• Journal of the Semiconductor & Display Technology
• /
• v.19 no.3
• /
• pp.12-18
• /
• 2020

In semiconductor manufacturing, defect detection is critical to maintain high yield. Currently, computer vision systems used in semiconductor photo lithography still have adopt to digital image processing algorithm, which often occur inspection faults due to sensitivity to external environment. Thus, we intend to handle this problem by means of using Mask R-CNN instead of digital image processing algorithm. Additionally, Mask R-CNN can be trained with image dataset pre-processed by means of the specific designed digital image filter to extract the enhanced feature map of Convolutional Neural Network (CNN). Our approach converged advantage of digital image processing and instance segmentation with deep learning yields more efficient semiconductor photo lithography inspection system than conventional system.

• Journal of information and communication convergence engineering
• /
• v.3 no.4
• /
• pp.184-186
• /
• 2005

The effect of power supply noise on the phase noise of a ring oscillator is studied. The power supply noise source in series with DC power supply voltage is applied to a 3 stage CMOS ring oscillator. The phase noise due to the power supply noise is modeled by the narrow band phase modulation. The model is verified by the fact that the spectrum of output of ring oscillator has two side bands at the frequencies offset from the frequency of the ring oscillator by the frequency of the power supply noise source. Simulations at several different frequency of the power supply noise reveals that the ring oscillator acts as a low pass filter to the power supply noise. This study, as a result, shows that the phase noise generated by the power supply noise is inversely proportional to the frequency offset from the carrier frequency.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1475-1478
• /
• 2002

This paper presents a performance analysis of an improved adaptive algorithm proposed by the authors recently. It is based on the normalized least mean square (NLMS) algorithm, which Is one of the major techniques to adapt the cofficients of a transversal filter. Generally, the performance of an adaptive algorithm is often discussed by investigating the mis-adjustment. In this paper, unlike these approaches, a novel analytical method is considered. letting the parameters so that the residual mean square error (MSE) after the convergence of the algorithm is equal to that of the NLMS algorithm, the MSE level is compared. It is shown that the theoretical analysis is agreed with the simulation results.

• Journal of information and communication convergence engineering
• /
• v.8 no.4
• /
• pp.405-410
• /
• 2010

Image processing operations like smoothing and edge detection, and many more are very widely used in areas like Computer Vision. We classify the image processing domain as seven branches-image acquirement and output, image coding and compression, image enhancement and restoration, image transformation, image segmentation, image description, and image recognition and description. We implemented algorithms of gaussian smoothing, laplace sharpening, image contrast effect, image black and white effect, image fog effect, image bright and dark effect, image median filter, and canny edge detection. Such experimental results show the figures respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.5C
• /
• pp.521-526
• /
• 2009

In this paper, a complex-valued recursive least squares escalator filter algorithm with reduced computational complexity for complex-valued signal processing applications is presented. The local tap weight of RLS-ESC algorithm is updated by incrementing its old value by an amount equal to the local estimation error times the local gain scalar, and for the gain scalar, the local input autocorrelation is calculated at the previous time. By deriving a new gain scalar that can be calculated by using the current local input autocorrelation, reduced computational complexity is accomplished. Compared with the computational complexity of the complex-valued version of RLS-ESC algorithm, the computational complexity of the proposed method can be reduced by 50% without performance degradation. The reduced computational complexity of the proposed algorithm is even less than that of the LMS-ESC. Simulation results for complex channel equalization in 64QAM modulation schemes demonstrate that the proposed algorithm has superior convergence and constellation performance.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.15 no.3
• /
• pp.809-823
• /
• 1991

본 연구에서는 Wiener 필터링 이론에 의하여 소음원의 입력신호에 대한 최적 한 예측을 할 수 있는 최적예측기(optimal predictor)로써 부가적인 음을 발생시키고 입력신호 및 출력신호 간의 차인 오차를 최소화시키도록 하는 적응신호처리방법에 대 해 설명하고 이러한 적응 신호처리 방법을 이용한 능동 소음 제어 방법을 제시하였다. 이와 아울러 제어계의 환경 변화에 따른 파라메타의 변화에 적응적으로 응답이 가능해 야 하는 적응 소음 제어계에서, 음향궤환과 함께 필히 고려해야하는 부가적인 전달함 수-모델과 스피커를 포함하는 보조경로 및 오차미이크로폰을 포함하는 오차경로의 전 달함수의 영향을 고려한 능동소음제어에 대해 연구하였다.

• Structural Engineering and Mechanics
• /
• v.27 no.1
• /
• pp.49-61
• /
• 2007

This paper presents a modified and improved bi-directional evolutionary structural optimization (BESO) method for topology optimization. A sensitivity filter which has been used in other optimization methods is introduced into BESO so that the design solutions become mesh-independent. To improve the convergence of the optimization process, the sensitivity number considers its historical information. Numerical examples show the effectiveness of the modified BESO method in obtaining convergent and mesh-independent solutions. A study of the effects of various BESO parameters on the solution is then conducted to determine the appropriate values for these parameters.

• Journal of information and communication convergence engineering
• /
• v.13 no.2
• /
• pp.113-122
• /
• 2015

Feature subset selection is as a pre-processing step in learning algorithms. In this paper, we propose an efficient algorithm, ModifiedFAST, for feature subset selection. This algorithm is suitable for text datasets, and uses the concept of information gain to remove irrelevant and redundant features. A new optimal value of the threshold for symmetric uncertainty, used to identify relevant features, is found. The thresholds used by previous feature selection algorithms such as FAST, Relief, and CFS were not optimal. It has been proven that the threshold value greatly affects the percentage of selected features and the classification accuracy. A new performance unified metric that combines accuracy and the number of features selected has been proposed and applied in the proposed algorithm. It was experimentally shown that the percentage of selected features obtained by the proposed algorithm was lower than that obtained using existing algorithms in most of the datasets. The effectiveness of our algorithm on the optimal threshold was statistically validated with other algorithms.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.35 no.2
• /
• pp.57-64
• /
• 1993

A modified kinematic wave model of the overland flow in vegetated filter strips was developed. The model can predict both flow depth and hydraulic radius of the flow. Existing models can predict only mean flow depth. By using the hydraulic radius, erosion, deposition and flow's transport capacity can be more rationally computed. Spacing hydraulic radius was used to compute flow's hydraulic radius. Numerical solution of the model was accomplished by using both a second-order nonlinear scheme and a linear solution scheme. The nonlinear portion of the model ensures convergence and the linear portion of the model provides rapid computations. This second-order nonlinear scheme minimizes numerical computation errors that may be caused by linearization of a nonlinear model. This model can also be applied to golf courses, parks, no-till fields to route runoff and production and attenuation of many nonpoint source pollutants.

• Journal of Wellbeing Management and Applied Psychology
• /
• v.4 no.3
• /
• pp.1-9
• /
• 2021

Purpose: This study was conducted to develop a module with higher removal efficiency and effectiveness by adapting two or more deodorization techniques for main cause of odor pollution exposed citizen living near water treatment facilities. Research design, data and methodology: To consider the standard, unity, electrical wire, compatibility of detachable device by installing two types of dry deodorization device within one module for easy replacement. Complex odor, H₂S, NH₃ were collected from sewage treatment facilities for evaluation of deodorization device. Results: Using the developed application in this study, removal efficiency of complex odor, H₂S, NH₃ were 93%, 100%, 82%, respectively. Conclusions: The H₂S removal efficiency of deodorization device was higher than bio-filter system, which were currently used by sewage treatment. Further, the device should be considered for use in efficient odor removal system.