• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 춘계종합학술대회 A
• /
• pp.877-881
• /
• 2008

In this paper, we propose a real-time people tracking system with multiple CCD cameras for security inside the building. The camera is mounted from the ceiling of the laboratory so that the image data of the passing people are fully overlapped. The implemented system recognizes people movement along various directions. To track people even when their images are partially overlapped, the proposed system estimates and tracks a bounding box enclosing each person in the tracking region. The approximated convex hull of each individual in the tracking area is obtained to provide more accurate tracking information. To achieve this goal, we propose a method for 3D walking human tracking based on the IMPRESARIO framework incorporating cascaded classifiers into hypothesis evaluation. The efficiency of adaptive selection of cascaded classifiers have been also presented. We have shown the improvement of reliability for likelihood calculation by using cascaded classifiers. Experimental results show that the proposed method can smoothly and effectively detect and track walking humans through environments such as dense forests.

• The Journal of the Acoustical Society of Korea
• /
• 제20권2E호
• /
• pp.17-24
• /
• 2001

In this paper, we propose the bootstrap and aggregating (bagging) vector quantization (VQ) classifier to improve the performance of the text-independent speaker recognition system. This method generates multiple training data sets by resampling the original training data set, constructs the corresponding VQ classifiers, and then integrates the multiple VQ classifiers into a single classifier by voting. The bagging method has been proven to greatly improve the performance of unstable classifiers. Through two different experiments, this paper shows that the VQ classifier is unstable. In one of these experiments, the bias and variance of a VQ classifier are computed with a waveform database. The variance of the VQ classifier is compared with that of the classification and regression tree (CART) classifier[1]. The variance of the VQ classifier is shown to be as large as that of the CART classifier. The other experiment involves speaker recognition. The speaker recognition rates vary significantly by the minor changes in the training data set. The speaker recognition experiments involving a closed set, text-independent and speaker identification are performed with the TIMIT database to compare the performance of the bagging VQ classifier with that of the conventional VQ classifier. The bagging VQ classifier yields improved performance over the conventional VQ classifier. It also outperforms the conventional VQ classifier in small training data set problems.

• 전자공학회논문지C
• /
• 제35C권6호
• /
• pp.88-98
• /
• 1998

부분공간 인식기는 Karhunen-Loeve (KL) 변환을 기반으로 하는 대표적인 패턴인식 방법이다. 이 부분 공간 인식기는 고차원의 패턴을 저차원의 부분공간에 나타내어 인식을 한다. 그러나 차원 감축으로 인한 정보의 손실로 principal components가 유사하게 나타나는 패턴간에는 분별이 어려워지는 단점이 있다. 본 논문에서는 이러한 부분공간 인식기의 단점을 해결하기 위해 일반적으로 무시되는 minor components로 표현되는 패턴의 노벨티 성분을 이용하는 결합된 다중 노벨티 신경망 인식기를 제안하고 부분공간 인식기와 결합을 통해 인식률을 제고하는 방법을 제시한다. 필기체 숫자 데이터베이스에 대해서 제안한 인식기를 구성하고 특성을 분석한다. 제안한 방법은 다른 인식기들에 비해서 신경망에 사용된 가중치의 수는 증가하지만 가장 우수한 인식 성능을 나타내었다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 컴퓨터소사이어티 추계학술대회논문집
• /
• pp.99-102
• /
• 2003

In this paper we propose an adaptive learning method for HMM classifiers by using splitting and merging techniques to overcome the problem of the conventional teaming, where one HMM classifier per class has been trained, individually. The experimental results demonstrate a possibility that the proposed mechanism could be applied for applications of having multiple clusters in a class.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2006년도 추계학술대회논문집
• /
• pp.411-416
• /
• 2006

Fault detection and diagnosis is the most important technology in condition-based maintenance(CBM) system that usually begins from collecting signatures of running machines using multiple sensors for subsequent accurate analysis. With the quick development in industry, there is an increasing requirement of selecting special sensors that are cheap, robust, and easy-installation. This paper experimentally investigated performances of four types of sensors used in induction motors faults diagnosis, which are vibration, current, voltage and flux. In addition, diagnostic effects of five popular classifiers also were evaluated. First, the raw signals from the four types of sensors are collected at the same time. Then the features are calculated from collected signals. Next, these features are classified through five classifiers using artificial intelligence techniques. Finally, conclusions are given based on the experiment results.

• 한국IT서비스학회지
• /
• 제14권4호
• /
• pp.121-135
• /
• 2015

Ensemble classification is to utilize multiple classifiers instead of using a single classifier. Recently ensemble classifiers have attracted much attention in data mining community. Ensemble learning techniques has been proved to be very useful for improving the prediction accuracy. Bagging, boosting and random subspace are the most popular ensemble methods. In random subspace, each base classifier is trained on a randomly chosen feature subspace of the original feature space. The outputs of different base classifiers are aggregated together usually by a simple majority vote. In this study, we applied the random subspace method to the bankruptcy problem. Moreover, we proposed a method for optimizing the random subspace ensemble. The genetic algorithm was used to optimize classifier subset of random subspace ensemble for bankruptcy prediction. This paper applied the proposed genetic algorithm based random subspace ensemble model to the bankruptcy prediction problem using a real data set and compared it with other models. Experimental results showed the proposed model outperformed the other models.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권3호
• /
• pp.163-169
• /
• 2008

