• Journal of Korean Society of Forest Science
• /
• v.104 no.1
• /
• pp.111-116
• /
• 2015

The purpose of this study was to evaluate mixed model of dbh-height relation containing random effect. Data were obtained from a survey site for Quercus variabilis in Gangwon region and remeasured the same site after three years. The mixed model were used to fixed effect in the dbh-height relation for Quercus variabilis, with random effect representing correlation of survey period were obtained. To verify the evaluation of the model for random effect, the akaike information criterion (abbreviated as, AIC) was used to calculate the variance-covariance matrix, and residual of repeated data. The estimated variance-covariance matrix, and residual were -0.0291, 0.1007, respectively. The model with random effect (AIC = -215.5) has low AIC value, comparison with model with fixed effect (AIC = -154.4). It is for this reason that random effect associated with categorical data is used in the data fitting process, the model can be calibrated to fit repeated site by obtaining measurements. Therefore, the results of this study could be useful method for developing model using repeated measurement.

• Journal of Korea Water Resources Association
• /
• v.54 no.5
• /
• pp.301-309
• /
• 2021

In this study, we tried to improve the performance of the existing U-net-based deep learning rainfall prediction model, which can weaken the meaning of time series order. For this, ConvLSTM2D U-Net structure model considering temporal consistency of data was applied, and we evaluated accuracy of the ConvLSTM2D U-Net model using a RainNet model and an extrapolation-based advection model. In addition, we tried to improve the uncertainty in the model training process by performing learning not only with a single model but also with 10 ensemble models. The trained neural network rainfall prediction model was optimized to generate 10-minute advance prediction data using four consecutive data of the past 30 minutes from the present. The results of deep learning rainfall prediction models are difficult to identify schematically distinct differences, but with ConvLSTM2D U-Net, the magnitude of the prediction error is the smallest and the location of rainfall is relatively accurate. In particular, the ensemble ConvLSTM2D U-Net showed high CSI, low MAE, and a narrow error range, and predicted rainfall more accurately and stable prediction performance than other models. However, the prediction performance for a specific point was very low compared to the prediction performance for the entire area, and the deep learning rainfall prediction model also had limitations. Through this study, it was confirmed that the ConvLSTM2D U-Net neural network structure to account for the change of time could increase the prediction accuracy, but there is still a limitation of the convolution deep neural network model due to spatial smoothing in the strong rainfall region or detailed rainfall prediction.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.1
• /
• pp.41-46
• /
• 2021

As data acquisition using unmanned aerial vehicles is widely used, as one of the ways to increase the accuracy of photogrammetry using unmanned aerial vehicles, a method of inputting both vertical and oblique images in bundle adjustment of aerial triangulation has been proposed. In this study, in order to find a suitable method for increasing the accuracy of photogrammetry, the accuracy of the case of adjusting the oblique images taken at different shooting angles and the case of adjusting the oblique images with different shooting angles at the same time with the vertical images were compared. As a result of the study, it was found that the error of the checkpoint decreases as the angle of the input oblique images increases. In particular, when the vertical images and oblique images are used together, the height error decreases significantly as the angle of the oblique images increases. The current 『Aerial Photogrammetry Work Regulation』 requires RMSE (Root Mean Square Error), which is the same as GSD (Ground Spatial Distance) of a vertical image. When using an oblique images with a shooting angle of 50°, a result close to this standard is obtained. If the vertical images and the 50° oblique images were adjusted at the same time, the work regulations could be satisfied. Using the results of this study, it is expected that photogrammetry using low-cost cameras mounted on unmanned aerial vehicles will become more active.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.2A
• /
• pp.115-120
• /
• 2011

The strength of reinforced concrete members has uncertainty from material properties of, concrete and reinforcements, section dimensions, and construction errors and so on. The accurate evaluation of these uncertainties is necessary to assure the reasonable safety. The uncertainties should be taken into account in design using structural reliability theory which requires probabilistic models for such uncertainties. In current Korean design code, most reliability evaluations were performed based on foreign data because of lack of local data. In this paper, the probabilistic models for yield strength of reinforcements were developed based on local data. The effects of various factors, nominal yield strength, diameter of reinforcements, and companies, on the models are also examined. According to data analysed, the effects of those factors are not significant. The probability model for yield strength of reinforcements in Korea can be expressed with Beta distribution based on collected data.

