins in Vietnam, was adopted as a case study. In addition, deep learning models were designed to forecast flowrate for one- and two-day ahead at Son Tay hydrological station on the Red River using a series of observed flowrate data at seven hydrological stations on three major river branches of the Red River system-Thao River, Da River, and Lo River-as the input data for training, validation, and testing. The comparison results have indicated that the four LSTM-based models exhibit significantly better performance and maintain stability than the FFNN and CNN models. Moreover, LSTM-based models may reach impressive predictions even in the presence of upstream reservoirs and dams. In the case of the stacked LSTM and BiLSTM models, the complexity of these models is not accompanied by performance improvement because their respective performance is not higher than the two standard models (LSTM and GRU). As a result, we realized that in the context of hydrological forecasting problems, simple architectural models such as LSTM and GRU (with one hidden layer) are sufficient to produce highly reliable forecasts while minimizing computation time because of the sequential data nature.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6C
• /
• pp.349-358
• /
• 2008

In performing the reliability analysis for predicting the settlement with time of alluvial clay layer at Kobe airport, the uncertainties of geotechnical properties were examined based on the stochastic and probabilistic theory. By using Terzaghi's consolidation theory as the objective function, the failure probability was normalized based on AFOSM method. As the result of reliability analysis, the occurrence probabilities for the cases of the target settlement of ${\pm}10%,\;{\pm}25%$ of the total settlement from the deterministic analysis were 30~50%, 60%~90%, respectively. Considering that the variation coefficients of input variable are almost similar as those of past researches, the acceptable error range of the total settlement would be expected in the range of 10% of the predicted total settlement. As the result of sensitivity analysis, the factors which affect significantly on the settlement analysis were the uncertainties of the compression coefficient Cc, the pre-consolidation stress Pc, and the prediction model employed. Accordingly, it is very important for the reliable prediction with high reliability to obtain reliable soil properties such as Cc and Pc by performing laboratory tests in which the in-situ stress and strain conditions are properly simulated.

• Earthquakes and Structures
• /
• v.27 no.1
• /
• pp.17-29
• /
• 2024

The traditional double story isolated structure is a derivative of the base isolated and inter-story isolated structures, while the new double story isolated structure represents a novel variation derived from the traditional double story isolated structure. In order to investigate the seismic response of the new double story isolated structure, a comprehensive structural model was developed. Concurrently, models for the basic fixed, base isolated, inter-story isolated, and traditional double story isolated structures were also established for comparative analysis. The nonlinear dynamic time-history response of the new double story isolated structure under rare earthquake excitations was analyzed. The findings of the study reveal that, in comparison to the basic fixed structure, the new double story isolated structure exhibits superior performance across all evaluated aspects. Furthermore, when compared to the base isolated and inter-story isolated structures, the new double story isolated structure demonstrates significant reductions in inter-story shear force, top acceleration, and inter-frame displacement. The horizontal displacement of the new double story isolated structure is primarily localized within the two isolation layers, effectively dissipating the majority of input seismic energy. In contrast to the traditional double story isolated structure, the new design minimizes displacements within the inter-isolation layer situated in the central part of the frame, as well as mitigates the overturning forces acting on the lower frame column. Consequently, this design ensures the structural integrity of the core tube, thereby preventing potential collapse and structural damage.