• Journal of Korea Multimedia Society
• /
• v.23 no.12
• /
• pp.1540-1551
• /
• 2020

Aiming at the problem that the existing human behavior recognition algorithm cannot fully utilize the multi-level spatio-temporal information of the network, a human behavior recognition algorithm based on a dense three-dimensional residual network is proposed. First, the proposed algorithm uses a dense block of three-dimensional residuals as the basic module of the network. The module extracts the hierarchical features of human behavior through densely connected convolutional layers; Secondly, the local feature aggregation adaptive method is used to learn the local dense features of human behavior; Then, the residual connection module is applied to promote the flow of feature information and reduced the difficulty of training; Finally, the multi-layer local feature extraction of the network is realized by cascading multiple three-dimensional residual dense blocks, and use the global feature aggregation adaptive method to learn the features of all network layers to realize human behavior recognition. A large number of experimental results on benchmark datasets KTH show that the recognition rate (top-l accuracy) of the proposed algorithm reaches 93.52%. Compared with the three-dimensional convolutional neural network (C3D) algorithm, it has improved by 3.93 percentage points. The proposed algorithm framework has good robustness and transfer learning ability, and can effectively handle a variety of video behavior recognition tasks.

• Advances in nano research
• /
• v.13 no.2
• /
• pp.165-174
• /
• 2022

In the present study, a hybrid model of artificial neural network (ANN) and internet of things (IoT) is proposed to overcome the difficulties in deriving governing equations and numerical solutions of the dynamical behavior of the nano-systems. Nano-structures manifest size-dependent behavior in response to static and dynamic loadings. Nonlocal and length-scale parameters alongside with other geometrical, loading and material parameters are taken as input parameters of an ANN to observe the natural frequency and damping behavior of micro sensors made from nanocomposite material with piezoelectric layers. The behavior of a micro-beam is simulated using famous numerical methods in literature under base vibrations. The ANN was further trained to correlate the output vibrations to the base vibration. Afterwards, using IoT, the electrical potential conducted in the sensors are collected and converted to numerical data in an embedded mini-computer and transferred to a server for further calculations and decision by ANN. The ANN calculates the base vibration behavior with is crucial in mechanical systems. The speed and accuracy of the ANN in determining base excitation behavior are the strengths of this network which could be further employed by engineers and scientists.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.39 no.3
• /
• pp.63-82
• /
• 2014

The main objective of this research is to investigate whether the RFM (recency-frequency-monetary value) information of a customer's redemption behavior of loyalty points can improve the prediction of future value of the customer. The conventional measurement of customer value has been primarily based on purchase transactions behavior although a customer's future behavior can be also influenced by other interactions between the customer and the firm such as redemption of rewards in a loyalty program. We theorize why a customer's redemption behavior can influence her future purchases and thereby the customer's total value based on operant learning theory, goal gradient hypothesis, and lock-in effect. Using a dataset from a major book store in Korea spanning three years between 2008 and 2010, we analyze both purchase transactions and redemption records of over 10,000 customers. The results show that the redemption-based RFM information does improve the prediction accuracy of the customer's future purchases. Based on this result, we also propose an improved estimate of customer lifetime value (CLV) by combining purchase transactions and loyalty points redemption data. Managerial implications will be also discussed for firms managing loyalty programs to maximize the total value customers.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06a
• /
• pp.873-878
• /
• 1994

Application of the filtered-x LMS adaptive filter to active noise cancellation requires to estimate the transfer characteristics between the output and the error signal of the adaptive canceller. We analyze the effects of estimation accuracy on the convergence behavior of the canceller when the input noise is modeled as a sinusoid.

• Journal of Microbiology and Biotechnology
• /
• v.34 no.4
• /
• pp.891-901
• /
• 2024

This study focuses on improving the 3D printability of pea protein with the help of food inks designed for jet-type 3D printers. Initially, the food ink base was formulated using nanocellulose-alginate with a gradient of native potato starch and its 3D printability was evaluated. The 3D-printed structures using only candidates for the food ink base formulated with or without potato starch exhibited dimensional accuracy exceeding 95% on both the X and Y axes. However, the accuracy of stacking on the Z-axis was significantly affected by the ink composition. Food ink with 1% potato starch closely matched the CAD design, with an accuracy of approximately 99% on the Z-axis. Potato starch enhanced the stacking of 3D-printed structures by improving the electrostatic repulsion, viscoelasticity, and thixotropic behavior of the food ink base. The 3D printability of pea protein was evaluated using the selected food ink base, showing a 46% improvement in dimensional accuracy on the Z-axis compared to the control group printed with a food ink base lacking potato starch. These findings suggest that starch can serve as an additive support for high-resolution 3D jet-type printing of food ink material.

