• Journal of KIISE:Software and Applications
• /
• v.37 no.2
• /
• pp.110-119
• /
• 2010

In this paper, we propose an automatic segmentation of the prostate using intensity distribution and gradient information in MR images. First, active shape model using adaptive intensity profile and multi-resolution technique is used to extract the prostate surface. Second, hole elimination using geometric information is performed to prevent the hole from occurring by converging the surface shape to the local optima. Third, the surface shape with large anatomical variation is corrected by using 2D gradient information. In this case, the corrected surface shape is often represented as rugged shape which is generated by the limited number of vertices. Thus, it is reconstructed by using surface modelling and smoothing. To evaluate our method, we performed the visual inspection, accuracy measures and processing time. For accuracy evaluation, the average distance difference and the overlapping volume ratio between automatic segmentation and manual segmentation by two radiologists are calculated. Experimental results show that the average distance difference was 0.3${\pm}$0.21mm and the overlapping volume ratio was 96.31${\pm}$2.71%. The total processing time of twenty patient data was 16 seconds on average.

• Korean Management Science Review
• /
• v.27 no.3
• /
• pp.197-211
• /
• 2010

This research applies Blattberg's CE (customer equity) model to B2B business context. Through the model we estimate customer lifetime value of a Korean semiconductor manufacturer. The results show that Blattberg model has limiting applicability to B2B business context. To overcome the limits, several suggestions were made. The main suggestion was predicting maintenance costs for different customers groups and including decision variables regarding marketing costs that would help building a differentiated CE model for the B2B business context.

• Journal of Korea Multimedia Society
• /
• v.23 no.7
• /
• pp.883-890
• /
• 2020

In this paper, we propose a variational expectation-maximization algorithm that computes posterior probabilities from Latent Dirichlet Allocation (LDA) model. The algorithm approximates the intractable posterior distribution of a document term matrix generated from a corpus made up by 50 papers. It approximates the posterior by searching the local optima using lower bound of the true posterior distribution. Moreover, it maximizes the lower bound of the log-likelihood of the true posterior by minimizing the relative entropy of the prior and the posterior distribution known as KL-Divergence. The experimental results indicate that documents clustered to image classification and segmentation are correlated at 0.79 while those clustered to object detection and image segmentation are highly correlated at 0.96. The proposed variational inference algorithm performs efficiently and faster than Gibbs sampling at a computational time of 0.029s.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.5 no.1
• /
• pp.99-105
• /
• 2009

Recently the research is much interested in about the inspection system using by computer vision system. In this paper, we deal with shape inspection technique for moving to be long and narrow object on conveyor belt. first, we are acquired for moving object on conveyor belt. then the object segmentation is using by color information for background and object. the object position be calculated by horizontal and a vertical histogram. second, we are checked for two hole in front part, widths and top/bottom side information in middle part, and finally checking for two holes in rear part. The performance of our proposed model is evaluated by experiments, within error of 1㎜, and can be checking to 17 object /min.

• Proceedings of the Korea Electromagnetic Engineering Society Conference
• /
• 2002.11a
• /
• pp.324-327
• /
• 2002

For the characterization of arbitrary shaped printed circuit board, lossy transmission line grid model based on SPICE netlist and analytical plane model based on the segmentation method are proposed in this paper. Two methods are compared with an arbitrary shaped power/ground plane. Furthermore, design considerations for the complete power distribution network structure are discussed to ensure the maximum value of the PDN impedance is low enough across the desired frequency range and to guide decoupling capacitor selection.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.55 no.4
• /
• pp.167-175
• /
• 2006

This paper presents a method to segment semiconductor wafer ID on poor quality images. The method is based on multiple templates and normalized gray-level correlation (NGC) method. If the lighting condition is not so good and hence, we can not control the image quality, target image to be inspected presents poor quality ID and it is not easy to identify and then recognize the ID characters. Conventional several method to segment the interesting ID regions fails on the bad quality images. In this paper, we propose a multiple template method, which uses combinational relation of multiple templates from model templates to match several characters of the inspection images. To find out the optimal solution of multiple template model in ID regions, we introduce newly-developed snake algorithm. Experimental results using images from real FA environment are presented.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.441-443
• /
• 2018

This paper deals with the analysis of leg motion using RGB image. We used RGB image as gait analysis element by using BMC(Background Model Challenge) method and by using combining object recognition segmentation algorithm and attitude detection algorithm. It is considered that gait analysis incorporating image can be used as a parameter for classification of gait pattern recognition and abnormal gait.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.18-21
• /
• 2003

In general it has been considered to be the difficult problem that we divide continuous speech into short interval with having identical phoneme quality. In this paper we used Gaussian Mixture Model (GMM) related to probability density to divide speech into phonemes, an initial, medial, and final sound. From them we peformed continuous speech recognition. Decision boundary of phonemes is determined by algorithm with maximum frequency in a short interval. Recognition process is performed by Continuous Hidden Markov Model(CHMM), and we compared it with another phoneme divided by eye-measurement. For the experiments result we confirmed that the method we presented is relatively superior in auto-segmentation in korean speech.

• Journal of Korea Multimedia Society
• /
• v.20 no.1
• /
• pp.10-16
• /
• 2017

This paper proposes efficient schemes for image segmentation using the region-based active contour model. The developed methods can approach the boundaries of the desired objects by evolving the curves through minimization of the Mumford-Shah energy functionals, given arbitrary curves as initial conditions. Topological changes such as splitting or merging of curves should be handled for the methods to work properly without prior knowledge of the number of objects to be segmented. This paper introduces how to change topological structure of the curves and shows experimental results by applying the methods to the images.

• Journal of Multimedia Information System
• /
• v.8 no.1
• /
• pp.31-34
• /
• 2021

The classical HMM is defined by a parameter triple �� = (��, A, B), where each parameter represents a collection of probability distributions: initial state, state transition and output distributions in order. This paper proposes a new stationary parameter e = (e1, e2, …, eN) where N is the number of states and et = P(|xt = i, y) for describing how an input pattern y ends in state xt = i at time t followed by nothing. It is often said that all is well that ends well. We argue here that all should end well. The paper sets the framework for the theory and presents an efficient inference and training algorithms based on dynamic programming and expectation-maximization. The proposed model is applicable to analyzing any sequential data with two or more finite segmental patterns are concatenated, each forming a context to its neighbors. Experiments on online Hangul handwriting characters have proven the effect of the proposed augmentation in terms of highly intuitive segmentation as well as recognition performance and 13.2% error rate reduction.