• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.787-790
• /
• 1998

In this paper, we presents a method for multi-level logic mainmization which is suitable for the minimization of look-up table type FPGAs. A pattern extraction algorithm is minimized AND/XOR multi-level circuits. The circuits apply to Roth-Karp decomposition which is most commonly used technique in the FPGA technology mapping. We tested the FPGA synthesis method using pattern extraction on a set of benchmark. The proposed method achieved reductions on the number of LUTs in mapping soultion as compared with MISII(or SIS) or previous results〔5〕

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.18 no.3
• /
• pp.225-244
• /
• 2014

In this work, we discuss segmentation algorithms based on the level set method that incorporates shape prior knowledge. Fundamental segmentation models fail to segment desirable objects from a background when the objects are occluded by others or missing parts of their whole. To overcome these difficulties, we incorporate shape prior knowledge into a new segmentation energy that, uses global and local image information to construct the energy functional. This method improves upon other methods found in the literature and segments images with intensity inhomogeneity, even when images have missing or misleading information due to occlusions, noise, or low-contrast. We consider the case when the shape prior is placed exactly at the locations of the desired objects and the case when the shape prior is placed at arbitrary locations. We test our methods on various images and compare them to other existing methods. Experimental results show that our methods are not only accurate and computationally efficient, but faster than existing methods as well.

• Journal of Advanced Marine Engineering and Technology
• /
• v.30 no.2
• /
• pp.247-252
• /
• 2006

Almost ship monitoring systems are event driven alarm system which warn only when the measurement value is over or under set point. These kinds of system cannot warn until signal is growing to abnormal state that the signal is over or under the set point. therefore cannot play a role for preventive maintenance system. This paper proposes fault diagnosis method which is able to diagnose and forecast the fault from present operating condition by analyzing monitored signals with present ship monitoring system without any additional sensors. By analyzing the data with high correlation coefficient(CC), correlation level of interactive data can be defined. Knowledge base of abnormal detection can be built by referring level of CC(Fault Detection CC. FDCC) to detect abnormal data among monitored data from monitoring system and knowledge base of diagnosis built by referring CC among interactive data for related machine each other to diagnose fault part.

• Proceedings of the Korean Society of Marine Engineers Conference
• /
• 2005.06a
• /
• pp.281-286
• /
• 2005

Almost ship monitoring systems are event driven alarm system which warn only when the measurement value is over or under set point. These kinds of system cannot warn while signal is growing to abnormal state until the signal is over or under the set point and cannot play a role for preventive maintenance system. This paper proposes fault diagnosis method which is able to diagnose and forecast the fault from present operating condition by analyzing monitored signals with present ship monitoring system without additional sensors. By analyzing this data having high correlation coefficient(CC), correlation level of interactive data can be understood. Knowledge base of abnormal detection can be built by referring level of CC(Fault Detection CC, FDCC) to detect abnormal data among monitored data from monitoring system and knowledge base of diagnosis built by referring CC among interactive data for related machine each other to diagnose fault part.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.408-410
• /
• 2021

In the NLP field, the pre-training model BERT launched by the Google team in 2018 has shown amazing results in various tasks in the NLP field. Subsequently, many variant models have been derived based on the original BERT, such as RoBERTa, ERNIEBERT and so on. In this paper, the WWMBERT (Whole Word Masking BERT) model suitable for Chinese text tasks was used as the baseline model of our experiment. The experiment is mainly for "Text-level Chinese text classification tasks" are improved, which mainly combines Tapt (Task-Adaptive Pretraining) and "Multi-Sample Dropout method" to improve the model, and compare the experimental results, experimental data sets and model scoring standards Both are consistent with the official WWMBERT model using Accuracy as the scoring standard. The official WWMBERT model uses the maximum and average values of multiple experimental results as the experimental scores. The development set was 97.70% (97.50%) on the "text-level Chinese text classification task". and 97.70% (97.50%) of the test set. After comparing the results of the experiments in this paper, the development set increased by 0.35% (0.5%) and the test set increased by 0.31% (0.48%). The original baseline model has been significantly improved.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.5
• /
• pp.447-455
• /
• 2014

In this paper, we consider a representative integral imaging method in glasses-free 3D image processing and propose a prioritized transmission scheme for guaranteeing a received video quality in error-prone environments. According to the correlation of pixels consisting of each voxel of integral image, we set the priority differently and apply the modulation level according to this priority value. That is to say, the corresponding pixels with small variance are set to a high priority and transmitted by using a low level modulation that is robust under transmission errors, but the corresponding pixels with greater variance are set to a lower priority and transmitted by using a high level modulation that has a high bit error rate but fast transmission rate. Result shows that the proposed scheme that applies the error-robust modulation level to the important image bit stream with the high priority improves the peak to sidelobe ratio (PSR) of the received 3D image, compared with a typical method that use the same modulation level without distinction of priorities.

