• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.1
• /
• pp.10-21
• /
• 2023

For various target tracking applications, it is well known that the Kalman filter is the optimal estimator(in the minimum mean-square sense) to predict and estimate the state(position and/or velocity) of linear dynamical systems driven by Gaussian stochastic noise. In the case of nonlinear systems, Extended Kalman filter(EKF) and/or Unscented Kalman filter(UKF) are widely used, which can be viewed as approximations of the(linear) Kalman filter in the sense of the conditional expectation. However, to implement EKF and UKF, the exact dynamical model information and the statistical information of noise are still required. In this paper, we propose the recurrent neural-network based Kalman filter, where its Kalman gain is obtained via the proposed GRU-LSTM based neural-network framework that does not need the precise model information as well as the noise covariance information. By the proposed neural-network based Kalman filter, the state estimation performance is enhanced in terms of the tracking error, which is verified through various linear and nonlinear tracking problems with incomplete model and statistical covariance information.

• Korean Journal of Radiology
• /
• v.25 no.1
• /
• pp.74-85
• /
• 2024

Objective: Idiopathic intracranial hypertension (IIH) is a condition of unknown etiology associated with venous sinus stenosis. This study aimed to develop a magnetic resonance venography (MRV)-based radiomics model for predicting a high trans-stenotic pressure gradient (TPG) in IIH patients diagnosed with venous sinus stenosis. Materials and Methods: This retrospective study included 105 IIH patients (median age [interquartile range], 35 years [27-42 years]; female:male, 82:23) who underwent MRV and catheter venography complemented by venous manometry. Contrast enhanced-MRV was conducted under 1.5 Tesla system, and the images were reconstructed using a standard algorithm. Shape features were derived from MRV images via the PyRadiomics package and selected by utilizing the least absolute shrinkage and selection operator (LASSO) method. A radiomics score for predicting high TPG (≥ 8 mmHg) in IIH patients was formulated using multivariable logistic regression; its discrimination performance was assessed using the area under the receiver operating characteristic curve (AUROC). A nomogram was constructed by incorporating the radiomics scores and clinical features. Results: Data from 105 patients were randomly divided into two distinct datasets for model training (n = 73; 50 and 23 with and without high TPG, respectively) and testing (n = 32; 22 and 10 with and without high TPG, respectively). Three informative shape features were identified in the training datasets: least axis length, sphericity, and maximum three-dimensional diameter. The radiomics score for predicting high TPG in IIH patients demonstrated an AUROC of 0.906 (95% confidence interval, 0.836-0.976) in the training dataset and 0.877 (95% confidence interval, 0.755-0.999) in the test dataset. The nomogram showed good calibration. Conclusion: Our study presents the feasibility of a novel model for predicting high TPG in IIH patients using radiomics analysis of noninvasive MRV-based shape features. This information may aid clinicians in identifying patients who may benefit from stenting.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.1031-1042
• /
• 2023

In this study, a deep learning model was developed to predict the yield of cabbage and radish, one of the five major supply and demand management vegetables, using satellite images of Landsat 8. To predict the yield of cabbage and radish in Gangwon-do from 2015 to 2020, satellite images from June to September, the growing period of cabbage and radish, were used. Normalized difference vegetation index, enhanced vegetation index, lead area index, and land surface temperature were employed in this study as input data for the yield model. Crop yields can be effectively predicted using satellite images because satellites collect continuous spatiotemporal data on the global environment. Based on the model developed previous study, a model designed for input data was proposed in this study. Using time series satellite images, convolutional neural network, a deep learning model, was used to predict crop yield. Landsat 8 provides images every 16 days, but it is difficult to acquire images especially in summer due to the influence of weather such as clouds. As a result, yield prediction was conducted by splitting June to July into one part and August to September into two. Yield prediction was performed using a machine learning approach and reference models , and modeling performance was compared. The model's performance and early predictability were assessed using year-by-year cross-validation and early prediction. The findings of this study could be applied as basic studies to predict the yield of field crops in Korea.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.32 no.7
• /
• pp.1076-1081
• /
• 2003

The effects of dietary supplementation of JR-22, formula containing distilled extracts of traditional herbs on maximal exercise performance and endurance were evaluated in mouse and rat model. In acute forced swimming test with 4%∼8% of the body weight attached to the tail, it was shown that dietary JR-22 supplementation increased endurance in exercise performance. There was no change of blood lactic acid, ammonia, inorganic phosphorous ion and creatine kinase activity, however ATP concentration in muscle was increased by JR-22 supplementation. Also, insulin-like growth factor-l (IGF-1) concentration in blood was significantly increased by JR-22 supplementation. In addition, the oxidative damage induced by exercise was reduced by JR-22 supplementation. In these results, we suggested that JR-22 supplementation enhanced maximal endurance exercise performance by the mechanism of increasing ATP and IGF-1 concentration and reducing oxidative damage.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.4
• /
• pp.523-528
• /
• 2013

In recent years, many efforts are increasingly being made to conserve the natural environment with enhanced emission standards and air quality standards. Also there are various methods necessary to be researched to minimize the emission of air pollutants. In particular, boilers of industrial facilities are major portions of the air pollution. The front duct which needs to be designed to reduce the gases to the electrostatic precipitator requires a bent tube, a reduction/extend tube and an auxiliary equipment, that is, a guide vane. This paper proposes an optimum design of the guide vane by a case study for electrostatic precipitator's flow uniformity. The operating conditions of this study are as follows: BMCR (Boiler Maximum Continuous Rate) and MGR (Maximum Guaranteed Rate) are 75%, 50%, and 30%; turbulent fluid dynamics model is based upon K-${\varepsilon}$ formulation. Presentation of the computed motion of particles is found to be quite useful to predict the precipitator performance by use CFD (Computational Fluid Dynamics).