• Journal of Korean Society for Geospatial Information Science
• /
• v.21 no.4
• /
• pp.117-123
• /
• 2013

The purpose of this research is to suggest the reasonable method of Network RTK surveying in future cadastral re-survey project through the accuracy analysis about Network RTK surveying achievement and the conventional TS-based confirmation surveying. To achieve it, we selected the experiment places and succeeded in achieving the result by Network RTK surveying about total of 307 parcel boundary point. We compared it with the result of confirmation surveying for cadastral, and it was shown that total connection errors of RMSE was ${\pm}0.1028m$ and total 48 places exceeded in the cadastral re-survey allowable error tolerance. The research suggested the practical alternatives in cadastral re-survey project after the comprehensive evaluation of those analysis results. Therefore, the author suggested development and adoptation of integrated electronic plane table surveying method. Moreover, we suggested unifying the first parcel boundary point method into the total station surveying and adopt the Network RTK surveying on the cadastral surveying inspection.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.39 no.2
• /
• pp.157-165
• /
• 2002

This paper presents the implementation of the speed control system for 3 phase induction motor using PD controller and neural networks. The PD controller is used to control the motor and to train neural networks at the first time. And neural networks are widely used as controllers because of a nonlinear mapping capability, we used feedforward neural networks(FNN) in order to simply design the speed control system of the 3 phase induction motor. Neural networks are tuned online using the speed reference, actual speed measured from an encoder and control input current to motor. PD controller and neural networks are applied to the speed control system for 3 phase induction motor, are compared with PI controller through computer simulation and experiment respectively. The results are illustrated that the output of the PD controller is decreased and feedforward neural networks act main controller, and the proposed hybrid controllers show better performance than the PI controller in abrupt load variation and the precise control is possible because the steady state error can be minimized by training neural networks.

• Journal of IKEEE
• /
• v.20 no.2
• /
• pp.188-191
• /
• 2016

In this paper, we propose energy-efficient automatic power controller which can power on and off the signboard at the specified light intensity using the Illuminance Detector. By using segmented section Classification algorithm, light intensity setup system propose variable resistor method which makes users more easy to control. Automatic light on-off system set a standard by measured illuminance data. Measured light-intensity through the Illuminance Detector are communicated with the signboard power controller with wireless communication, and it controls lighting system. In this paper, we evaluated the Energy-Efficient Automatic Power Controller of The Signboard using illuminance detector. Experimental results in lightless environment shows that the error rate is less than 3% by Accredited Testing Laboratories.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.555-562
• /
• 2019

This paper presents soil deformation measurement in model chamber based on photogrammetry. We created an aluminum framed acrylic model chamber with soil inside and applied photogrammetry to measure soil deformation caused by loading tests. The soil consists of 40% black and 60% regular sand to create image contrast in soil images. In preprocessing, the self camera calibration was carried out for IOPs (Interior Orientation Parameters), followed by the space resection to estimate EOPs (Exterior Orientation Parameters) using control points located along the aluminum frame. Image matching was applied to measure the soil displacement. We tested different matching window sizes and the effect of image smoothing. Experimental results showed that 65x65 pixels of window size produced better soil deformation map and the image smoothing was useful to suppress the matching outliers. In conclusion, photogrammetry was able to efficiently generated soil deformation map.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.14 no.1
• /
• pp.94-101
• /
• 2000

In this paper, a multi-channel active noise control(ANC) system using subband hybrid control techniques is proposed. Subband techniques could reduce computational burden and improve the performance of ANC systems by dividing several frequency subband and adjusting adaptive filter coefficients. So it can effectively cancel noises at wanted frequency range and use lower order adaptive filter than the existing algorithms. The adjoint LMS algorithm, which prefilter the error signals instead of the divided reference signals in frequency band, is also used for adaptive filter algorithms to reduce the computational burden of the subband adaptive systems. To improve performance of the ANC system, a weighted hybrid control technique, which has weightily properties of feedforward control systems and feedback control systems, is applied. This algorithm shows higher stability and good noise attenuation property in broad band ANC systems. Computer simulations were performed to show the effectiveness of the proposed algorithm.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.1
• /
• pp.59-65
• /
• 2012

In this paper, we proposed a method for automatic detection of a updating object from spatial data sets of different scale and updating cycle by using areal feature matching based on shape similarity. For this, we defined a updating object by analysing matching relationships between two different spatial data sets. Next, we firstly eliminated systematic errors in different scale by using affine transformation. Secondly, if any object is overlaid with several areal features of other data sets, we changed several areal features into a single areal feature. Finally, we detected the updating objects by applying areal feature matching based on shape similarity into the changed spatial data sets. After applying the proposed method into digital topographic map and a base map of Korean Address Information System in South Korea, we confirmed that F-measure is highly 0.958 in a statistical evaluation and that significant updating objects are detected from a visual evaluation.