• 인터넷정보학회논문지
• /
• 제19권4호
• /
• pp.35-44
• /
• 2018

Monitoring through Synthesis Aperture Radar (SAR) is responsible for marine safety from floating icebergs. However, there are limits to distinguishing between icebergs and ships in SAR images. Convolutional Neural Network (CNN) is used to distinguish the iceberg from the ship. The goal of this paper is to increase the accuracy of identifying icebergs from SAR images. The metrics for performance evaluation uses the log loss. The two-layer CNN model proposed in research of C.Bentes et al.[1] is used as a benchmark model and compared with the four-layer CNN model using data augmentation. Finally, the performance of the final CNN model using the VGG-16 pre-trained model is compared with the previous model. This paper shows how to improve the benchmark model and propose the final CNN model.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2004년도 추계학술대회 및 정기총회
• /
• pp.407-429
• /
• 2004

As documents in webs are increasing explosively due to the rapid development of electronic documents, an efficient system classifying documents automatically is required. In this study, a new document classification method, which is called Document Finger Print Method, is suggested to classify web documents automatically and efficiently. The performance of the suggested method is evaluated alone with other existing methods such as key words based method, weighted key words based method, neural networks, and decision trees. An experiment is designed with 10 documents categories and 59 randomly selected words. The result shows that the suggested algorithm has a superior classifying performance compared to other methods. The most important advantage of this method is that the suggested method works well without the size limits of the number of words in documents.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.2059-2062
• /
• 2002

For detection of digitally modulated signals, the receiver must be provide with accurate carrier phase and symbol timing estimates. So far, tots of algorithms have been suggested for those purposes. In general, a interpolation filter with TED(Timing Error Detection) like Gardner algorithm is popularly used for symbol timing estimate of digital communication receiver. Apart from the performance point of view, a multiplicative operation of any interpolation filter limits the symbol rate of the system. Hence, we suggest a new symbol timing estimate algorithm for high speed burst-mode fixed wireless communication system and analyze its performance in the AWGN channel.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 춘계학술대회 학술발표 논문집
• /
• pp.81-84
• /
• 1998

Genetic Algorithms (GAs) are stochastic algorithms whose search methods model some natural phenomena. The procedure of GAs may be divided into two sub-procedures : Operation and Selection. Chromosomes can produce new offspring by means of operation, and the fitter chromosomes can produce more offspring than the less fit ones by means of selection. However, operation which is executed randomly and has some limits to its execution can not guarantee to produce fitter chromosomes. Thus, we propose a method which gives a directional information to the genetic operator by reinforcement learning. It can be achived by using neural networks to apply reinforcement learning to the genetic operator. We use the amount of fitness change which can be considered as reinforcement signal to calcualte the error terms for the output units. Then the weights are updated using backpropagtion algorithm. The performance improvement of GAs using reinforcement learning can be measured by applying the pr posed method to GA-hard problem.

• Journal of Information Processing Systems
• /
• 제7권1호
• /
• pp.63-74
• /
• 2011

As a powerful and flexible processor, the Graphic Processing Unit (GPU) can offer a great faculty in solving many high-performance computing applications. Sweep3D, which simulates a single group time-independent discrete ordinates (Sn) neutron transport deterministically on 3D Cartesian geometry space, represents the key part of a real ASCI application. The wavefront process for parallel computation in Sweep3D limits the concurrent threads on the GPU. In this paper, we present multi-dimensional optimization methods for Sweep3D, which can be efficiently implemented on the finegrained parallel architecture of the GPU. Our results show that the overall performance of Sweep3D on the CPU-GPU hybrid platform can be improved up to 4.38 times as compared to the CPU-based implementation.

• Journal of Mechanical Science and Technology
• /
• 제15권11호
• /
• pp.1517-1523
• /
• 2001

The effect of power supply voltage on the performance limits in a laboratory magneto-rheological fluid based device was identified by experiments. It suggests that the frequency range of motion for control is limited by the voltage attenuation due to the coil inductance and the maximum power supply voltage set for practical use of an MRF devices. In this work, the magnetic and electrical characteristics of the MRF device are investigated and a design procedure is formulated to achieve the desired performance for a given power supply.

• Bulletin of the Korean Chemical Society
• /
• 제33권12호
• /
• pp.3971-3976
• /
• 2012

A simple method was developed for the analysis of seven stilbene-type fluorescent whitening agents (FWAs) in paper materials by ion-pair reversed-phase high-performance liquid chromatography with fluorescence detection. These stilbene-type FWAs included two disulfonate, two tetrasulfonate, and three hexasulfonate compounds. After optimization of chromatographic conditions, the FWAs were satisfactorily separated using a reversed-phase column (RP-18) with the following isocratic mobile phase: methanol-water (60:40) containing 17.5 mM TBABr and 10 mM citrate buffer (pH = 7.0). The calibration plot was linear in the range from 5 to 500 ng/mL for two disulfo-FWAs and from 1 to 500 ng/mL for the other five FWAs. Precision levels of the calibration curve as indicated by RSD of response factors were 1.2 and 8.1%. Limits of quantitation (LOQ) ranged from 1.2 to 11 ng/mL.

• Structural Engineering and Mechanics
• /
• 제12권6호
• /
• pp.573-594
• /
• 2001

A two-level displacement-based design procedure is developed. To obtain the displacement demands, elastic spectra for occasional earthquakes and inelastic spectra for rare earthquakes are used. Minimum global stiffness and strength to be supplied to the structure are based on specified maximum permissible drift limits and on the condition that the structure responds within the elastic range for occasional earthquakes. The performance of the structure may be assessed by an inelastic push-over analysis to the required displacement and the evaluation of damage indices. The approach is applied to the design of a five-story reinforced concrete coupled wall structure located in the most hazardous seismic region of Argentina. The inelastic dynamic response of the structure subjected to real and artificially generated acceleration time histories is also analyzed. Finally, advantages and limitations of the proposed procedure from the conceptual point of view and practical application are discussed.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2002년도 추계학술대회 및 정기총회
• /
• pp.118-121
• /
• 2002

The problem of maneuvering target tracking has been studied in the field of the state estimation over decades. The Kalman filter has been widely used to estimate the state of the target, but in the presence of a maneuver, its performance may be seliously degraded. In this paper, to solve this problem and track a maneuvering target effectively, DNA coding-based intelligent Kalman filter (DNA coding-based IKF) is proposed. The proposed method can overcome the mathematical limits of conventional methods and can effectively track a maneuvering target with only one filter by using the fuzzy logic based on DNA coding method. The tracking performance of the proposed method is compared with those of the adaptive interacting multiple model (AIMM) method and the GA-based IKF in computer simulations.

• 대한산업공학회지
• /
• 제21권2호
• /
• pp.255-265
• /
• 1995

Due to advances in automated manufacturing systems and automatic inspection equipment, complete inspection has drown increased attention recently and has become a widespread practice. In a complete inspection plan all of the items are subject to acceptance inspection If an item fails to meet the predetermined specifications, it is rejected. In this paper, economic complete inspection plans are developed in situations where rejected items are reworked Complete inspections based on the performance variable of interest or a variable which is correlated with the performance variable are considered. Cost models are constructed which involve cost incurred by imperfect quality rework cast and quality inspection cost Methods of finding the optimal complete inspection plans are presented and numerical examples are given.