• Asia pacific journal of information systems
• /
• v.23 no.1
• /
• pp.107-132
• /
• 2013

Despite the importance of knowledge complementarities (KC) to firm performance, very little is known about exactly what constitutes KC and how synergistic knowledge is created in KC. This research looks into the dimensionality of KC and how synergistic knowledge as an essential component of KC is generated in a process innovation (PI) project. We propose that KC consists of potential complementary knowledge, collaborative elaboration (CE) process, and synergistic knowledge. The model is investigated quantitatively, using a sample of 26 matched-pairs of client and consultant who participated in a PI project, and then qualitatively using interviews of a sub-sample of 7 matched-pairs of client and consultant. Data were collected in a longitudinal way at four different points during the four month project period. Results show that consultant's learning about the client's business occurs first and then client learning about IT capabilities follows through CE. With this enhanced clients' knowledge about IT capabilities, clients play an initiative role in designing the To-Be business processes, while consultants play a supporting role by introducing best practices or making suggestions based on their experiences. Future research implications as well as practical implications are also discussed.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.452-457
• /
• 2003

An add-on type output regulator is proposed in this paper. By an add-on controller we mean an additional controller which operates harmonically with a pre-designed one. The role of the add-on controller is to reject a sinusoidal disturbance of unknown magnitude and phase but with known frequency. Advantages of the proposed controller include that (1) it can be used only when the performance of disturbance rejection needs to be enhanced, (2) when it is turned on or off, unwanted transient can be avoided (i.e., bumpless transfer), (3) it is designed for perfect disturbance rejection not just for disturbance reduction, (4) ability for perfect rejection is preserved even with uncertain plant model. This design may be promising for optical disc drive (ODD) systems in which disc eccentricity results in a sinusoidal disturbance. For ODD systems, the sensitivity function obtained by the pre-designed controller, which may have been designed by the lead-lag, $H_{\infty}$, or DOB (disturbance observer) technique, does not change much with the add-on controller except at the frequency of the disturbance. Since the add-on controller does the job of rejecting major eccentricity disturbance, the gain of the pre-designed controller does not have to be too high.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.3
• /
• pp.57-67
• /
• 2015

In this paper, we proposed the geolocation method using target detection information in infrared images. Our method was applied to geolocation system of hostile targets in ground-to-ground field. The major distortion that has bad effect of geolocation was composed of optic, topography, GPS(Global Positioning System) and IMU(Inertial Measurement Unit) of reconnaissance unit. We proposed enhanced geolocation method to cope with optic and topography distortion using polynomial fitting and slant-range calculation model to overcome earth curvature problem, and the result showed that the performance of our method was good for system requirements.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.9 no.1
• /
• pp.122-130
• /
• 2001

In order to improve the efficiency of the design process and the quality of the resulting design, this study proposes a design method for determining design variables of an automotive wheel-bearing unit of double-row angular-contact ball bearing type by using a genetic algorithm. The desired performance of the wheel-bearing unit is to maximize system life while satisfying geometrical and operational constraints without enlarging mounting spae. The use of gradient-based optimization methods for the design of the unit is restricted because this design problem is characterized by the presence of discrete design variables such as the number of balls and standard ball diameter. Therefore, the design problem of rolling element bearings is a constrained discrete optimization problem. A genetic algorithm using real coding and dynamic mutation rate is used to efficiently find the optimum discrete design values. To effectively deal with the design constraints, a ranking method is suggested for constructing a fitness function in the genetic algorithm. A computer program is developed and applied to the design of a real wheel-bearing unit model to evaluate the proposed design method. Optimum design results demonstrate the effectiveness of the design method suggested in this study by showing that the system life of an optimally designed wheel-bearing unit is enhanced in comparison with that of the current design without any constraint violations.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.2 s.314
• /
• pp.37-45
• /
• 2007

We propose an adaptive image enhancement method for epigraphic images, which is based on local statistics. Local statistics of the image are utilized for adaptive realization of the enhancement, that controls the contribution of the smoothing or sharpening paths. Image contrast enhancement occurs in details and noises are suppressed in smooth areas. For modeling the epigraphic image, pre~process is achieved by HSDI(Hanzi squeezed digital image). We have calculated the local statistics from this HSDI model. Application of this approach to HSDI has shown that processing not only smooths the background areas but also improves the subtle variations of edges, so that the word regions can be enhanced. Experimental results show that the proposed algorithm has better performance than the conventional image enhancement ones.