• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.163-168
• /
• 2009

Most image segmentation methods are to represent observed feature vectors at each pixel, which are assumed as appropriated probability models. These models can be used by statistical estimating or likelihood clustering algorithms of feature vectors. EM algorithms have some calculation problems of maximum likelihood for unknown parameters from incomplete data and maximum value in post probability distribution. First, the performance is dependent upon starting positions and likelihood functions are converged on local maximum values. To solve these problems, we mixed the Gausian function and histogram at all the level values at the image, which are proposed most suitable image segmentation methods. This proposed algoritms are confirmed to classify most edges clearly and variously, which are implemented to MFC programs.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.346-348
• /
• 2004

In this rarer, we introduce a new Fuzzy Polynomial Neural Networks (FPNNs)-like structure whose neuron is based on the Fuzzy Set-based Fuzzy Inference System (FS-FIS) and is different from that of FPNNs based on the Fuzzy relation-based Fuzzy Inference System (FR-FIS) and discuss the ability of the new FPNNs-like structurenamed Fuzzy Set-based Polynomial Neural Networks (FSPNN). The premise parts of their fuzzy rules are not identical, while the consequent parts of the both Networks (such as FPNN and FSPNN) are identical. This difference results from the angle of a viewpoint of partition of input space of system. In other word, from a point of view of FS-FIS, the input variables are mutually independent under input space of system, while from a viewpoint of FR-FIS they are related each other. In considering the structures of FPNN-like networks such as FPNN and FSPNN, they are almost similar. Therefore they have the same shortcomings as well as the same virtues on structural side. The proposed design procedure for networks' architecture involves the selection of appropriate nodes with specific local characteristics such as the number of input variables, the order of the polynomial that is constant, linear, quadratic, or modified quadratic functions being viewed as the consequent part of fuzzy rules, and a collection of the specific subset of input variables. On the parameter optimization phase, we adopt Information Granulation (IG) based on HCM clustering algorithm and a standard least square method-based learning. Through the consecutive process of such structural and parametric optimization, an optimized and flexible fuzzy neural network is generated in a dynamic fashion. To evaluate the performance of the genetically optimized FSPNN (gFSPNN), the model is experimented with using gas furnace process dataset.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.6 s.38
• /
• pp.93-102
• /
• 2005

This paper proposes an approach to verify signature using autonomous self-organized Neural Network Model , fused with fuzzy membership equation of fuzzy c-means algorithm, based on the features of the signature. To overcome limitations of the functional approach and Parametric approach among the conventional on-line signature recognition approaches, this Paper presents novel autonomous signature classification approach based on clustering features. Thirty-six globa1 features and twelve local features were defined, so that a signature verifying system with FE-SONN that learns them was implemented. It was experimented for total 713 signatures that are composed of 155 original signatures and 180 forged signatures yet 378 original signatures written by oneself. The success rate of this test is more than 97.67$\%$ But, a few forged signatures that could not be detected by human eyes could not be done by the system either.

• Journal of the Korean Society for Library and Information Science
• /
• v.40 no.2
• /
• pp.333-355
• /
• 2006

Network generation methods to visualize bibliometric data for examining the intellectual structure of knowledge domains are investigated in some detail. Among the four methods investigated in this study, pathfinder network algorithm is the most effective method in representing local details as well as global intellectual structure. The nearest neighbor graph, although never used in bibliometic analysis, also has some advantages such as its simplicity and clustering ability. The effect of input data preparation process on resulting intellectual structures are examined, and concluded that unlike MDS map with clusters, the network structure could be changed significantly by the differences in data matrix preparation process. The network generation methods investigated in this paper could be alternatives to conventional multivariate analysis methods and could facilitate our research on examining intellectual structure of knowledge domains.

• Journal of Korean Society for Atmospheric Environment
• /
• v.23 no.1
• /
• pp.14-28
• /
• 2007

This study presented the selection of ozone ($O_3$) potential factors and designed and assessed its potential prediction model using multiple-linear regression equations in Ulsan area during the springtime from April to June, $2000{\sim}2004$. $O_3$ potential factors were selected by analyzing the relationship between meterological parameters and surface $O_3$ concentrations. In addition, cluster analysis (e.g., average linkage and K-means clustering techniques) was performed to identify three major synoptic patterns (e.g., $P1{\sim}P3$) for an $O_3$ potential prediction model. P1 is characterized by a presence of a low-pressure system over northeastern Korea, the Ulsan was influenced by the northwesterly synoptic flow leading to a retarded sea breeze development. P2 is characterized by a weakening high-pressure system over Korea, and P3 is clearly associated with a migratory anticyclone. The stepwise linear regression was performed to develop models for prediction of the highest 1-h $O_3$ occurring in the Ulsan. The results of the models were rather satisfactory, and the high $O_3$ simulation accuracy for $P1{\sim}P3$ synoptic patterns was found to be 79, 85, and 95%, respectively ($2000{\sim}2004$). The $O_3$ potential prediction model for $P1{\sim}P3$ using the predicted meteorological data in 2005 showed good high $O_3$ prediction performance with 78, 75, and 70%, respectively. Therefore the regression models can be a useful tool for forecasting of local $O_3$ concentration.

