• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.714-720
• /
• 2021

The ITL (information theoretic learning) based on the kernel density estimation method that has successfully been applied to machine learning and signal processing applications has a drawback of severe sensitiveness in choosing proper kernel sizes. For the maximization of correntropy criterion (MCC) as one of the ITL-type criteria, several methods of adapting the remaining kernel size ( ) after removing the term have been studied. In this paper, it is shown that the main cause of sensitivity in choosing the kernel size derives from the term and that the adaptive adjustment of in the remaining terms leads to approach the absolute value of error, which prevents the weight adjustment from continuing. Thus, it is proposed that choosing an appropriate constant as the kernel size for the remaining terms is more effective. In addition, the experiment results when compared to the conventional algorithm show that the proposed method enhances learning performance by about 2dB of steady state MSE with the same convergence rate. In an experiment for channel models, the proposed method enhances performance by 4 dB so that the proposed method is more suitable for more complex or inferior conditions.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.3
• /
• pp.11-20
• /
• 2022

Currently, the black-white QR (Quick Response) codes have been used widely in consumer advertising fields and the study of color QR codes have received a growing demand because of much higher data encoding capacity. Color QR codes can be reproduced by the printing and scanning processes, however, these encounter colors distortion caused by insufficient lighting, low resolution of camera and geometric deformation during the capturing processes. In order to overcome these problems, this paper proposes an efficient decoding algorithm for color QR codes with inserted patterns, which are dealt with conventional studies. These are evaluated in view of the recognition rate under different noise conditions, for example, Gaussian noises/blurring and geometric deformation. Experimental results demonstrate that the color QR codes with simple pattern can resist the distortion of Gaussian noises/blurrings.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.199-206
• /
• 2021

This paper describes the feasibility and reliability verification of installing a protective structure to protect attack periscopes. The attack periscope is the critical equipment of a submarine to enable the user to monitor surface and air activity, collect navigational data, and detect and identify targets. The attack periscope provides target information acquired through TV, IR camera, and laser range finder to the combat system. In the product improvement program, the upper part of the masts was exposed to the outside of the sail because the existing attack periscope was replaced with a new one. On the other hand, the head sensor can be damaged by floating objects, such as fishing nets, during sea navigation. Therefore, the installation of a protective structure for an attack periscope improved the equipment operation performance. The feasibility and reliability of the installation of the protective structure were verified by examining the influence of URN.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.2
• /
• pp.15-25
• /
• 2024

For effective analysis of animal ecosystems, technology that can automatically identify the current status of animal habitats is crucial. Specifically, animal sound classification, which identifies species based on their sounds, is gaining great attention where video-based discrimination is impractical. Traditional studies have relied on a single deep learning model to classify animal sounds. However, sounds collected in outdoor settings often include substantial background noise, complicating the task for a single model. In addition, data imbalance among species may lead to biased model training. To address these challenges, in this paper, we propose an animal sound classification scheme that combines predictions from multiple models using Focal Loss, which adjusts penalties based on class data volume. Experiments on public datasets have demonstrated that our scheme can improve recall by up to 22.6% compared to an average of single models.

• 전자공학회논문지 IE
• /
• v.45 no.1
• /
• pp.58-64
• /
• 2008

This Paper provides perceptual metrics for video quality based on properties of human visual system, and audio quality based on human audition. All metrics work without reference signals, allowing non-intrusive, in-service measurements. A simple and easy-to-learn user interface displays the metrics and saves them in popular file formats like CSV. In this paper, proposed method was able to various and corrective measurement for the multimedia service video quality. As that it was able to application to set up service guide line and the methode of measurement and system for the set up standardization of the high quality video service.

