• Journal of the Korea Society of Computer and Information
• /
• v.7 no.4
• /
• pp.75-82
• /
• 2002

Video viewers can not understand enough entire video contents because most video is long length data of large capacity. This paper Propose efficient scene change detection and video abstracting using new shot clustering to solve this problem. Scene change detection is extracted by method that was merged color histogram with χ2 histogram. Clustering is performed by similarity measure using difference of local histogram and new shot merge algorithm. Furthermore, experimental result is represented by using Real TV broadcast program.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.717-718
• /
• 2008

Fisher Linear Discriminant(FLD) is a sample and intuitive linear feature extraction method in pattern recognition. But in some special cases, such as un-separable case, one class data dispersed into several clustering case, FLD doesn't work well. In this paper, a new discriminant named K-means Fisher Linear Discriminant, which combines FLD with K-means clustering is proposed. It could deal with this case efficiently, not only possess FLD's global-view merit, but also K-means' local-view property. Finally, the simulation results also demonstrate its advantage against K-means and FLD individually.

• Journal of Korea Multimedia Society
• /
• v.20 no.2
• /
• pp.344-351
• /
• 2017

In this paper, we propose the obstacle database management module for obstacle estimation and clustering. The proposed G-eye manager system can create customized walking route for blind people using the UI manager and verify the coordinates of the path. Especially, G-eye management system designed a regional information module. The regional information module can improve the loading speed of the obstacle data by classifying the local information by clustering the coordinates of the obstacle. In this paper, we evaluate the reliability of the walking route generated from the obstacle map. We obtain the coordinate value of the path avoiding the virtual obstacle from the proposed system and analyze the error rate of the path avoiding the obstacle according to the size of the obstacle. And we analyze the correlation between obstacle size and route by classifying virtual obstacles into sizes.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2002.05a
• /
• pp.949-952
• /
• 2002

A regression model is used in predicting the response variable given predictor variables However, in case of large number of predictor variables, a regression model has some problems such as multicollinearity, interpretation of the functional relationship between the response and predictors and prediction accuracy. A clustering method and stepwise regression could be used to reduce the amount of data by grouping predictors having similar properties and by selecting the subset of predictors. respectively. This paper proposes a prediction method combining clustering method and stepwise regression. The proposed method fits a global model and local models and predicts responses given new observations by using both models. The paper also compares the performance of proposed method with stepwise regression via a real data of ample obtained in a steel process.

• Journal of KIISE:Software and Applications
• /
• v.26 no.1
• /
• pp.130-130
• /
• 1999

In this paper, we propose a novel design method for improving performance of existing FCM-type clustering algorithms. First, we define the performance measure which focuses on bothcompactness and separation of clusters. Next, we optimize this measure using evolution program.Especially the proposed method has following merits: ① using evolution program, it solves suchproblems as initialization, number of clusters, and convergence to local optimum ② it reduces searchspace and improves convergence speed of algorithm since it represents chromosome with possiblepotential centers which are selected possible candidates of centers by density measure ③ it improvesperformance of clustering algorithm with the performance index which embedded both compactnessand separation Properties ④ it is robust to noise data since it minimizes its effect on center search.

• Korean Journal of Remote Sensing
• /
• v.20 no.3
• /
• pp.215-225
• /
• 2004

A multistage hierarchical clustering technique, which is an unsupervised technique, was suggested in this paper for classifying large remotely-sensed imagery. The multistage algorithm consists of two stages. The 'local' segmentor of the first stage performs region-growing segmentation by employing the hierarchical clustering procedure of CN-chain with the restriction that pixels in a cluster must be spatially contiguous. The 'global' segmentor of the second stage, which has not spatial constraints for merging, clusters the segments resulting from the previous stage, using the conventional agglomerative approach. Using simulation data, the proposed method was compared with another hierarchical clustering technique based on 'mutual closest neighbor.' The experimental results show that the new approach proposed in this study considerably increases in computational efficiency for larger images with a low number of bands. The technique was then applied to classify the land-cover types using the remotely-sensed data acquired from the Korean peninsula.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.40 no.4
• /
• pp.203-210
• /
• 2017

