• The Korean Journal of Applied Statistics
• /
• v.24 no.2
• /
• pp.335-345
• /
• 2011

In this paper, we propose a new hybrid speed function for image segmentation using level set. A new proposed speed function uses the region and boundary information of image object for the exact result of segmentation. The region information is defined by the probability information of pixel intensity in a ROI(region-of-interest), and the boundary information is defined by the gradient vector flow obtained from the gradient of image. We show the results of experiment for an various artificial image and real medical image to verify the accuracy of segmentation using proposed method.

• Nuclear Engineering and Technology
• /
• v.52 no.7
• /
• pp.1436-1442
• /
• 2020

Nuclear infrastructure systems play an important role in national security. The functions and missions of nuclear infrastructure systems are vital to government, businesses, society and citizen's lives. It is crucial to design nuclear infrastructure for scalability, reliability and robustness. To do this, we can use machine learning, which is a state of the art technology used in various fields ranging from voice recognition, Internet of Things (IoT) device management and autonomous vehicles. In this paper, we propose to design and develop a machine learning algorithm to perform predictive maintenance of nuclear infrastructure. Support vector machine and logistic regression algorithms will be used to perform the prediction. These machine learning techniques have been used to explore and compare rare events that could occur in nuclear infrastructure. As per our literature review, support vector machines provide better performance metrics. In this paper, we have performed parameter optimization for both algorithms mentioned. Existing research has been done in conditions with a great volume of data, but this paper presents a novel approach to correlate nuclear infrastructure data samples where the density of probability is very low. This paper also identifies the respective motivations and distinguishes between benefits and drawbacks of the selected machine learning algorithms.

• Journal of Mechanical Science and Technology
• /
• v.16 no.3
• /
• pp.292-301
• /
• 2002

Ultrasound waves interact strongly with the orientation and sequence of the plies in a layup when propagating in the thickness direction of composite laminates. Also the layup orientation greatly influences its properties in a composite laminate. If the layup orientation of a ply is misaligned, it could result in the part being rejected and discarded. Now, most researchers cut a small coupon from the waste edge and use a microscope to optically verify the ply sequences on important parts. This may add a substantial cost to the production since the test is both labor intensive and performed after the part is cured. A nondestructive technique would be very beneficial, which could be used to test the part after curing and requires less time than the optical test. Therefore we have developed, reduced, and implemented a novel ply-by-ply vector decomposition model for composite laminates fabricated from unidirectional plies. This model decomposes the transmission of a linearly polarized ultrasound wave into orthogonal components through each ply of a laminate. High probability is found, by comparisons between the model and tests, in characterizing cured layups of the laminates by using the proposed method.

• Annual Conference on Human and Language Technology
• /
• 2014.10a
• /
• pp.95-99
• /
• 2014

본 논문에서는 최근 급속히 증가하여 사회적 이슈가 되고 있는 SMS 스팸 필터링을 위한 듀얼 SMS 스팸필터링 기법을 제안한다. 지속적으로 증가하고 새롭게 변형되는 SMS 문자 필터링을 위해서는 패턴 및 스팸 단어 사전을 통한 필터링은 많은 수작업을 요구하여 부적합하다. 그리하여 기계 학습을 이용한 자동화 시스템 구축이 요구되고 있으며, 효과적인 기계 학습을 위해서는 자질 선택과 자질의 가중치 책정 방법이 중요하다. 하지만 SMS 문자 특성상 문장들이 짧기 때문에 출현하는 자질의 수가 적어 분류의 어려움을 겪게 된다. 이 같은 문제를 개선하기 위하여 본 논문에서는 슬라이딩 윈도우 기반 N-gram 확장을 통해 자질을 확장하고, 확장된 자질로 그래프를 구축하여 얕은 구조적 특징을 표현한다. 학습 데이터에 출현한 N-gram 자질을 정점(Vertex)으로, 자질의 출현 빈도를 그래프의 간선(Edge)의 가중치로 설정하여 햄(HAM)과 스팸(SPAM) 그래프를 각각 구성한다. 이렇게 구성된 그래프를 바탕으로 노드의 중요도와 간선의 가중치를 활용하여 최종적인 자질의 가중치를 결정한다. 입력 문자가 도착하면 스팸과 햄의 그래프를 각각 이용하여 입력 문자의 2개의 자질 벡터(Vector)를 생성한다. 생성된 자질 벡터를 지지 벡터 기계(Support Vector Machine)를 이용하여 각 SVM 확률 값(Probability Score)을 얻어 스팸 여부를 결정한다. 3가지의 실험환경에서 바이그램 자질과 이진 가중치를 사용한 기본 시스템보다 F1-Score의 약 최대 2.7%, 최소 0.5%까지 향상되었으며, 결과적으로 평균 약 1.35%의 성능 향상을 얻을 수 있었다.

• Proceedings of the KSME Conference
• /
• 2001.06a
• /
• pp.401-406
• /
• 2001

When propagating the thickness direction of composite laminates ultrasound waves interacts strongly with the orientation and sequence of the plies in a layup. Also the layup orientation greatly influences its properties in a composite laminate. If one ply of the layup orientation is misaligned, it could result in the part being rejected and discarded. Now, most researchers cut a small coupon from the waste edge and use a microscope to optically verify the ply sequences on important parts. Those may add a substantial cost to the product since the test is both labor hard and performed after the part is cured. A nondestructive technique would be very beneficial, which could be used to test the part after curing and require less time than the optical test. Therefore we have developed, reduced, and implemented a novel ply-by-ply vector decomposition model for composite lam mates fabricated from unidirectional plies. This model decomposes the transmission of a linearly polarized ultrasound wave into orthogonal components through each ply of a laminate. It is found that a high probability shows between the model and tests developed in characterizing cured layups of the laminates.

