• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.220-225
• /
• 2007

This paper presents an efficient input variable selection method using both principal component analysis(PCA) and adaptive partition mutual information(AP-MI) estimation. PCA which is based on 2nd order statistics, is applied to prevent a overestimation by quickly removing the dependence between input variables. AP-MI estimation is also applied to estimate an accurate dependence information by equally partitioning the samples of input variable for calculating the probability density function. The proposed method has been applied to 2 problems for selecting the input variables, which are the 7 artificial signals of 500 samples and the 24 environmental pollution signals of 55 samples, respectively. The experimental results show that the proposed methods has a fast and accurate selection performance. The proposed method has also respectively better performance than AP-MI estimation without the PCA and regular partition MI estimation.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.53.6-53
• /
• 2001

In this paper, we are concerned with a tracking filter algorithm which can track a maneuvering target. Among the novel tracking filter algorithms, the input estimation (IE) filter can be summarized as estimating the unknown maneuver input and compensating the state according to the estimated input, and the variable dimension filter (VDF) can be summarized as detecting the maneuver of target and changing the dimension of the target dynamics to accomodate the maneuver of target They have some goods and bads with respect to each other. The variable dimension filter with input estimation (VDIEF) is constructed by combining the two filtering algorithms. However, it requires too much computational burden while it has good performance. We propose another variable dimension with input estimation ...

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1996.04a
• /
• pp.119-126
• /
• 1996

In this paper, a method of substructural identification is presented for the estimation of localized structural parameters. for this purpose, an auto-regressive and moving average with stochastic input (ARMAX) model is derived for the substructure to process the measurement data impaired by noises. The sequential prediction error method is used fer the estimation of unknown localized parameters. Using the substructural method, the number of unknown parameters can be reduced and the convergence and accuracy of estimation can be improved. For some substructures, the effect of the input excitation is expressed in terms of the responses at the inferences with the main structure, and substructural identification may be carried out without measuring the actual input excitation to the whole structure. Example analysis is carried out for idealized structural models of a multistory building and a truss bridge. The results indicate that the present method is effective and efficient for local damage estimation of complex structures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2328-2344
• /
• 2022

2D human pose estimation still faces difficulty in low-resolution images. Most existing top-down approaches scale up the target human bonding box images to the large size and insert the scaled image into the network. Due to up-sampling, artifacts occur in the low-resolution target images, and the degraded images adversely affect the accurate estimation of the joint positions. To address this issue, we propose a multi-resolution input feature fusion network for human pose estimation. Specifically, the bounding box image of the target human is rescaled to multiple input images of various sizes, and the features extracted from the multiple images are fused in the network. Moreover, we introduce a guiding channel which induces the multi-resolution input features to alternatively affect the network according to the resolution of the target image. We conduct experiments on MS COCO dataset which is a representative dataset for 2D human pose estimation, where our method achieves superior performance compared to the strong baseline HRNet and the previous state-of-the-art methods.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2015.05a
• /
• pp.109-110
• /
• 2015

Estimation of rebar quantity may be used as an index to evaluate the economic feasibility of structural designs. However, when using the software to estimate the rebar quantity, there may be some limitations such as data loss caused by human errors and estimation delays caused by increased input time, since the information on arrangement of rebar is inserted manually. To solve the problems of such quantity estimation software, it is necessary to develop a method on automatic input/output of structural design information for quantity estimation and an algorithm for accurate estimation of rebar quantity. The purpose of this study is to improve the existing rebar quantity estimation by connecting with the database on information related to rebar estimation and the algorithm for rebar estimation, in order to develop an algorithm to estimate an accurate, net rebar quantity. The study result can be used as basic data for development of software for efficient structural designs and automatic framework estimation of buildings.

• The Journal of the Acoustical Society of Korea
• /
• v.27 no.1E
• /
• pp.25-29
• /
• 2008

The projection approximation subspace tracking (PAST) is one of the attractive subspace tracking algorithms, because it estimatesthe signal subspace adaptively and continuously. Furthermore, the computational complexity is relatively low. However, the algorithm still has room for improvement in the subspace estimation accuracy. In this paper, we propose a new algorithm to improve the subspace estimation accuracy using a normally ordered input vector and a reversely ordered input vector simultaneously.

