• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.161-166
• /
• 2018

In this paper, we propose a method for the random selection of pooling operations for the regularization and reduction of cross validation in convolutional neural networks. The pooling operation in convolutional neural networks is used to reduce the size of the feature map and for its shift invariant properties. In the existing pooling method, one pooling operation is applied in each pooling layer. Because this method fixes the convolution network, the network suffers from overfitting, which means that it excessively fits the models to the training samples. In addition, to find the best combination of pooling operations to maximize the performance, cross validation must be performed. To solve these problems, we introduce the probability concept into the pooling layers. The proposed method does not select one pooling operation in each pooling layer. Instead, we randomly select one pooling operation among multiple pooling operations in each pooling region during training, and for testing purposes, we use probabilistic weighting to produce the expected output. The proposed method can be seen as a technique in which many networks are approximately averaged using a different pooling operation in each pooling region. Therefore, this method avoids the overfitting problem, as well as reducing the amount of cross validation. The experimental results show that the proposed method can achieve better generalization performance and reduce the need for cross validation.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.9
• /
• pp.217-224
• /
• 2019

Recently, sensor networks are built and used on many kinds of fields such as home, traffic, medical treatment and power grid. Sensing data manipulation on these fields could be a serious threat on property and safety. Thus, a proper way to block sensing data manipulation is necessary. In this paper, we propose IoT(Internet of Things) sensing data validation mechanism based on data obfuscation and variance analysis to remove manipulated sensing data effectively. IoT sensor device modulates sensing data with obfuscation function and sends it to a user. The user demodulates received data to use it. Fake data which are not modulated with proper obfuscation function show different variance aspect with valid data. Our proposed mechanism thus can detect fake data by analyzing data variance. Finally, we measured data validation time for performance analysis. As a result, block rate for false data was improved by up to 1.45 times compared with the existing technique and false alarm rate was 0.1~2.0%. In addition, the validation time on the low-power, low-performance IoT sensor device was measured. Compared to the RSA encryption method, which increased to 2.5969 seconds according to the increase of the data amount, the proposed method showed high validation efficiency as 0.0003 seconds.

• Journal of Korean Society for Atmospheric Environment
• /
• v.21 no.E1
• /
• pp.37-39
• /
• 2005

It is well perceived that micrometeorological approach is one of the most reliable method for the quantification of vertical fluxes of trace components in the atmosphere. In this study, the feasibility of relaxed eddy accumulation (REA) method is discussed with respect to its reliability in the field application. Knowing that the use of micrometeorological approaches requires validation of analytical uncertainties involved, the problems and issues associated with its application are discussed to stimulate the proper employment of such technique in the field study.

• Proceedings of the KIPE Conference
• /
• 1999.07a
• /
• pp.282-286
• /
• 1999

This paper presents a vector control of parallel drive, a beatless control and a low switching PWM technique for the propulsion system of Electric car as transient state which are included interrupting line voltage, changing line voltage slowly, suddenly, regenerating light load and starting from backward running. Improved performance and a validation of proposed method is shown by the experimental results using a 1.65MVA IGBT VVVF inverter and inertia load equivalent to 160 tons electric cars through a running and transient state test.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.30A no.11
• /
• pp.48-57
• /
• 1993

The ANSI recently adopted normalized site attenuation(NSA) as a validation procedure of open area test sites. In this paper, the theoretical NSA values specified in the truth vlues of NSA were calculated for an ideal test site when tuned dipoles are used. A method of moments technique is employed to evaluate these NSA values. The calculated NSA values are more accurate than that of the ANSI.

• Proceedings of the KIEE Conference
• /
• 1998.07c
• /
• pp.868-870
• /
• 1998

This is Part II of a two part paper dealing with control design in petter systems using the method of normal forms. The companion paper (part I) [1] describes the general theory and control procedure. This paper depicts numeral results to show the validation of the proposed method and to observe the effects of controller setting changes.

• Journal of the Korean Data and Information Science Society
• /
• v.9 no.1
• /
• pp.77-88
• /
• 1998

Smoothing spline estimators are modified to remove boundary bias effects using the technique proposed in Eubank and Speckman (1991). An O(n) algorithm is developed for the computation of the resulting estimator as well as associated generalized cross-validation criteria, etc. The asymptotic properties of the estimator are studied for the case of a linear smoothing spline and the upper bound for the average mean squared error of the estimator given in Eubank and Speckman (1991) is shown to be asymptotically sharp in this case.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.606-608
• /
• 2003

This paper investigates the accuracy of three-dimensional positioning for EROS A stereo pairs when different numbers of ground control points are employed. The major works of the proposed schemes include: (1) initialization of orientation parameters (2) preliminary orbit fitting, (3) orbit refinement using the least squares filtering technique, and (4) space intersection. The experiment includes validation of positioning accuracy for an EROS A in-track stereo pair when different number of check points are employed.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.21 no.7
• /
• pp.939-947
• /
• 1997

An experimental study of the skin friction measurement in a turbulent boundary-layer has been carried out. The skin friction measurements are made using the laser interferometer skin friction (LISF) meter, which optically detects the rate of thinning of an oil applied to the test surface. This technique produces reliable skin friction data over a wide range of flow situations up to 3-dimensional complicated flows with separation, where traditional skin friction measurement techniques are not applicable. The present measured data in a turbulent boundary-layer on a flat plate using the LISF technique shows a good comparison with the result from the previous velocity profile techniques, which proves the validity of the present technique. An extensive error analysis is carried out for the present technique yielding an uncertainty of about .+-.8%, which makes them suitable for CFD code validation purposes. Finally the measurements of the skin friction in a separated region after a surface-mounted obstacle are also presented.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1996.04a
• /
• pp.361-369
• /
• 1996

The accurate prediction of modal parameters of a rotating blade is an important requirement in the assessment of the dynamics of a helicopter rotor. Indeed, predictions of flight loads and stability are normally dependent on initially predicting the undamped mode shapes. A measuring technique, known as Strain Pattern Analysis (SPA), appears to be the most successful technique for measuring the mode shapes of rotating blades. This method was developed to be used on actual aircraft so no attempt was made to measure rotating mode shapes directly in order to validate the SPA method. This report summarizes results from experimental investigations which were carried out to validate the SPA method for the prediction of aerodynamically damped modes of a rotating blade. A series of modal tests were carried out on two rotor models in which the non-rotating, undamped and aerodynamically damped rotating modes were measured directly (strain and displacement patterns). It is shown that the SPA method to be very successful in itself but there are a number of limitations in validating this technique. To provide data which could be used to confidently validate theoretical prediction codes, existing limitations should be addressed.