Data clustering is one of the most difficult and challenging problems and can be formally considered as a particular kind of NP-hard grouping problems. The K-means algorithm is one of the most popular and widely used clustering method because it is easy to implement and very efficient. However, it has high possibility to trap in local optimum and high variation of solutions with different initials for the large data set. Therefore, we need study efficient computational intelligence method to find the global optimal solution in data clustering problem within limited computational time. The objective of this paper is to propose a combined artificial bee colony (CABC) with K-means for initialization and finalization to find optimal solution that is effective on data clustering optimization problem. The artificial bee colony (ABC) is an algorithm motivated by the intelligent behavior exhibited by honeybees when searching for food. The performance of ABC is better than or similar to other population-based algorithms with the added advantage of employing fewer control parameters. Our proposed CABC method is able to provide near optimal solution within reasonable time to balance the converged and diversified searches. In this paper, the experiment and analysis of clustering problems demonstrate that CABC is a competitive approach comparing to previous partitioning approaches in satisfactory results with respect to solution quality. We validate the performance of CABC using Iris, Wine, Glass, Vowel, and Cloud UCI machine learning repository datasets comparing to previous studies by experiment and analysis. Our proposed KABCK (K-means+ABC+K-means) is better than ABCK (ABC+K-means), KABC (K-means+ABC), ABC, and K-means in our simulations.

• Journal of Information Processing Systems
• /
• v.16 no.3
• /
• pp.612-628
• /
• 2020

This paper presents a new approach for detecting abnormal behaviors in complex surveillance scenes where anomalies are subtle and difficult to distinguish due to the intricate correlations among multiple objects' behaviors. Specifically, a cascaded probabilistic topic model was put forward for learning the spatial context of local behavior and the temporal context of global behavior in two different stages. In the first stage of topic modeling, unlike the existing approaches using either optical flows or complete trajectories, spatio-temporal correlations between the trajectory fragments in video clips were modeled by the latent Dirichlet allocation (LDA) topic model based on Markov random fields to obtain the spatial context of local behavior in each video clip. The local behavior topic categories were then obtained by exploiting the spectral clustering algorithm. Based on the construction of a dictionary through the process of local behavior topic clustering, the second phase of the LDA topic model learns the correlations of global behaviors and temporal context. In particular, an abnormal behavior recognition method was developed based on the learned spatio-temporal context of behaviors. The specific identification method adopts a top-down strategy and consists of two stages: anomaly recognition of video clip and anomalous behavior recognition within each video clip. Evaluation was performed using the validity of spatio-temporal context learning for local behavior topics and abnormal behavior recognition. Furthermore, the performance of the proposed approach in abnormal behavior recognition improved effectively and significantly in complex surveillance scenes.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.58-66
• /
• 2007

This paper introduces the new generic dynamic neuro-fuzzy local modeling system (DNFLMS) that is based on a dynamic Takagi-Sugeno (TS) type fuzzy inference system for complex dynamic hydrological modeling tasks. The proposed DNFLMS applies a local generalization principle and an one-pass training procedure by using the evolving clustering method to create and update fuzzy local models dynamically and the extended Kalman filtering learning algorithm to optimize the parameters of the consequence part of fuzzy local models. The proposed DNFLMS is applied to develop the inference model to forecast the flow of Waikoropupu Springs, located in the Takaka Valley, South Island, New Zealand, and the influence of the operation of the 32 Megawatts Cobb hydropower station on springs flow. It is demonstrated that the proposed DNFLMS is superior in terms of model accuracy, model complexity, and computational efficiency when compared with a multi-layer perceptron trained with the back propagation learning algorithm and well-known adaptive neural-fuzzy inference system, both of which adopt global generalization.

• World Technopolis Review
• /
• v.2 no.2
• /
• pp.81-95
• /
• 2013

This paper seeks to build on previous work conducted by Porter, Devol, Florida, Bahrami and Evans, Wennberg and Lindqvist, and others contained in the literature, to construct a new way of looking at innovation cluster development. It seeks to describe the key elements contained in the research that serve as building blocks for innovation clustering, adding analysis dimensions that aim to further illuminate understanding of this process. It compares those building block characteristics to the innovation topography of U.S. urban centers, to shed light on a new framework through which urban innovation cluster formation can be considered. It identifies three building block analysis categories: 1) Technological Capability and Capacity (TCC); 2) Intellectual Propulsion Capacity (IPC); and 3) Structural Creative Inspiration (SCI). These three pillars form the architecture for creation of a Strategic Innovation Network (SIN), upon which clustering can be systematically analysed and built. The purpose of the SIN is to optimally organize and connect all available resources that include physical, financial, and human, such that innovation clustering is inspired, encouraged, nurtured, and ultimately constructed as fully functioning socio-economic organisms that provide both local and regional benefits. It is designed to aid both private enterprise and public policy leaders in their strategic planning considerations, and to enhance urban economic development opportunities.