• Journal of Broadcast Engineering
• /
• v.25 no.7
• /
• pp.1095-1106
• /
• 2020

In this paper, we propose an algorithm that extracts a high-precision 3D skeleton using a model generated using a distributed RGB-D camera. When information about a 3D model is extracted through a distributed RGB-D camera, if the information of the 3D model is used, a skeleton with higher precision can be obtained. In this paper, in order to improve the precision of the 2D skeleton, we find the conditions to obtain the 2D skeleton well using the PCA. Through this, high-quality 2D skeletons are obtained, and high-precision 3D skeletons are extracted by combining the information of the 2D skeletons. Even though this process goes through, the generated skeleton may have errors, so we propose an algorithm that removes these errors by using the information of the 3D model. We were able to extract very high accuracy skeletons using the proposed method.

• Journal of Korean Association for Spatial Structures
• /
• v.21 no.2
• /
• pp.99-110
• /
• 2021

This study presents the estimation of crack depth by analyzing temperatures extracted from thermal images and environmental parameters such as air temperature, air humidity, illumination. The statistics of all acquired features and the correlation coefficient among thermal images and environmental parameters are presented. The concrete crack depths were predicted by four different machine learning models: Multi-Layer Perceptron (MLP), Random Forest (RF), Gradient Boosting (GB), and AdaBoost (AB). The machine learning algorithms are validated by the coefficient of determination, accuracy, and Mean Absolute Percentage Error (MAPE). The AB model had a great performance among the four models due to the non-linearity of features and weak learner aggregation with weights on misclassified data. The maximum depth 11 of the base estimator in the AB model is efficient with high performance with 97.6% of accuracy and 0.07% of MAPE. Feature importances, permutation importance, and partial dependence are analyzed in the AB model. The results show that the marginal effect of air humidity, crack depth, and crack temperature in order is higher than that of the others.

• Journal of the Korean Geotechnical Society
• /
• v.36 no.12
• /
• pp.27-34
• /
• 2020

Use of axisymmetric shell element for the structure increases the efficiency and accuracy in finite element analysis of the interaction between the ground and the structure. This paper derived the force balance equation and the moment balance equation for an axisymmetric shell element based on Kirchhoff's theory. The governing equation for the axial deformation used the isoparametric shape function in the Galerkin formulation, and the governing equation for the shell bending used the higher-order shape function. The developed axisymmetric shell element was combined with Geo-COUS, a geotechnical finite element program for the coupled analysis with the ground. The accuracy of the developed element was confirmed through the example analyses of the circular plate and the liquid storage tank. And the energy balance equation for the axisymmetric shell element is presented.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.293-298
• /
• 2023

The military is facing a continuous decrease in personnel, and in order to cope with potential accidents and challenges in operations, efforts are being made to reduce the direct involvement of personnel by utilizing the latest technologies. Recently, the use of various sensors related to Manned-Unmanned Teaming and artificial intelligence technologies has gained attention, emphasizing the need for flexible utilization methods. In this paper, we propose four dataset construction methods that can be used for effective training of robots that can be deployed in military operations, utilizing not only RGB image data but also data acquired from IR image sensors. Since there is no publicly available dataset that combines RGB and IR image data, we directly acquired the dataset within buildings. The input values were constructed by combining RGB and IR image sensor data, taking into account the field of view, resolution, and channel values of both sensors. We compared the proposed method with conventional RGB image data classification training using the same learning model. By employing the proposed image data fusion method, we observed improved stability in training loss and approximately 3% higher accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.2
• /
• pp.200-206
• /
• 2024

Computing music similarity is an indispensable component in developing music search service. This paper proposes a relevance weight of each chromagram vector for cover song identification in computing a music similarity function in order to boost identification accuracy. We derive a music similarity function using the relevance weight based on the probabilistic relevance model, where higher relevance weights are assigned to less frequently-occurring discriminant chromagram vectors while lower weights to more frequently-occurring ones. Experimental results performed on two cover music datasets show that the proposed music similarity improves the cover song identification performance.

