• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.4
• /
• pp.1714-1729
• /
• 2018

Change detection of remote sensing images is a profound challenge in the field of remote sensing image analysis. This paper proposes a novel change detection method for bitemporal remote sensing images based on feature fusion and fuzzy c-means (FCM). Different from the state-of-the-art methods that mainly utilize a single image feature for difference image construction, the proposed method investigates the fusion of multiple image features for the task. The subsequent problem is regarded as the difference image classification problem, where a modified fuzzy c-means approach is proposed to analyze the difference image. The proposed method has been validated on real bitemporal remote sensing data sets. Experimental results confirmed the effectiveness of the proposed method.

• Journal of information and communication convergence engineering
• /
• v.20 no.3
• /
• pp.153-159
• /
• 2022

Plain radiographic analysis is the initial imaging modality for suspected small bowel obstruction. Among the many features that affect the diagnosis of small bowel obstruction (SBO), the presence of gas-filled or fluid-filled small bowel loops is the most salient feature that can be automatized by computer vision algorithms. In this study, we compare three frequently applied pixel-clustering algorithms for extracting gas-filled areas without human intervention. In a comparison involving 40 suspected SBO cases, the Possibilistic C-Means and Fuzzy C-Means algorithms exhibited initialization-sensitivity problems and difficulties coping with low intensity contrast, achieving low 72.5% and 85% success rates in extraction. The Adaptive Resonance Theory 2 algorithm is the most suitable algorithm for gas-filled region detection, achieving a 100% success rate on 40 tested images, largely owing to its dynamic control of the number of clusters.

• Journal of the Korean Statistical Society
• /
• v.31 no.4
• /
• pp.519-532
• /
• 2002

K-means clustering is a well-known partitioning method of multivariate observations. Recently, the method is implemented broadly in data mining softwares due to its computational efficiency in handling large data sets. However, it does not yield a suitable visual display of multivariate observations that is important especially in exploratory stage of data analysis. The aim of this study is to develop a K-means clustering method that enables visual display of multivariate observations in a low-dimensional space, for which the projection pursuit method is adopted. We propose a computationally inexpensive and reliable algorithm and provide two numerical examples.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2007.11a
• /
• pp.201-202
• /
• 2007

In this paper we investigated the partial discharge distribution using the K-means clustering according to the needle of tilt and void at the cross linked polyethylene(XLPE) insulators. As a result, the specimen with tilt $45^{\circ}$ has highest breakdown voltage and the specimen with air void has lower breakdown voltage than the specimen with on void. In K-menas clustering distribution of clusters concentrates at inception condition, but the distribution spreads widely at breakdown.

• Journal of Korean Academy of Nursing
• /
• v.27 no.2
• /
• pp.329-340
• /
• 1997

Resourcefulness was analyzed by Walker and Avant's method to make a theoretical framework for nursing. Also, the appropriate Korean terminology was identified.“Resourcefulness”means the ability to use internal and external resources to eliminate or to control stress. '자원동원성' is chosen to be the most appropriate term to reflect the concept of resourcefulness. Upon the concept analysis, availability, controllability, confidence, and self-instruction were identified as the defining characteristics of resourcefulness. Contrary to other work, "availability" means the ability to use the social resources as well as the internal cognitive-behavioral resources. “Controllability” means the ability to delay or control immediate gratification of one's needs or to divert one's own mood in order to solve problem. "Confidence" is the self-efficacy belief in one's control ability. "Self-instruction" is the cognitive ability to instruct oneself positively. Resourcefulness is learned by active experience, vicarious experience, and formal or informal instruction, and it is cued in the presence of a stressful situation or a need for change. As a result of the use of resourcefulness, the target behavior performed immediately and ultimately improve the quality of life or adaptation. Therefore, it is necessary to develop the instrument including the defining attributes identified in this study. Also, it is necessary to simultaneously analyze the related concepts of self-efficacy, self-control, and self-regulation for appropriate use.