• Proceedings of the KSRS Conference
• /
• v.1
• /
• pp.126-129
• /
• 2006

The marine environment is under considerable threat from intentional or accidental oil spills, ballast water discharged, dredging and infilling for coastal development, and uncontrolled sewage and industrial wastewater discharges. Monitoring spills and illegal oil discharges is an important component in ensuring compliance with marine protection legislation and general protection of the coastal environments. For the monitoring task an image processing system is needed that can efficiently perform the detection and the tracking of oil spills and in this direction a significant amount of research work has taken place mainly with the use of radar (SAR) remote sensing data. In this paper the level set image segmentation technique was tested for the detection of oil spills. Level set allow the evolving curve to change topology (break and merge) and therefore boundaries of particularly intricate shapes can be extracted. Experimental results demonstrated that the level set segmentation can be used for the efficient detection and monitoring of oil spills, since the method coped with abrupt shape’s deformations and splits.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2005.05a
• /
• pp.43-46
• /
• 2005

Micro molding technology is a promising mass production technology for polymer based microstructures. Mass production technologies such as the micro injection/compression molding, hot embossing, and micro reaction molding are already in use. In the present study, we have developed a numerical analysis system to simulate three-dimensional non-isothermal cavity filling for hot embossing, with a special emphasis on the free surface capturing. Precise free surface capturing has been successfully accomplished with the level set method, which is solved by means of the Runge-Kutta discontinuous Galerkin (RKDG) method. The RKDG method turns out to be excellent from the viewpoint of both numerical stability and accuracy of volume conservation. The Stokes equations are solved by the stabilized finite element method using the equal order tri-linear interpolation function. To prevent possible numerical oscillation in temperature Held we employ the streamline upwind Petrov-Galerkin (SUPG) method. With the developed code we investigated the detailed change of free surface shape in time during the mold filling. In the filling simulation of a simple rectangular cavity with repeating protruded parts, we find out that filling patterns are significantly influenced by the geometric characteristics such as the thickness of base plate and the aspect ratio and pitch of repeating microstructures. The numerical analysis system enables us to understand the basic flow and material deformation taking place during the cavity filling stage in microstructure fabrications.

• Proceedings of the KIEE Conference
• /
• 2008.11a
• /
• pp.50-52
• /
• 2008

We have limited the transfer power to capital area below a certain level which is called "The Capital Area Transfer Power Limit", and calculated on every Thursday for the application next week. This level is very important in our network operation, because if this level is not set properly, our power network can be fallen under great danger in case of a fault among the transfer power line. But the calculation procedure for the limit level is so complicated and iterative that it mace us spend much time and do much work. So, when a sudden trip of the related facility to the limit level we can't recalculate the limit level fast enough. And this can drop our network reliability below our standards, therefore our network can be dangerous. To avoid this kind of problems, we have figured out a method to calculate simply the limit level. That method uses the index related to the level. We think this method can make short of the calculation procedures for the level. This paper deals with the simplified method for the calculation of the level limit.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.21 no.3
• /
• pp.239-245
• /
• 2008

A level set based topological shape optimization using extended B-spline basis functions is developed for steady-state heat conduction problems. The only inside of complicated domain identified by the level set functions is taken into account in computation, so we can remove the effects of domain outside parts in heat conduction problem. The solution of Hamilton-Jacobi equation leads to an optimal shape according to the normal velocity field determined from the sensitivity analysis, minimizing a thermal compliance while satisfying a volume constraint. To obtain exact shape sensitivity, the precise normal and curvature of geometry need to be determined using the level set and B-spline basis functions. Using topological derivative concept, the nucleation of holes for topological changes can be made whenever and wherever necessary during the optimization.