• Structural Engineering and Mechanics
• /
• v.73 no.6
• /
• pp.667-684
• /
• 2020

In this article, the static responses of layered magneto-electro-thermo-elastic (METE) plates in thermal environment have been investigated through FE methods. By using Reddy's third order shear deformation theory (TSDT) in association with the Hamilton's principle, the direct and derived quantities of the coupled system have been obtained. The coupled governing equations of METE plates have been derived through condensation technique. Three layered METE plates composed of piezoelectric and piezomagnetic phases are considered for evaluation. For investigating the correctness and accuracy, the results in this article are validated with previous researches. In addition, a special attention has been paid to evaluate the influence of different electro-magnetic boundary conditions and pyrocoupling on the coupled response of METE plates. Finally, the influence of stacking sequences, magnitude of temperature load and aspect ratio on the coupled static response of METE plates are investigated in detail.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.7
• /
• pp.424-433
• /
• 2003

Evolutionary design related to the optimal design of Polynomial Neural Networks (PNNs) structure for model identification of complex and nonlinear system is studied in this paper. The PNN structure is consisted of layers and nodes like conventional neural networks but is not fixed and can be changable according to the system environments. three types of polynomials such as linear, quadratic, and modified quadratic is used in each node that is connected with various kinds of multi-variable inputs. Inputs and order of polynomials in each node are very important element for the performance of model. In most cases these factors are decided by the background information and trial and error of designer. For the high reliability and good performance of the PNN, the factors must be decided according to a logical and systematic way. In the paper evolutionary algorithm is applied to choose the optimal input variables and order. Evolutionary (genetic) algorithm is a random search optimization technique. The evolved PNN with optimally chosen input variables and order is not fixed in advance but becomes fully optimized automatically during the identification process. Gas furnace and pH neutralization processes are used in conventional PNN version are modeled. It shows that the designed PNN architecture with evolutionary structure optimization can produce the model with higher accuracy than previous PNN and other works.

• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.52 no.7
• /
• pp.424-424
• /
• 2003

Evolutionary design related to the optimal design of Polynomial Neural Networks (PNNs) structure for model identification of complex and nonlinear system is studied in this paper. The PNN structure is consisted of layers and nodes like conventional neural networks but is not fixed and can be changable according to the system environments. three types of polynomials such as linear, quadratic, and modified quadratic is used in each node that is connected with various kinds of multi-variable inputs. Inputs and order of polynomials in each node are very important element for the performance of model. In most cases these factors are decided by the background information and trial and error of designer. For the high reliability and good performance of the PNN, the factors must be decided according to a logical and systematic way. In the paper evolutionary algorithm is applied to choose the optimal input variables and order. Evolutionary (genetic) algorithm is a random search optimization technique. The evolved PNN with optimally chosen input variables and order is not fixed in advance but becomes fully optimized automatically during the identification process. Gas furnace and pH neutralization processes are used in conventional PNN version are modeled. It shows that the designed PNN architecture with evolutionary structure optimization can produce the model with higher accuracy than previous PNN and other works.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.369-384
• /
• 2021

In a personalized product recommendation system, when the amount of log data is large or sparse, the accuracy of model recommendation will be greatly affected. To solve this problem, a personalized product recommendation method using deep factorization machine (DeepFM) to analyze user behavior is proposed. Firstly, the K-means clustering algorithm is used to cluster the original log data from the perspective of similarity to reduce the data dimension. Then, through the DeepFM parameter sharing strategy, the relationship between low- and high-order feature combinations is learned from log data, and the click rate prediction model is constructed. Finally, based on the predicted click-through rate, products are recommended to users in sequence and fed back. The area under the curve (AUC) and Logloss of the proposed method are 0.8834 and 0.0253, respectively, on the Criteo dataset, and 0.7836 and 0.0348 on the KDD2012 Cup dataset, respectively. Compared with other newer recommendation methods, the proposed method can achieve better recommendation effect.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.38 no.4
• /
• pp.196-203
• /
• 2022

Purpose: This study aimed to evaluate the accuracy and clinical usability of an identification model using deep learning for 79 dental implant types. Materials and Methods: A total of 45396 implant fixture images were collected through panoramic radiographs of patients who received implant treatment from 2001 to 2020 at 30 dental clinics. The collected implant images were 79 types from 18 manufacturers. EfficientNet and Meta Pseudo Labels algorithms were used. For EfficientNet, EfficientNet-B0 and EfficientNet-B4 were used as submodels. For Meta Pseudo Labels, two models were applied according to the widen factor. Top 1 accuracy was measured for EfficientNet and top 1 and top 5 accuracy for Meta Pseudo Labels were measured. Results: EfficientNet-B0 and EfficientNet-B4 showed top 1 accuracy of 89.4. Meta Pseudo Labels 1 showed top 1 accuracy of 87.96, and Meta pseudo labels 2 with increased widen factor showed 88.35. In Top5 Accuracy, the score of Meta Pseudo Labels 1 was 97.90, which was 0.11% higher than 97.79 of Meta Pseudo Labels 2. Conclusion: All four deep learning algorithms used for implant identification in this study showed close to 90% accuracy. In order to increase the clinical applicability of deep learning for implant identification, it will be necessary to collect a wider amount of data and develop a fine-tuned algorithm for implant identification.