• Smart Structures and Systems
• /
• v.15 no.2
• /
• pp.489-503
• /
• 2015

Response estimation at unmeasured locations using the limited number of measurements is an attractive topic in the field of structural health monitoring (SHM). Because of increasing complexity and size of civil engineering structures, measuring all structural responses from the entire body is intractable for the SHM purpose; the response estimation can be an effective and practical alternative. This paper investigates a response estimation technique based on the Kalman state estimator to combine multi-sensor data under non-zero mean input excitations. The Kalman state estimator, constructed based on the finite element (FE) model of a structure, can efficiently fuse different types of data of acceleration, strain, and tilt responses, minimizing the intrinsic measurement noise. This study focuses on the effects of (a) FE model error and (b) combinations of multi-sensor data on the estimation accuracy in the case of non-zero mean input excitations. The FE model error is purposefully introduced for more realistic performance evaluation of the response estimation using the Kalman state estimator. In addition, four types of measurement combinations are explored in the response estimation: strain only, acceleration only, acceleration and strain, and acceleration and tilt. The performance of the response estimation approach is verified by numerical and experimental tests on a simply-supported beam, showing that it can successfully estimate strain responses at unmeasured locations with the highest performance in the combination of acceleration and tilt.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.31 no.1
• /
• pp.15-24
• /
• 2006

In simulation input modeling, it is important to identify a probability distribution to represent the input process of interest. In this paper, an appropriate sample size is determined for parameter estimation associated with some typical probability distributions frequently encountered in simulation input modeling. For this purpose, a statistical measure is proposed to evaluate the effect of sample size on the precision as well as the accuracy related to the parameter estimation, square rooted mean square error to parameter ratio. Based on this evaluation measure, this sample size effect can be not only analyzed dimensionlessly against parameter's unit but also scaled regardless of parameter's magnitude. In the Monte Carlo simulation experiments, three continuous and one discrete probability distributions are investigated such as ; 1) exponential ; 2) gamma ; 3) normal ; and 4) poisson. The parameter's magnitudes tested are designed in order to represent distinct skewness respectively. Results show that ; 1) the evaluation measure drastically improves until the sample size approaches around 200 ; 2) up to the sample size about 400, the improvement continues but becomes ineffective ; and 3) plots of the evaluation measure have a similar plateau pattern beyond the sample size of 400. A case study with real datasets presents for verifying the experimental results.

• ETRI Journal
• /
• v.15 no.2
• /
• pp.35-51
• /
• 1993

This paper presents function approximation based on nonparametric estimation. As an estimation model of function approximation, a three layered network composed of input, hidden and output layers is considered. The input and output layers have linear activation units while the hidden layer has nonlinear activation units or kernel functions which have the characteristics of bounds and locality. Using this type of network, a many-to-one function is synthesized over the domain of the input space by a number of kernel functions. In this network, we have to estimate the necessary number of kernel functions as well as the parameters associated with kernel functions. For this purpose, a new method of parameter estimation in which linear learning rule is applied between hidden and output layers while nonlinear (piecewise-linear) learning rule is applied between input and hidden layers, is considered. The linear learning rule updates the output weights between hidden and output layers based on the Linear Minimization of Mean Square Error (LMMSE) sense in the space of kernel functions while the nonlinear learning rule updates the parameters of kernel functions based on the gradient of the actual output of network with respect to the parameters (especially, the shape) of kernel functions. This approach of parameter adaptation provides near optimal values of the parameters associated with kernel functions in the sense of minimizing mean square error. As a result, the suggested nonparametric estimation provides an efficient way of function approximation from the view point of the number of kernel functions as well as learning speed.

• Journal of Information Processing Systems
• /
• v.13 no.4
• /
• pp.1029-1042
• /
• 2017

A joint channel estimation and data detection technique for a multiple input multiple output (MIMO) wireless communication system is proposed. It combines the least square (LS) training based channel estimation (TBCE) scheme with sphere decoding. In this new approach, channel estimation is enhanced with the help of blind symbols, which are selected based on their correctness. The correctness is determined via sphere decoding. The performance of the new scheme is studied through simulation in terms of the bit error rate (BER). The results show that the proposed channel estimation has comparable performance and better computational complexity over the existing semi-blind channel estimation (SBCE) method.