• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.392-412
• /
• 2018

Due to the semantic gap problem across different modalities, automatically retrieval from multimedia information still faces a main challenge. It is desirable to provide an effective joint model to bridge the gap and organize the relationships between them. In this work, we develop a deep image annotation and classification by fusing multi-modal semantic topics (DAC_mmst) model, which has the capacity for finding visual and non-visual topics by jointly modeling the image and loosely related text for deep image annotation while simultaneously learning and predicting the class label. More specifically, DAC_mmst depends on a non-parametric Bayesian model for estimating the best number of visual topics that can perfectly explain the image. To evaluate the effectiveness of our proposed algorithm, we collect a real-world dataset to conduct various experiments. The experimental results show our proposed DAC_mmst performs favorably in perplexity, image annotation and classification accuracy, comparing to several state-of-the-art methods.

• Proceedings of the KIEE Conference
• /
• 2007.04a
• /
• pp.204-206
• /
• 2007

In this study, we introduce the optimization method of fuzzy inference systems that is based on Hierarchical Fair Competition-based Parallel Genetic Algorithms (HFCGA) and information data granulation, The granulation is realized with the aid of the Hard C-means clustering and HFCGA is a kind of multi-populations of Parallel Genetic Algorithms (PGA), and it is used for structure optimization and parameter identification of fuzzy model. It concerns the fuzzy model-related parameters such as the number of input variables to be used, a collection of specific subset of input variables, the number of membership functions, the order of polynomial, and the apexes of the membership function. In the optimization process, two general optimization mechanisms are explored. The structural optimization is realized via HFCGA and HCM method whereas in case of the parametric optimization we proceed with a standard least square method as well as HFCGA method as well. A comparative analysis demonstrates that the proposed algorithm is superior to the conventional methods. Particularly, in parameter identification, we use the UNDX operator which uses multiple parents and generate offsprings around the geographic center off mass of these parents.

• Smart Structures and Systems
• /
• v.26 no.4
• /
• pp.533-544
• /
• 2020

Effective bridge maintenance reduces bridge operation costs and extends its service life. The possibility of storing bridge life-cycle data in a 3D parametric model of the bridge through Bridge Information Modeling (BrIM) provides new opportunities to enhance current practices of bridge maintenance management. This study develops a Decision Support System (DSS), namely BrDSS, which employs BrIM and an efficient optimization model for bridge maintenance planning. The BrIM model in BrDSS extracts basic data of elements required for the optimization process and visualizes the inspection data and the optimization results to the user to help in decision makings. In the optimization module of the DSS, the specifically formulated Genetic Algorithm (GA) eliminates the chances of producing infeasible solutions for faster convergence. The practicality of the presented DSS was explored by utilizing the DSS in the maintenance planning of a bridge under operation in the southwest of Iran.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.2
• /
• pp.703-722
• /
• 2016

Passive wireless devices have increasing civilian and military applications, especially in the scenario with wearable devices and Internet of Things. In this paper, we study indoor localization of a target equipped with radio-frequency identification (RFID) device in ultra-wideband (UWB) wireless networks. With known room layout, deterministic multipath components, including the line-of-sight (LOS) signal and the reflected signals via multipath propagation, are employed to locate the target with one transmitter and a single inaccurate receiver. A factor graph corresponding to the joint posterior position distribution of target and receiver is constructed. However, due to the mixed distribution in the factor node of likelihood function, the expressions of messages are intractable by directly applying belief propagation on factor graph. To this end, we approximate the messages by Gaussian distribution via minimizing the Kullback-Leibler divergence (KLD) between them. Accordingly, a parametric message passing algorithm for indoor passive localization is derived, in which only the means and variances of Gaussian distributions have to be updated. Performance of the proposed algorithm and the impact of critical parameters are evaluated by Monte Carlo simulations, which demonstrate the superior performance in localization accuracy and the robustness to the statistics of multipath channels.

• Earthquakes and Structures
• /
• v.8 no.6
• /
• pp.1499-1528
• /
• 2015

The yielding mechanisms of reinforced concrete (RC) structures are the main cause of the collapse of RC buildings during earthquake excitation. Nowadays, the application of earthquake energy dissipation devices, such as viscous dampers (VDs), is being widely considered to protect RC structures which are designed to withstand severe seismic loads. However, the effect of VDs on the formation of plastic hinges and the yielding criteria of RC members has not been investigated extensively, due to the lack of an analytical model and a numerical means to evaluate the seismic response of structures. Therefore, this paper offers a comprehensive investigation of how damper devices influence the yielding mechanisms of RC buildings subjected to seismic excitation. For this purpose, adapting the Newmark method, a finite element algorithm was developed for the nonlinear dynamic analysis of reinforced concrete buildings equipped with VDs that are subjected to earthquake. A special finite element computer program was codified based on the developed algorithm. Finally, a parametric study was conducted for a three-story RC building equipped with supplementary VD devices, performing a nonlinear analysis in order to evaluate its effect on seismic damage and on the response of the structure. The results of this study showed that implementing VDs substantially changes the mechanism and formation of plastic hinges in RC buildings.

