• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권7호
• /
• pp.949-958
• /
• 2004

본 연구는 대용량 훈련 데이타를 사용하는 얼굴 검출 분류기의 학습과정에서 새로운 데이터의 추가 학습이 가능한 새로운 방법을 제안한다. 추가되는 데이타로부터 새로운 정보를 학습하여 이미 습득된 기존의 지식을 갱신하는 것이 점진적 학습의 목표이다. 이러한 학습 기법에 기반한 분류기의 설계에서는 최종 분류기가 전체 훈련 데이타 집합의 특성을 반영하는 것이 매우 중요한 문제이다. 제안하는 알고리즘은 최적화된 최종 분류기 획득을 위하여 훈련 집합의 전역적인 특성을 대표하는 검증집합을 생성하고, 이 집단 내에서의 분류성능을 기준으로 중간단계 분류기들의 가중치를 결정한다. 각 중간단계 분류기는 개변 데이타 집합의 학습 결과로써 가중치 기반 결합 방식에 의해 최종 분류기로 구성된다. 반복적인 실험을 통해, 제안한 알고리즘을 사용하여 학습한 얼굴 검출 분류기의 성능이 AdaBoost 및 Learn＋＋기반의 분류기보다 우수한 검출 성능을 보임을 확인하였다.

• International Journal of Control, Automation, and Systems
• /
• 제1권4호
• /
• pp.453-458
• /
• 2003

The anomaly-detection algorithm based on negative selection of T cells is representative model among self-recognition methods and it has been applied to computer immune systems in recent years. In immune systems, T cells are produced through both positive and negative selection. Positive selection is the process used to determine a MHC receptor that recognizes self-molecules. Negative selection is the process used to determine an antigen receptor that recognizes antigen, or the nonself cell. In this paper, we propose a novel self-recognition algorithm based on the positive selection of T cells. We indicate the effectiveness of the proposed algorithm by change-detection simulation of some infected data obtained from cell changes and string changes in the self-file. We also compare the self-recognition algorithm based on positive selection with the anomaly-detection algorithm.

• 스마트미디어저널
• /
• 제1권2호
• /
• pp.34-39
• /
• 2012

본 연구에서는 노이즈를 제거하고 자연 영상에서 자동으로 꽃을 분할하는 후처리방법을 제시한다. 레벨 셋 알고리즘을 이용한 자연영상 꽃 분할에서는 레벨 셋이 에지 정보에만 의존하기 때문에 기대하지 않았던 분리된 노이즈들이 발생한다. 실험 결과는 제안 방법이 꽃 영역과 배경 영역의 많은 노이즈를 성공적으로 제거하였음을 보여준다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.28.5-28
• /
• 2001

This paper investigates a classification method of the electrocardiogram (ECG) into different disease categories. The features for the classification of the ECG are the coefficients of the discrete wavelet transform (DWT) of ECG signals. The coefficients are calculated with Haar wavelet, and after DWT we can get 64 coefficients. Each coefficient has morphological information and they may be good features when conventional time-domain features are not available. Since all of them are not meaningful, it is needed to reduce the size of meaningful coefficients set. The distributions of each coefficient can be the rules to classify ECG signal. The optimally reduced feature set is obtained by fuzzy c-means algorithm and rough set theory. First, the each coefficient is clustered by fuzzy c-means algorithm and the clustered ...

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권2호
• /
• pp.835-838
• /
• 2016

Background: Lung cancer remains one of the most common cancers in the world, both in terms of new cases (about 13% of total per year) and deaths (nearly one cancer death in five), because of the high case fatality. Errors in lung cancer type or malignant growth determination lead to degraded treatment efficacy, because anticancer strategy depends on tumor morphology. Materials and Methods: We have made an attempt to evaluate effectiveness of machine learning algorithms in the task of lung cancer classification based on gene expression levels. We processed four publicly available data sets. The Dana-Farber Cancer Institute data set contains 203 samples and the task was to classify four cancer types and sound tissue samples. With the University of Michigan data set of 96 samples, the task was to execute a binary classification of adenocarcinoma and non-neoplastic tissues. The University of Toronto data set contains 39 samples and the task was to detect recurrence, while with the Brigham and Women's Hospital data set of 181 samples it was to make a binary classification of malignant pleural mesothelioma and adenocarcinoma. We used the k-nearest neighbor algorithm (k=1, k=5, k=10), naive Bayes classifier with assumption of both a normal distribution of attributes and a distribution through histograms, support vector machine and C4.5 decision tree. Effectiveness of machine learning algorithms was evaluated with the Matthews correlation coefficient. Results: The support vector machine method showed best results among data sets from the Dana-Farber Cancer Institute and Brigham and Women's Hospital. All algorithms with the exception of the C4.5 decision tree showed maximum potential effectiveness in the University of Michigan data set. However, the C4.5 decision tree showed best results for the University of Toronto data set. Conclusions: Machine learning algorithms can be used for lung cancer morphology classification and similar tasks based on gene expression level evaluation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5344-5356
• /
• 2018

Nowadays with the help of Location-Based Social Networks (LBSNs), users of Point-of-Interest (POI) recommendation service in LBSNs are able to publish their geo-tagged information and physical locations in the form of sign-ups and share their experiences with friends on POI, which can help users to explore new areas and discover new points-of-interest, and promote advertisers to push mobile ads to target users. POI recommendation service in LBSNs is attracting more and more attention from all over the world. Due to the sparsity of users' activity history data set and the aggregation characteristics of sign-in area, conventional recommendation algorithms usually suffer from low accuracy. To address this problem, this paper proposes a new recommendation algorithm based on a novel Preference-Content-Region Model (PCRM). In this new algorithm, three kinds of information, that is, user's preferences, content of the Point-of-Interest and region of the user's activity are considered, helping users obtain ideal recommendation service everywhere. We demonstrate that our algorithm is more effective than existing algorithms through extensive experiments based on an open Eventbrite data set.

