• Journal of information and communication convergence engineering
• /
• v.13 no.3
• /
• pp.215-219
• /
• 2015

Cells in a culture process transform with time and produce many overlapping cells in their vicinity. We are interested in a separation algorithm for images of overlapping cells taken using a fluorescence optical microscope system during a cell culture process. In this study, all cells are assumed to have an ellipse-like shape. For an ellipse fitting-based method, an improved least squares method is used by decomposing the design matrix into quadratic and linear parts for the separation of overlapping cells. Through various experiments, the improved least squares method (numerically stable direct least squares fitting [NSDLSF]) is compared with the conventional least squares method (direct least squares fitting [DLSF]). The results reveal that NSDLSF has a successful separation ratio with an average accuracy of 95% for two overlapping cells, an average accuracy of 91% for three overlapping cells, and about 82% accuracy for four overlapping cells.

• ETRI Journal
• /
• v.38 no.3
• /
• pp.510-517
• /
• 2016

Recent developments in the field of separation of mixed signals into music/voice components have attracted the attention of many researchers. Recently, iterative kernel back-fitting, also known as kernel additive modeling, was proposed to achieve good results for music/voice separation. To obtain minimum mean square error (MMSE) estimates of short-time Fourier transforms of sources, generalized spatial Wiener filtering (GW) is typically used. In this paper, we propose an advanced music/voice separation method that utilizes a generalized weighted ${\beta}$-order MMSE estimation (WbE) based on iterative kernel back-fitting (KBF). In the proposed method, WbE is used for the step of mixed music signal separation, while KBF permits kernel spectrogram model fitting at each iteration. Experimental results show that the proposed method achieves better separation performance than GW and existing Bayesian estimators.

• Proceedings of the KSR Conference
• /
• 2007.05a
• /
• pp.219-222
• /
• 2007

The Full-Off Fitting of Movable Bracket using conventional line have received a dynamic contact impact between contact wire and pantograph of electric rolling stock, so this impact cause a rapid vibration of foreword and back near the supporting point of movable bracket. The Full-Off Fitting on Movable Bracket in conventional line differs from that in high speed line. Now conventional line have required a speed up for new electric locomotive and electric car. In speed-up track and big impact place, the Full-Off Fitting and the dropper often get demage. We have done a study of buffering function to solve that problem in conventional line. We have find a need adopting buffer fitting. It is decrease a dynamic impact between contact wire and pantograph of electric rolling stock.

• Journal of the Society of Naval Architects of Korea
• /
• v.46 no.1
• /
• pp.87-95
• /
• 2009

The applicability of optimization techniques for hull surface fitting has been important in the ship design process. In this research, the Genetic Algorithm has been used as a searching technique for solving surface fitting problem and minimizing errors between B-spline surface and the ship's offset data. The encoded design variables are the location of the vertex points and parametric values. The sufficient accuracy in surface fitting implies not only various techniques for computer-aided design, but also the future production design.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1999.10a
• /
• pp.113-118
• /
• 1999

The design program of fitting and thrust block including various shapes and sizes was developed for the GMR project (Great Man-made River Project in Lybia). It can automatically recalculated the minimum size of fitting required for the structural performance and can also determine the proper size of fitting for the safe and economic design by users. It was found that the results of this program was accurate and efficient when it was compared to other references. This program is also developed with GUI (Graphic User Interface) technology for the convenience of users.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.20 no.3
• /
• pp.17-29
• /
• 1995

Recently, neural network models have been employed as an alternative to regression analysis for point estimation or function fitting in various field. Thus far, however, no theoretical or empirical guides seem to exist for selecting the tool which the most suitable one for a specific function-fitting problem. In this paper, we evaluate performance of three major function-fitting techniques, regression analysis and two neural network models, back-propagation and linear-Hebbian-learning neural networks. The functions to be fitted are simple linear ones of a single independent variable. The factors considered are size of noise both in dependent and independent variables, portion of outliers, and size of the data. Based on comutational results performed in this study, some guidelines are suggested to choose the best technique that can be used for a specific problem concerned.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.42
• /
• pp.68-77
• /
• 2000

Analyses, in which load was regarded as instant load and gradual step load, respectively, were performed with data measured on a gradually loaded field, and the results were inspected to find the effect of load conditions, and the final settlements which were predicted by Hyperbolic, Tan's, Asaoka's, and Monden's methods were compared with each other. Settlement curves in which load was regarded as instant load and gradual step load being to coincide at twice the time of duration of embankment. On the ground installed vertical drain, from the results of Hyperbolic, Tan's, Asaoka's, Monden's, Curve fitting I, and Curve fitting II (simple, carrillo) methods it was concluded that Asaoka, Curve fitting I, and Curve fitting II methods are reliable for prediction final settlement with back analysis.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.17 no.2
• /
• pp.77-82
• /
• 2008

Rubbing-roller is used for manufacturing liquid crystal display, and static displacement of the rubbing-roller becomes bigger as length of the rubbing roller made of aluminum is getting longer. Therefore, material of the rubbing-roller is changed from aluminum to CFRP(Carbon Fiber Reinforced plastic). Recently thermal spraying is applied to manufacturing process of long rubbing-roller. The thermal spraying has disadvantages such as increment of manufacturing time and fraction defective caused by density of stainless steel particle. In this study, fitting process by drawing was suggested and FEM analysis with Tsai-Wu failure theory and fitting experiments are carried out to find adequate shrink allowance. The suggested shrink allowance gives proper adhesive force, and CFRP failure is not occurred. Furthermore, the fitting process is applied to long rubbing-roller and availability of the fitting process is studied by measurement of roundness, straightness and shear strength.

• IE interfaces
• /
• v.13 no.4
• /
• pp.703-708
• /
• 2000

In manufacturing processes, manufacturing features always deviate somewhat from their nominal design specifications due to several types of errors. This study suggests a fitting algorithm of the geometric profile parameters of leading and trailing edges for turbine compressor airfoils. In reality, industry personnels inspect the airfoil profile by trial-and-error method to determine the geometric feature parameters. In this study we propose an exploration approach based on factorial design with center point to minimize the effect of measurement errors caused by probe slip. By adopting the fitting method developed in this paper, one can enhance the precision and efficiency of fitting the airfoil edge profile.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.159-161
• /
• 2003

Fitting the projected wire-frame model to the detected edge pixels on images by using least-squares approach, called Least-squares Model-image Fitting (LSMIF), is the key of the Model-based Building Extraction (MBBE). It is implemented by iteratively adjusting the model parameters to minimize the squares sum of distances from the extracted edge pixels to the projected wire-frame. This paper describes a series of experiments and studies on various factors affect the fitting results, including the edge detectors, the weighting rules, the initial value of parameters, and the number of overlapped images. The experimental result is not only helpful to clarify the influences of each factor, but is also able to enhance the robustness of the LSMIF algorithm.