• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.357-363
• /
• 2023

In underwater signal processing, separating individual signals from mixed signals has long been a challenge due to low signal quality. The common method using Short-time Fourier transform for spectrogram analysis has faced criticism for its complex parameter optimization and loss of phase data. We propose a Triple-path Recurrent Neural Network, based on the Dual-path Recurrent Neural Network's success in long time series signal processing, to handle three-dimensional tensors from multi-channel sensor input signals. By dividing input signals into short chunks and creating a 3D tensor, the method accounts for relationships within and between chunks and channels, enabling local and global feature learning. The proposed technique demonstrates improved Root Mean Square Error and Scale Invariant Signal to Noise Ratio compared to the existing method.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3A
• /
• pp.457-464
• /
• 2006

Fiber-reinforced composite materials due to their high specific strength, high stiffness and light weight are becoming increasingly used in many engineering industry, especially in the aerospace, marin and civil, etc. In this paper, the buckling load and mode shapes of composite laminates with elliptical cross-section including transverse shear deformations are analyzed. For solving this problems, a versatile flat shell element has been developed by combining a membrane element with drilling degree-of-freedom and a plate bending element. Also, an improved shell element has been established by the combined use of the addition of enhanced assumed strain and the substitute shear strain fields. The combined influence of shell geometry and elliptical cross-sectional parameter, fiber angle, and lay-up on the buckling loads of elliptical cylinder is examined. The critical buckling loads and mode shapes analyzed here may serve as a benchmark for future investigations.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.205-208
• /
• 2022

Multi-exposure high dynamic range (HDR) image reconstruction, the task of reconstructing an HDR image from multiple low dynamic range (LDR) images in a dynamic scene, often produces ghosting artifacts caused by camera motion and moving objects and also cannot deal with washed-out regions due to over or under-exposures. While there has been many deep-learning-based methods with motion estimation to alleviate these problems, they still have limitations for severely moving scenes. They also require large parameter counts, especially in the case of state-of-the-art methods that employ attention modules. To address these issues, we propose a frequency domain approach based on the idea that the transform domain coefficients inherently involve the global information from whole image pixels to cope with large motions. Specifically we adopt Residual Fast Fourier Transform (RFFT) blocks, which allows for global interactions of pixels. Moreover, we also employ Depthwise Overparametrized convolution (DO-conv) blocks, a convolution in which each input channel is convolved with its own 2D kernel, for faster convergence and performance gains. We call this LFFNet (Lightweight Frequency Fusion Network), and experiments on the benchmarks show reduced ghosting artifacts and improved performance up to 0.6dB tonemapped PSNR compared to recent state-of-the-art methods. Our architecture also requires fewer parameters and converges faster in training.

• Smart Structures and Systems
• /
• v.30 no.5
• /
• pp.479-500
• /
• 2022

In this study, a new type of isolation bearing is proposed by combining S-shaped steel plate dampers (SSDs) with a spherical steel bearing, and the seismic control effect of a five-span standard high-speed railway bridge is investigated. The advantages of the proposed S-shaped steel damping friction bearing (SSDFB) are that it cannot only lengthen the structural periods, dissipate the seismic energy, but also prevent bridge unseating due to the restraint effectiveness of SSDs in the large relative displacements between the girders and piers. This study first presents a detailed description and working principle of the SSDFB. Then, mechanical modeling of the SSDFB was derived to fundamentally define its cyclic behavior and obtain key mechanical parameters. The numerical model of the SSDFB's critical component SSD was verified by comparing it with the experimental results. After that, parameter studies of the dimensions and number of SSDs, the friction coefficient, and the gap length of the SSDFBs were conducted. Finally, the longitudinal seismic responses of the bridge with SSDFBs were compared with the bridge with spherical bearing and spherical bearing with strengthened shear keys. The results showed that the SSDFB can not only significantly mitigate the shear force responses and residual displacement in bridge substructures but also can effectively reduce girder displacement and prevent bridge unseating, at a cost of inelastic deformation of the SSDs, which is easy to replace. In conclusion, the SSDFB is expected to be a cost-effective option with both multi-stage energy dissipation and restraint capacity, making it particularly suitable for seismic isolation application to high-speed railway bridges.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.93-101
• /
• 2024

Accurate field crop classification is essential for various agricultural applications, yet existing methods face challenges due to diverse crop types and complex field conditions. This study aimed to address these issues by combining support vector machine (SVM) models with multi-seasonal unmanned aerial vehicle (UAV) images, texture information extracted from Gray Level Co-occurrence Matrix (GLCM), and RGB spectral data. Twelve high-resolution UAV image captures spanned March-October 2021, while field surveys on three dates provided ground truth data. We focused on data from August (-A), September (-S), and October (-O) images and trained four support vector classifier (SVC) models (SVC-A, SVC-S, SVC-O, SVC-AS) using visual bands and eight GLCM features. Farm maps provided by the Ministry of Agriculture, Food and Rural Affairs proved efficient for open-field crop identification and served as a reference for accuracy comparison. Our analysis showcased the significant impact of hyperparameter tuning (C and gamma) on SVM model performance, requiring careful optimization for each scenario. Importantly, we identified models exhibiting distinct high-accuracy zones, with SVC-O trained on October data achieving the highest overall and individual crop classification accuracy. This success likely stems from its ability to capture distinct texture information from mature crops.Incorporating GLCM features proved highly effective for all models,significantly boosting classification accuracy.Among these features, homogeneity, entropy, and correlation consistently demonstrated the most impactful contribution. However, balancing accuracy with computational efficiency and feature selection remains crucial for practical application. Performance analysis revealed that SVC-O achieved exceptional results in overall and individual crop classification, while soybeans and rice were consistently classified well by all models. Challenges were encountered with cabbage due to its early growth stage and low field cover density. The study demonstrates the potential of utilizing farm maps and GLCM features in conjunction with SVM models for accurate field crop classification. Careful parameter tuning and model selection based on specific scenarios are key for optimizing performance in real-world applications.