• Journal of Applied Reliability
• /
• v.17 no.4
• /
• pp.355-361
• /
• 2017

Purpose: The purpose of this research is to establish the life test method for ceramic heater and identify the failure mechanisms. Methods: We do accelerated life test in the condition of thermal shock and failure analysis for failed samples. Conclusion: The main failure mechanisms of ceramic heater are identified as overstress failure mechanisms as results of failure analysis and the shape parameters of weibull distribution by accelerated life test are identified as 0.8, 1.2 and 0.4 each at $400^{\circ}C$, $600^{\circ}C$ and $900^{\circ}C$. At $900^{\circ}C$, the shape parameter 0.4 means that It is exactly initial failure caused that the stress exceeds the strength of ceramic heater highly and the shape parameters 0.8, 1.2 at $400^{\circ}C$, $600^{\circ}C$ means that the shape parameters are around 1.0 so that the main failure mechanism is overstress failure which is same result as failure analysis. It means that the appropriate life test method for ceramic heater is reliability qualification test method rather than accelerated life test.

• The Korean Journal of Community Living Science
• /
• v.18 no.4
• /
• pp.707-714
• /
• 2007

This study was conducted by surveying 300 female college students in Yanbian University and 248 surveys were considered for final analysis. Each question was rated according to 5 point scale in which 1 means 'not at all' and 5 means 'definitely'. The data of this study was statistically analyzed by SAS PC program, the factor analysis, cluster analysis, ANOVA and Duncan-test and t-test. The results are summarized as follows; The clothing attitude of the college students is classified into four factors: fashionable, brand-oriented, aesthetic and modest. The behavior pattern was categorized into four groups: conservative modest group, aesthetic modest group, fashionable aesthetic group and brand oriented group. The perception and buying desire of the fashionable aesthetic group and the brand oriented group on clothing made in Korea were higher than those of the other two groups. As the perception and buying desire on clothing made in Korea depended on the average monthly expense for clothing, the group with high expense for clothing showed higher means compared to the low expense group for clothing.

• Geotechnical Engineering
• /
• v.13 no.4
• /
• pp.13-24
• /
• 1997

A predicting method for ultimate vertical and horizontal bearing capacity by means of PAR(Pile Analysis Routines) was suggested. Based on the static pile load test data, case studies by means of PAR were performed. Ultimate pile capacity predicted by PAR was within 15% error range of that determined by stairs pile load tests. Also, the results of static pile load test, statnamic tests and PDA data performed on pipe piles were compared and, by using PAR, ultimate pile capacity was determined. Distributions of atrial pile load could be predicted and load transfer analysis could be done approximately by those distributions.

• Communications for Statistical Applications and Methods
• /
• v.12 no.2
• /
• pp.497-508
• /
• 2005

This paper aims at studying on K-means Clustering focusing on initialization which affect the clustering results in K-means cluster analysis. The four different methods(the MA method, the KA method, the Max-Min method and the Space Partition method) were compared and the clustering result shows that there were some differences among these methods, especially that the MA method sometimes leads to incorrect clustering due to the inappropriate initialization depending on the types of data and the Max-Min method is shown to be more effective than other methods especially when the data size is large.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 1998.11a
• /
• pp.295-300
• /
• 1998

Recently graphical modelling is being studied as a useful process analysis tool for exploratory causal analysis. Graphical modelling is a presentation method that uses graphs to describe statistical models of the structures of multivariate data. This paper describes an application of this graphical modeling with two cases from the automotive parts industry. One case is the unbalance problem of the pulley, an automotive generator part. There is multivariate data of the product from each of the processes which are connected in the series. By means of exploratory causal analysis between the variables using graphical modeling, the key processes which causes the variation of the final characteristics and their mechanism of the causal relationship have become clear. Another case is, also, the unbalanced problem of automotive starter parts which consists of many parts and is manufactured by complex machinery and assembling process. By means of the similar technique, the key processes are obtained easily and the results are reasonable from technical knowledge.