• Journal of Satellite, Information and Communications
• /
• v.8 no.1
• /
• pp.77-82
• /
• 2013

The object tracking method using particle filtering has been proved successful since it is based on the Monte Carlo simulation to estimate the posterior distribution of the state vector that is nonlinear and non-Gaussian in the real-world situation. In this paper, we present two nobel methods that can improve the performance of the object tracking algorithm based on the particle filtering. First one is the feedback method that replace the low-weighted tracking sample by the estimated state vector in the previous frame. The second one is an tracking box correction method to find an confidence interval of back projection probability on the estimated candidate object area. An sample propagation equation is also presented, which is obtained by experiments. We designed well-organized test data set which reflects various challenging circumstances, and, by using it, experimental results proved that the proposed methods improves the traditional particle filter based object tracking method.

• Korean Journal of Acupuncture
• /
• v.33 no.1
• /
• pp.18-32
• /
• 2016

Objectives : This study aims to analyze the co-occurrence of pathological symptoms and corresponding acupoints as documented by the comprehensive acupuncture and moxibustion records in the classical texts of Far East traditional medicine as an aid to a more efficient understanding of the tacit treatment principles of ancient physicians. Methods : The Classic of Nourishing Life with Acupuncture and Moxibustion(Zhenjiu Zisheng Jing; hereinafter ZZJ) was selected as the primary reference book for the analysis. The pathology-acupoint co-occurrence analysis was performed by applying 4 values of vector space measures(weighted Euclidean distance, Euclidean distance, $Cram\acute{e}r^{\prime}s$ V and Canberra distance), which measure the distance between the observed and expected co-occurrence counts, and 3 values of probabilistic measures(association strength, Fisher's exact test and Jaccard similarity), which measure the probability of observed co-occurrences. Results : The treatment records contained in ZZJ were preprocessed, which yielded 4162 pathology-acupoint sets. Co-occurrence was performed applying 7 different analysis variables, followed by a prediction simulation. The prediction simulation results revealed the Weighted Euclidean distance had the highest prediction rate with 24.32%, followed by Canberra distance(23.14%) and association strength(21.29%). Conclusions : The weighted Euclidean distance among the vector space measures and the association strength among the probabilistic measures were verified to be the most efficient analysis methods in analyzing the correlation between acupoints and pathologies found in the classical medical texts.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.6_2
• /
• pp.623-629
• /
• 2012

During the construction of crossing engineering one of the important measures to ensure the safety of subway operation is the implementation of deformation surveying to the existing subway tunnel. Guangzhou new subway line 2 engineering which crosses the existing tunnel is taken as the background. How to achieve intelligent and automatic deformation surveying forecast during the subway tunnel construction process is studied. Because large amount of surveying data exists in the subway construction, deformation analysis is difficult and prediction has low accuracy, a subway intelligent deformation prediction model based on the PBIL and support vector machine is proposed. The PBIL algorithm is used to optimize the exact key parameters combination of support vector machine though probability analysis and thereby the predictive ability of the model deformation is greatly improved. Through applications on the Guangzhou subway across deformation surveying deformation engineering the prediction method's predictive ability has high accuracy and the method has high practicality. It can support effective solution to the implementation of the comprehensive and accurate surveying and early warning under subway operation conditions with the environmental interference and complex deformation.

• The Journal of the Korea Contents Association
• /
• v.11 no.5
• /
• pp.9-14
• /
• 2011

This paper presents a new fast motion estimation method where search ranges are determined by the probabilities of motion vector differences (MVDs), which is an adaptive/dynamic search range (ASR) method. The MVDs' distribution is investigated and its parameter is estimated by the maximum likelihood estimator. With the estimated distribution, we show that the search ranges can be efficiently restricted by a prefixed probability for MVDs. Experimental results showed that the performance of the proposed method is very similar to that of the full search algorithm in PSNR but it enables significant reduction in the computational complexity. In addition, they revealed that the proposed method determine the search ranges much more efficiently than the conventional ASR methods.

• Structural Engineering and Mechanics
• /
• v.66 no.3
• /
• pp.399-409
• /
• 2018

This paper demonstrates the potential application of machine learning algorithms for approximate prediction of the load and deflection capacities of the novel type of Laced Steel Concrete-Composite (LSCC) beams proposed by Anandavalli et al. (Engineering Structures 2012). Initially, global and local responses measured on LSCC beam specimen in an experiment are used to validate nonlinear FE model of the LSCC beams. The data for the machine learning algorithms is then generated using validated FE model for a range of values of the identified sensitive parameters. The performance of four well-known machine learning algorithms, viz., Support Vector Regression (SVR), Minimax Probability Machine Regression (MPMR), Relevance Vector Machine (RVM) and Multigene Genetic Programing (MGGP) for the approximate estimation of the load and deflection capacities are compared in terms of well-defined error indices. Through relative comparison of the estimated values, it is demonstrated that the algorithms explored in the present study provide a good alternative to expensive experimental testing and sophisticated numerical simulation of the response of LSCC beams. The load carrying and displacement capacity of the LSCC was predicted well by MGGP and MPMR, respectively.