• Journal of Korean Society of Steel Construction
• /
• v.11 no.1 s.38
• /
• pp.13-22
• /
• 1999

The objective of this paper is to develop a program for distortional analysis of steel box girder bridges. This program is formulated by using MSDM(modified slope deflection method). Two examples are carried out to verify the validity of the developed technique and its computation procedures. The analyzed results are compared with the previously proposed methods, BEF(beam on elastic foundation) and EBEF(equivalent beam on elastic foundation). The BEF method is limited only to prismatic straight box girders. In the EBEF method, stiffness of the intermediate diaphragms is infinitely considered. On the other hand, stiffness of the intermediate diaphragms is idealized as spring contant in this study. And then, nonprismatic straight box girders can be analyzed using the same procedure. Therefore, the comparison shows that the MSDM algorithm proposed in this paper is more efficient and reliable. Also parametric studies are perfomed using the proposed algorithm.

• Journal of the Society of Naval Architects of Korea
• /
• v.45 no.4
• /
• pp.403-410
• /
• 2008

This paper describes the optimization of the rudder section by the genetic algorism based on VLM(Vortex Lattice Method) and panel method. The developed propeller-rudder analysis program has been validated by comparing with experimental data. The research extends to optimize the anti-erosion rudder section of the large container ship. The object function is the amount of pressure at leading edge of rudder which is closely related with erosion phenomena. The optimized rudder has been compared with conventional rudder with NACA 0021 section by analyzing with the developed program. The finally optimized section has low and mild pressure distribution in comparison with the NACA rudder. The experiments is expected to be carried out for the validation of the present optimization and more parametric study of section geometry is also expected to be conducted in the near future.

• The KIPS Transactions:PartB
• /
• v.15B no.6
• /
• pp.533-542
• /
• 2008

Background subtraction from video data is one of the most important task in various realtime machine vision applications. In this paper, a new scheme for background subtraction based on the hexagonal pixel sampling is proposed. Generally it has been found that hexagonal spatial sampling yields smaller quantization errors and remarkably improves the understanding of connectivity. We try to apply the hexagonally sampled image to the LBP based non-parametric background subtraction algorithm. Our scheme makes it possible to omit the bilinear pixel interpolation step during the local binary pattern generation process, and, consequently, can reduce the computation time. Experimental results revealed that our approach based on hexagonal spatial sampling is very efficient and can be utilized in various background subtraction applications.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.4
• /
• pp.310-315
• /
• 2015

A single channel VAD (Voice Activity Detection) algorithm for nonstationary noise environment is proposed in this paper. Threshold values of the feature parameter for VAD decision are updated adaptively based on estimates of means and standard deviations of past non-speech frames. The feature parameter, SPD-TE (Spectral Power Difference-Teager Energy), is obtained by applying the Teager energy to the WPD (Wavelet Packet Decomposition) coefficients. It was reported previously that the SPD-TE is robust to noise as a feature for VAD. Experimental results by using TIMIT speech and NOISEX-92 noise databases show that decision accuracy of the proposed algorithm is comparable to several typical VAD algorithms including standards for SNR values ranging from 10 to -10 dB.

• Structural Engineering and Mechanics
• /
• v.10 no.6
• /
• pp.517-532
• /
• 2000

A direct output feedback control scheme was recently proposed by the authors for single-story building structures resting on flexible soil body. In this paper, the control scheme is extended to mitigate the seismic responses of multi-story buildings. Soil-structure interaction is taken into account in two parts: input at the soil-structure interface/foundation and control algorithm. The former reflects the effect on ground motions and is monitored in real time with accelerometers at foundation. The latter includes the effect on the dynamic characteristics of structures, which is formulated by modifying the classical linear quadratic regulator based on the fundamental mode shape of the soil-structure system. Numerical result on the study of a $\frac{1}{4}$-scale three-story structure, supported by a viscoelastic half-space of soil mass, have demonstrated that the proposed algorithm is robust and very effective in suppressing the earthquake-induced vibration in building structures even supported on a flexible soil mass. Parametric studies are performed to understand how soil damping and flexibility affect the effectiveness of active tendon control. The selection of weighting matrix and effect of soil property uncertainty are investigated in detail for practical applications.