• International Journal of Aeronautical and Space Sciences
• /
• 제10권2호
• /
• pp.1-11
• /
• 2009

In this paper, two algorithms to solve the path planning problem with constraints from obstacles are presented. One proposed Algorithm is "Grid point-based path planning". The first step of this algorithm is to set points which can be the waypoints around the field. These points can be located inside or outside of the field or the obstacles. Therefore, we should determine whether those points are located in the field or not. Using the equations of boundary lines for a region that we are interested in is an effective approach to handle. The other algorithm is based on the boundary lines of the agricultural field, and the concept of this algorithm is well known as "boustrophedon method". These proposed algorithms are simple but powerful for complex cases since it can generate a plausible path for the complex shape which cannot be represented by using geometrical approaches efficiently and for the case that some obstacles or forbidden regions are located on the field by using a skill of discriminants about set points. As will be presented, this proposed algorithm could exhibit a reasonable accuracy to perform an agricultural mission.

• 한국도로학회논문집
• /
• 제14권2호
• /
• pp.11-17
• /
• 2012

본 연구는 실험계획법(예: 반응표면계획법) 및 하모니 검색 알고리즘을 통하여 다양한 아스팔트 콘크리트 포장 구조체에 있어 피로균열의 공용성 인자인 인장변형률을 예측하는 모델을 개발하는 방법에 대한 연구이다. 인장변형률을 산정하기 위하여 한국건설기술연구소에서 개발한 유한요소 축대칭해석 프로그램인 KICTPAVE를 이용하여 아스팔트 층과 린콘크리트 층의 접속면에서 발생되는 변형률을 구하여 데이터베이스(D/B)화 하였다. 아스팔트 포장에서 입력변수인 층별 탄성계수 및 두께를 다양한 조건에서 KICTPAVE 프로그램을 수행하여 훈련용 D/B(Training Set)인 변형률의 값들을 구축한 후 반응표면계획법에 근거하여 회귀방정식을 정의하였으며 방정식에 필요한 계수값을 결정하기 위하여 하모니 검색 알고리즘을 이용하였다. 최종적으로 결정된 회귀방정식의 계수값들의 정확성을 검증하기 위해서 훈련용 D/B가 아닌 다른 조건의 입력변수를 이용하여 검증용 D/B(Testing Set)를 구축하고 이를 이용하여 개발된 모델을 검증하였다.

• 한국측량학회지
• /
• 제31권1호
• /
• pp.1-9
• /
• 2013

본 연구는 유전 알고리즘을 이용하여 다대다 면 객체 정합을 수행하는 방법을 제안한다. 동일한 지형 지물을 표현하는 객체 집합의 형상은 서로 동일하다는 가정 하에 형상 유사도를 최적화하는 객체 집합을 두 지도 사이에서 탐색함으로써 정합을 수행한다. 이 때 어떤 객체가 객체 집합에 포함되는지의 여부를 이진 부호로 표현하고, 이진 부호들을 결합한 이진 문자열로 후보해를 표현한다. 초기 후보해들로 해집단을 생성한 뒤, 유전 알고리즘에 의하여 점진적으로 해집단의 품질을 개선함으로써 최적해를 탐색하였다. 제안된 방법을 평가하기 위하여 수원시 도심지역의 수치지형도와 지적도에서 가구계 대응 면 객체 집합을 탐색하였으며 제안된 알고리즘의 효용성을 확인할 수 있었다. 또한 수작업에 의한 탐색결과를 이용하여 평가한 결과 0.946의 정확도를 얻었다.

• Investigative Magnetic Resonance Imaging
• /
• 제24권3호
• /
• pp.123-131
• /
• 2020

Purpose: Joint label fusion (JLF) is a popular multi-atlas-based segmentation algorithm, which compensates for dependent errors that may exist between atlases. However, in order to get good segmentation results, it is very important to set the several free parameters of the algorithm to optimal values. In this study, we first investigate the feasibility of a JLF algorithm for prostate segmentation in MR images, and then suggest the optimal set of parameters for the automatic prostate segmentation by validating the results of each parameter combination. Materials and Methods: We acquired T2-weighted prostate MR images from 20 normal heathy volunteers and did a series of cross validations for every set of parameters of JLF. In each case, the atlases were rigidly registered for the target image. Then, we calculated their voting weights for label fusion from each combination of JLF's parameters (r_pxy, r_pz, r_sxy, r_sz, β). We evaluated the segmentation performances by five validation metrics of the Prostate MR Image Segmentation challenge. Results: As the number of voxels participating in the voting weight calculation and the number of referenced atlases is increased, the overall segmentation performance is gradually improved. The JLF algorithm showed the best results for dice similarity coefficient, 0.8495 ± 0.0392; relative volume difference, 15.2353 ± 17.2350; absolute relative volume difference, 18.8710 ± 13.1546; 95% Hausdorff distance, 7.2366 ± 1.8502; and average boundary distance, 2.2107 ± 0.4972; in parameters of r_pxy = 10, r_pz = 1, r_sxy = 3, r_sz = 1, and β = 3. Conclusion: The evaluated results showed the feasibility of the JLF algorithm for automatic segmentation of prostate MRI. This empirical analysis of segmentation results by label fusion allows for the appropriate setting of parameters.