In this paper, we propose a real-time people tracking system with multiple CCD cameras for security inside the building. The camera is mounted from the ceiling of the laboratory so that the image data of the passing people are fully overlapped. The implemented system recognizes people movement along various directions. To track people even when their images are partially overlapped, the proposed system estimates and tracks a bounding box enclosing each person in the tracking region. The approximated convex hull of each individual in the tracking area is obtained to provide more accurate tracking information. To achieve this goal, we propose a method for 3D walking human tracking based on the IMPRESARIO framework incorporating cascaded classifiers into hypothesis evaluation. The efficiency of adaptive selection of cascaded classifiers have been also presented. We have shown the improvement of reliability for likelihood calculation by using cascaded classifiers. Experimental results show that the proposed method can smoothly and effectively detect and track walking humans through environments such as dense forests.

• 대한원격탐사학회지
• /
• 제31권3호
• /
• pp.205-214
• /
• 2015

이 연구에서는 자료변환기법을 이용해 추출된 여러 특징과 다양한 분류방법론을 결합하여 다중시기 SAR 자료를 위한 새로운 토지피복 분류기법을 제안하였다. 먼저, 다중시기 SAR 자료로부터 원본자료와는 다른 새로운 정보를 추출하기 위해 주성분분석과 3차원 웨이블렛 변환을 이용한 자료변환을 수행하였다. 그리고 나서 최대우도법 분류자, 신경망, support vector machine을 포함한 세 가지 다른 분류자를 변환된 특징자료들과 원본 후방산란계수 자료를 포함한 세가지 자료에 적용하여 다양한 초기 분류 결과를 얻도록 한다. 이후 다수결규칙을 통해 모든 초기결과를 결합하여 최종 분류 결과를 생성하게 된다. 다중시기 ENVISAT ASAR 자료를 이용한 사례연구에서 모든 초기 결과는 사용한 특징자료와 분류자의 종류에 따라 매우 다양한 분류정확도를 보였다. 이러한 9개의 초기 분류 결과를 결합한 최종 분류 결과는 가장 높은 분류 정확도를 보여주고 있는데, 이는 각 초기 분류 결과가 토지피복을 결정하기 위한 상호 보완적인 정보를 제공하기 때문이다. 이 연구에서의 분류정확도 향상은 주로 자료변환을 통해 얻어진 각기 다른 특징자료와 다른 분류자를 결합에 의한 다양성 확보에서 기인한다. 그러므로 이 연구에서 제안한 토지피복 분류방법론은 다중시기 SAR자료의 분류에 효과적으로 적용가능하며, 또한 다중센서 원격탐사 자료융합으로 확장이 가능하다.

• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1175-1186
• /
• 2017

Human emotion recognition is one of the promising applications in the era of artificial super intelligence. Thus far, facial expression traits are considered to be the most widely used information cues for realizing automated emotion recognition. This paper proposes a novel facial expression recognition (FER) method that works well for recognizing emotion from image sequences. To this end, we develop the so-called weighted soft voting classification (WSVC) algorithm. In the proposed WSVC, a number of classifiers are first constructed using different and multiple feature representations. In next, multiple classifiers are used for generating the recognition result (namely, soft voting) of each face image within a face sequence, yielding multiple soft voting outputs. Finally, these soft voting outputs are combined through using a weighted combination to decide the emotion class (e.g., anger) of a given face sequence. The weights for combination are effectively determined by measuring the quality of each face image, namely "peak expression intensity" and "frontal-pose degree". To test the proposed WSVC, CK+ FER database was used to perform extensive and comparative experimentations. The feasibility of our WSVC algorithm has been successfully demonstrated by comparing recently developed FER algorithms.

• 전자공학회논문지CI
• /
• 제47권2호
• /
• pp.21-28
• /
• 2010

본 논문에서는 동적시간교정법(dynamic time warping: DTW)과 다중퓨전기법(multiple fusion strategy: MFS)을 연속 적용하여 비유사도기반 분류법(dissimilarity-based classification: DBC)을 최적화시키는 방법의 실험결과를 보고한다. DBC란 샘플패턴을 분류하기 위하여 샘플의 특징 값을 이용하는 대신에 샘플들 사이의 비유사도를 측정하여 분류기를 설계하는 방법이다. DTW에서는 다음과 같이 두 단계로 나누어 비유사도를 측정한다. 먼저 상관계수를 이용하여 객체 샘플들을 대응시키기 위한 최적의 대응경로를 찾을 수 있도록 샘플들을 조정한다. 그리고 기존의 거리측정법으로 조정된 샘플들 사이의 비유사도를 측정한다. MFS에서는 분류기결합 뿐만 아니라 비유사도 행렬생성에서도 퓨전기법을 적용한다. 즉, DTW 기법으로 작성한 다수의 비유사도 행렬들을 결합하여 새로운 비유사도 행렬을 생성한 다음, 이 행렬공간에서 여러 개의 베이스 분류기를 학습하여 다시 결합한다. 본 논문에서 제안한 방법을 벤취마크 영상 데이터베이스를 대상으로 실험한 결과, 기존의 방법과 비교하여 분류성능을 향상시킬 수 있음을 확인하였다. 이와 같은 실험결과로 볼 때, 제안 방법을 멀티미디어 정보검색 등과 같은 다른 고차원 응용에도 활용할 수 있을 것으로 사료된다.