• Korean Journal of Computational Design and Engineering
• /
• v.20 no.2
• /
• pp.193-209
• /
• 2015

For reasons such as global warming, depletion of fossil fuels and the danger of nuclear energy the research and development of renewable energy is actively underway. Wind energy has advantages over another renewable energy in terms of location requirements, energy efficiency and reliability. Nowadays the research and development area is expanded to offshore because it can supply more wind reliability and free from noise pollution. In this study, the monopile offshore wind turbine transportation and installation (T&I) process are investigated. In addition, the schedule and cost for the process are estimated by discrete event simulation. For the simulation, simulation models for various means of T&I are developed. The optimum T&I execution plan with shortest duration and lowest cost can be found by using different mission start day and T&I means.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.7
• /
• pp.56-66
• /
• 1999

Thermoacoustic oscillation is a significant problem in cylindrical-type combustors such as common internal combustion engines, industrial furnaces, gas turbine, etc. This kind of low frequency oscillation can lead to serious consequences such as destruction of the combustor and production of strong noise. The accurate numerical simulation of thermoacoustic phenomena is a complex and challenging problem, especially when considering the chemical reaction of mixtures. As with other simulations of aerodynamics and aeroacoustics, the direct computation of thermoacoustic phenomena requires that Navier-Stokes equations be solved using accurate numerical differentiation and time-marching schemes, with non-reflecting boundary conditions. The numerical approach used here aims at qualitative analysis and efficient prediction of those problems, not at the development of an accurate scheme. The numerical prediction developed in this work is shown to be reasonably matched with experimental result.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.687-701
• /
• 2020

In the production of industrial fabric, it needs automatic real-time system to detect defects on the fabric for assuring the defect-free products flow to the market. At present, many visual-based methods are designed for detecting the fabric defects, but they usually lead to high false alarm. Base on this reason, we propose a Sobel operator combined with patch statistics (SOPS) algorithm for defects detection. First, we describe the defect detection model. mean filter is applied to preprocess the acquired image. Then, Sobel operator (SO) is applied to deal with the defect image, and we can get a coarse binary image. Finally, the binary image can be divided into many patches. For a given patch, a threshold is used to decide whether the patch is defect-free or not. Finally, a new image will be reconstructed, and we did a loop for the reconstructed image to suppress defects noise. Experiments show that the proposed SOPS algorithm is effective.

• Advances in materials Research
• /
• v.6 no.1
• /
• pp.79-91
• /
• 2017

In the present work, a design of experiment (DOE) technique using Taguchi method, has been applied to optimize the properties of ODS tungsten heavy alloys(WHAs). In this work Taguchi method involves nine experiments groups for four processing parameters (compaction pressure, sintering temperature, binding material type, and oxide type) with three levels was implemented. The signal-to-noise (S/N) ratio and analysis of variance (ANOVA) were employed to obtain the optimal process parameter levels and to analyze the effect of these parameters on density, electrical conductivity, hardness and compressive strength values. The results showed that all the chosen factors have significant effects on all properties of ODS tungsten heavy alloys samples. The density, electrical conductivity and hardness increases with the increase in sintering temperature. The analysis of the verification experiments for the physical properties (density and Electrical conductivity) has shown that Taguchi parameter design can successfully verify the optimal parameters, where the difference between the predicted and the verified values of relative density and electrical conductivity is about 1.01% and 1.15% respectively.

• Journal of Magnetics
• /
• v.21 no.3
• /
• pp.330-339
• /
• 2016

A magnetic field sensing system with a single primary sensor and multiple reference sensors deployed locally and orthogonally, was proposed for downlink signal reception and interference cancelling for Through-the-Earth Communication (TEC). This paper mathematically analyzes a design optimization process for a search coil magnetometer (SCM), and applies that process to minimize the bandwidth of the primary SCM for TEC signal reception and the volume of reference SCMs for multiple distributions. The primary SCM achieves a 3-dB bandwidth of 7 Hz, a sensitivity threshold of 120 fT/${\surd}$Hz, and a volume of $2.32{\times}10^{-4}m^3$. The entire sensing system volume is as small as $10^{-2}m^3$. Experiments with interference from industrial frequency harmonics demonstrated an average of 36 dB and 18 dB improvements in signal-to-interference ratio and signal-to-interference plus noise ratio, respectively, using multichannel recursive-least-squares algorithm. Thus, the proposed sensing system can reduce the interference effectively and allows reliable downlink signal reception.