• Advances in nano research
• /
• v.15 no.3
• /
• pp.215-224
• /
• 2023

Nonlinearity plays an important role in control systems and the application of design. For this reason, in addition to linear vibrations, nonlinear vibrations of the stepped nanobeam are also discussed in this manuscript. This study investigated the vibrations of stepped nanobeams according to Eringen's nonlocal elasticity theory. Eringen's nonlocal elasticity theory was used to capture the nanoscale effect. The nanoscale stepped Euler Bernoulli beam is considered. The equations of motion representing the motion of the beam are found by Hamilton's principle. The equations were subjected to nondimensionalization to make them independent of the dimensions and physical structure of the material. The equations of motion were found using the multi-time scale method, which is one of the approximate solution methods, perturbation methods. The first section of the series obtained from the perturbation solution represents a linear problem. The linear problem's natural frequencies are found for the simple-simple boundary condition. The second-order part of the perturbation solution is the nonlinear terms and is used as corrections to the linear problem. The system's amplitude and phase modulation equations are found in the results part of the problem. Nonlinear frequency-amplitude, and external frequency-amplitude relationships are discussed. The location of the step, the radius ratios of the steps, and the changes of the small-scale parameter of the theory were investigated and their effects on nonlinear vibrations under simple-simple boundary conditions were observed by making comparisons. The results are presented via tables and graphs. The current beam model can assist in designing and fabricating integrated such as nano-sensors and nano-actuators.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.369-377
• /
• 2023

In the industrial manufacturing sector, quality control is pivotal for minimizing defect rates; inadequate management can result in additional costs and production delays. This study underscores the significance of detecting texture defects in manufactured goods and proposes a more precise defect detection technique. While the DFR(Deep Feature Reconstruction) model adopted an approach based on feature map amalgamation and reconstruction, it had inherent limitations. Consequently, we incorporated a new loss function using statistical methodologies, integrated a skip connection structure, and conducted parameter tuning to overcome constraints. When this enhanced model was applied to the texture category of the MVTec-AD dataset, it recorded a 2.3% higher Defect Segmentation AUC compared to previous methods, and the overall defect detection performance was improved. These findings attest to the significant contribution of the proposed method in defect detection through the reconstruction of feature map combinations.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.2
• /
• pp.17-26
• /
• 2024

In recent times, accidents involving structural elements, formwork, and shore have been persistently occurring during concrete pouring, especially in multi-story reinforced concrete (RC) structures. In previous studies, research on construction load analysis was mainly conducted for cases where the thickness of all slabs is constant. However, when the thickness of some slabs is different, the variation in the stiffness of slab cross-sections can lead to different distributions of construction loads, necessitating further investigation. In this study, the slab thickness was set as a variable, and the analysis of the distribution of construction loads was conducted, taking into account the influence of changes in slab thickness on the concrete stiffness and structure. It was confirmed that not only the concrete material stiffness but also the slab cross-section stiffness should be considered in the estimation of construction loads when the slab thickness changes. As the slab thickness increases, the maximum construction load and maximum damage parameter on the layer with increased thickness significantly increase, and it was observed that a thicker slab results in a higher proportion of construction load.

• The Journal of the Acoustical Society of Korea
• /
• v.27 no.6
• /
• pp.296-302
• /
• 2008

This study investigated that the preference model of soloist performers for both vocal and instrumental types in terms of a stage support parameter, ST1. The test was carried out on a stage of a fan-shaped multi-purpose hall with orchestra shell. Objective measurements were carried out at 15 positions on the stage to evaluate various stage condition. The results showed that ST1 varies between -19.9 and -11.3 dB. Vocal and instrumental players participated in performance evaluation test as they played at 5 selected positions according to ST1 values. Players' preference was evaluated by 5-point rating and rank ordering method. As a result, it was found that the preference model of vocalist is different from that of instrumentalist. It was also found that the ST1 does not correlate well with the performer's preference.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.3
• /
• pp.179-186
• /
• 2024

In this paper, a modification coefficient for equivalent single degree of freedom (SDOF), considering the plasticity range of the member subjected to shock wave type of blast load, was developed. The modification coefficient for the equivalent SDOF was determined through comparison with the analysis of a multi-degree of freedom (MDOF) system. The parameters influencing the equivalent SDOF system analysis were chosen as the boundary conditions of the member and the ratio of the duration of blast load to the natural period of the member. The modification coefficient was calculated based on the elastic load-mass transformation factor. The modification coefficient curve was derived using an elliptical equation to ensure it exists between the upper and lower parameter bounds. Using the modification coefficient on examples with varying cross sections and boundary conditions reduced the SDOF analysis error rate from 15% to 3%. This study shows that using the modification coefficient significantly improves the accuracy of SDOF analysis. The modification coefficient proposed in this study can be used for blast analysis.