• Structural Engineering and Mechanics
• /
• v.82 no.2
• /
• pp.245-258
• /
• 2022

Many Structural Health Monitoring (SHM) methods have been proposed for structural damage diagnosis and prognosis. However, SHM for pinched hysteretic structures can be problematic due to the high level of nonlinearity. The model-free hysteresis loop analysis (HLA) has displayed notable robustness and accuracy in identifying damage for full-scaled and scaled test buildings. In this paper, the performance of HLA is compared with seven other SHM methods in identifying lateral elastic stiffness for a six-story numerical building with highly nonlinear pinching behavior. Two successive earthquakes are employed to compare the accuracy and consistency of methods within and between events. Robustness is assessed across sampling rates 50-1000 Hz in noise-free condition and then assessed with 10% root mean square (RMS) noise added to responses at 250 Hz sampling rate. Results confirm HLA is the most robust method to sampling rate and noise. HLA preserves high accuracy even when the sampling rate drops to 50 Hz, where the performance of other methods deteriorates considerably. In noisy conditions, the maximum absolute estimation error is less than 4% for HLA. The overall results show HLA has high robustness and accuracy for an extremely nonlinear, but realistic case compared to a range of leading and recent model-based and model-free methods.

• Elastomers and Composites
• /
• v.44 no.4
• /
• pp.378-383
• /
• 2009

In this study, the use of a condensed model is proposed for the simulation of forming of sandwich sheet with pyramid core. A corresponding finite element analysis for L-type bending is carried out to prove the accuracy and the effectiveness. In order to improve the accuracy of forming analysis, more precise modeling of core shape and consideration of work-hardening of previous core forming are carried out. Simulation results are compared with those of experiment. Deformation shape and post-buckling behavior by simulation are in good agreement with those of experiment for the considerable range of deformation. From the comparison of force-displacement curve, it is shown that the proposed model shows good prediction of post-buckling behavior.

• Geomechanics and Engineering
• /
• v.34 no.5
• /
• pp.529-545
• /
• 2023

The coupled soil-pile-structure seismic response is recently in the spotlight of researchers because of its extensive applications in the different fields of engineering such as bridges, offshore platforms, wind turbines, and buildings. In this paper, a simple analytical model is developed to evaluate the dynamic performance of seismically isolated bridges considering triple interactions of soil, piles, and bridges simultaneously. Novel expressions are proposed to present the dynamic behavior of pile groups in inhomogeneous soils with various shear modulus along with depth. Both cohesive and cohesionless soil deposits can be simulated by this analytical model with a generalized function of varied shear modulus along the soil depth belonging to an inhomogeneous stratum. The methodology is discussed in detail and validated by rigorous dynamic solution of 3D continuum modeling, and time history analysis of centrifuge tests. The proposed analytical model accuracy is guaranteed by the acceptable agreement between the experimental/numerical and analytical results. A comparison of the proposed linear model results with nonlinear centrifuge tests showed that during moderate (frequent) earthquakes the relative differences in responses of the superstructure and the pile cap can be ignored. However, during strong excitations, the response calculated in the linear time history analysis is always lower than the real conditions with the nonlinear behavior of the soil-pile-bridge system. The current simple and efficient method provides the accuracy and the least computational costs in comparison to the full three-dimensional analyses.

• Tribology and Lubricants
• /
• v.24 no.1
• /
• pp.27-33
• /
• 2008

A micro-impression creep machine was designed and developed, adopting a small punch in diameter of 150 um, displacement gage with an accuracy of sub-${\mu}m$ scale, and load-cell with an accuracy of mN scale in order to investigate creep behavior of small solder ball in diameter of less than 1 mm. Creep behavior of lead-free solder ball(Sn-3.0Ag-0.5Cu) in diameter of $760\;{\mu}m$ was investigated in the stress range of $8{\sim}60\;MPa$ and at $303\;K{\sim}393\;K$. The applied load became decreased slightly and continuously in the creep rate of $10^{-4}/s$ range during the current experiments. Also, the machine frame was so sensitive to the environmental temperature that nm scaled displacement recording was unstable according to the change in environmental temperature.

• Journal of the Korean Society for Precision Engineering
• /
• v.18 no.4
• /
• pp.129-134
• /
• 2001

It is essential to predict the welding deformation at assembly stage, to increase productivity through mechanization and automation effectively. A practical analysis method appled for production engineering was proposed to simulate the deformation of arc welding, with an analytical model using finite element method solving thermal-elastic-plastic behavior. In this research, for accurate assembling, 3-D thermal-elastic-plastic finite element model is used to simulate the out-of-plane deformation caused by arc welding. Efforts have been made to find out the efficient method to improve the reliability and accuracy of the numerical calculation. Each of theories of small and large deformation is applied in solving 3-D thermal-elastic-plastic problem to compare with their efficiency about calculation imes and solution accuracy. When solid elements are used in a bending problem of a plate, phenomenon that the predictive deformation is more than that of actual survey is observed. To prevent this phenomenon, reduced integration method for element is employed instead of full integration that is generally used in 3-D thermal-elastic-plastic analysis.