• Journal of Photoscience
• /
• v.12 no.2
• /
• pp.67-73
• /
• 2005

Fluorescent probe techniques were used to evaluate the effect of dibucaine.HCl on the physical properties (transbilayer asymmetric lateral mobility, annular lipid fluidity and protein distribution) of synaptosomal plasma membrane vesicles (SPMV) isolated from bovine cerebral cortex. An experimental procedure was used based on selective quenching of 1,3-di(l-pyrenyl)propane (Py-3-Py) by trinitrophenyl groups, and radiationless energy transfer from the tryptophans of membrane proteins to Py-3-Py. Dibucaine.HCl increased the bulk lateral mobility, and annular lipid fluidity in SPMV lipid bilayers, and had a greater fluidizing effect on the inner monolayer than the outer monolayer. The magnitude of increasing effect on annular lipid fluidity in SPMV lipid bilayer induced by dibucaine.HCl was significantly far greater than magnitude of increasing effect of the drug on the lateral mobility of bulk SPMV lipid bilayer. It also caused membrane proteins to cluster. These effects of dibucaine.HCl on neuronal membranes may be responsible for some, though not all, of the local anesthetic actions of dibucaine.HCl.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2003.11a
• /
• pp.135-135
• /
• 2003

The fast evolution in the fold of optical communication systems demands powerful optical information treatment. These functions can be performed by integrated optical systems. A key component of such systems is erbium doped waveguide amplifier(EDWA). The intra 4f radiative transition of Er at 1.5 $\mu\textrm{m}$ is particularly interesting because this wavelength is standard in optical telecommunications. The fabrication of waveguide amplifier for integrated optics using sol-gel process has received an increasing attention. Potential advantage of lower cost by less capital equipment and easy processing makes this process an attractive alternatives to conventional technologies like flame hydrolysis deposition, ion exchange and chemical vapor deposition, etc. In addition, sol-gel process has been found to be extremely suitable for the control of composition and refractive index related directly with optical properties. The main drawback of such an amplifier with respect to the EDWA is the need for a much higher Er3+ concentration to compensate for the smaller interaction length. However, the high doping of Er might be resulted in the non-radiative relaxation by clustering of Er ions End co-operative upconversion. In order to solve this problem, we investigate the possibility of avoiding short Er-Er distances by encapsulation of Er3+ ions in hosts such as organic-inorganic hybrid materials. For inorganic-organic hybrid sols, methacryloxypropyltrimethoxysilane (MPTS), zirconyl chloride octahydrate and erbium(III) chloride hexahydrate were used as starting materials, followed by conventional sol-gel process. It was observed by TEM that nano sols having core/shell toplology were formed, depending on the mole ratio of Zr/Er. The surface roughness for the coatings on Si substrate was investigated by AFM as a function of Zr/Er ratio. The local environment and vibrational Properties of Er3+ ions were studied using Near-IR, FT-IR, and UV/Vis spectroscopy. Nano hybrid coatings derived from polymer and Er doped encapsulation Eave the good luminescence at 1.55$\mu\textrm{m}$.

• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.401-408
• /
• 2006

In this paper, we propose a new class of segmentation technique for feature extraction based on the statistical and regional classification at each vertical or horizontal line of digital image data. Data is processed and clustered at each line, different from the point or space process. They are designed to segment gray-scale sectional images using a horizontal and vertical line process due to their statistical and property differences, and to extract the feature. The techniques presented here show efficient results in case of the gray level overlap and not having threshold image. Such images are also not easy to be segmented by the global or local threshold methods. Line pixels inform us the sectionable data, and can be set according to cluster quality due to the differences of histogram and statistical data. The total segmentation on line clusters can be obtained by adaptive extension onto the horizontal axis. Each processed region has its own pixel value, resulting in feature extraction. The advantage and effectiveness of the line-cluster approach are both shown theoretically and demonstrated through the region-segmental carotid artery medical image processing.

• Journal of KIISE
• /
• v.42 no.6
• /
• pp.699-706
• /
• 2015

In recent years, as near-duplicate image has been increasing explosively by the spread of Internet and image-editing technology that allows easy access to image contents, related research has been done briskly. However, BoF (Bag-of-Feature), the most frequently used method for near-duplicate image detection, can cause problems that distinguish the same features from different features or the different features from same features in the quantization process of approximating a high-level local features to low-level. Therefore, a post-verification method for BoF is required to overcome the limitation of vector quantization. In this paper, we proposed and analyzed the performance of a post-verification method for BoF, which converts SIFT (Scale Invariant Feature Transform) descriptors into 128 bits binary codes and compares binary distance regarding of a short ranked list by BoF using the codes. Through an experiment using 1500 original images, it was shown that the near-duplicate detection accuracy was improved by approximately 4% over the previous BoF method.

• Annual Conference of KIPS
• /
• 2003.11a
• /
• pp.201-204
• /
• 2003

At a consequence of WWW large popularity, the internet has suffered from various performance problems, such as network congestion and overloaded servers. These days, it is not uncommon to find servers refusing connections because they are overloaded. Web server performance has always been a key issue in the design and operation of on-line systems. With regard to Internet, performance is also critical, because users want fast and easy access to all objects (e.g., documents, graphics, audio, and video) available on the net. To solve this problem, a number of companies are exploring the benefits of having multiple geographically or locally distributed Internet sites. This requires a comprehensive scheme for traffic management, which includes the principle of an optimal load balancing of client requests across multiple clusters of real servers. This paper focuses on the performance analysis of Web server and we apply these results to load balancing in clustering web server. It also discusses the mam steps needed to carry out a WWW performance analysis effort and shows relations between the workload characteristics and system resource usage. Also, we will introduce an optimal load balancing algorithm base on the RTSM (Real-Time Server Monitor) and Fuzzy Inference Engine for the local status of a real server, and the benefits is provided with of the suggested method.