• The KIPS Transactions:PartB
• /
• v.13B no.7 s.110
• /
• pp.689-698
• /
• 2006

Pattern recognition is the study of how machines can observe the environment, learn to distinguish patterns of interest from their background, and make sound and reasonable decisions about the categories of the patterns. Color image consists of various color patterns. And most pattern recognition methods use the information of color which has been trained and extract the feature of the color. This thesis extracts adaptively specific color feature from images with several limited colors. Because the number of the color patterns is limited, the distribution of the color in the image is similar. But, when there are some noises and distortions in the image, its distribution can be various. Therefore we cannot extract specific color regions in the standard image that is well expressed in special color patterns to extract, and special color regions of the image to test. We suggest new method to reduce the error of recognition by extracting the specific color feature adaptively for images with the low distortion, and six test images with some degree of noises and distortion. We consequently found that proposed method shouws more accurate results than those of statistical pattern recognition.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1994.06c
• /
• pp.121-124
• /
• 1994

음소를 인식의 기본으로 하는 한국어 음성인식 시스템을 구현하기 위한 기초 연구의 일환으로서 각 음소의 특징 가장 잘 표현하는 기준프레임 추출을 위한 연구를 수행하였다. 이를 위하여 먼저 선행 실험과 분산비 분석을 통해서 인식에 필요로한 시간 패턴의 길이를 추출한 후 이를 바탕으로 통계적 인식방법인 베이즈 결정법칙을 이용하여 시단 프레임으로부터 3프레임씩 시점을 1프레임씩 옮기면서 인식 실험을 해？여, 각 음소별 특징이 가장 풍부한 기준 프레임을 추출하였다. 그리고 이 기준 프레임을 중심으로 각 음소군별 인식 실험을 수행하여 그 결과를 시단을 기준으로 한 경우와 비교 검토하고 한국어 전 음소별로 확장하여 인식 실험을 실시하였다. 이 실험 결과 모음의 경우 시단으로부터 5프레임, 파열음은 시단에서부터 5프레임사이, 마찰음은 3프레임에서부터 10프레임까지, 파찰음은 5프레임까지, 비음과 유음의 경우 초성은 시단 프레임에서 6프레임, 종성은 종단으로부터 전 4프레임 구간이 인식률이 높게 나타나 이 부분의 특징이 인식에 가장 유효함을 알 수 있었다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.144-152
• /
• 1996

In this paper, a pulse-mode digital multilayer neural network with a massively parallel yet compact and flexible network architecture is presented. Algebraicneural operations are replaced by stochastic processes using pseudo-random pulse sequences and simple logic gates are used as basic computing elements. The distributions of the results from the stochastic processes are approximated using the hypergeometric distribution. A statistical model of the noise(error) is developed to estimate the relative accuracy associated with stochastic computing in terms of mean and variance. Numerical character recognition problems are applied to the network to evaluate the network performance and to justify the validity of analytic results based on the developed statistical model. The network architectures are modeled in VHDL using the mixed descriptions of gate-level and register transfer level (RTL). Experiments show that the statistical model successfully predicts the accuracy of the operations performed in the network and that the character classification rate of the network is competitive to that of ordinary Back-Propagation networks.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.592-594
• /
• 2001

본 연구에서는 음성인식에 있어서 음향모델의 고정도화를 위해 통계적 방법인 HMM과 시간동기형 Viterbi 알고리즘을 기반으로 한 세그멘트되지 않은 음성의 자동 세그멘테이션에 관한 연구를 수행하였다. 본 연구에서는 소량의 세그멘트된 음성에 대해 연속분포형 HMM 기본모델을 작성한 후 이를 표준패턴으로 사용하고, 세그멘트되지 않은 입력음성의 특징 피라미터에 대해 시간동기형 Viterbi 알고리즘의 프레임마다 최대가 되는 지점을 최적경계로 설정하고, 앞에서 구현 최적 경계 정보와 언어학적 지식인 발음사전 정보를 이용하여 음성을 세그멘테이션 하는 것이다. 본 연구와의 비교를 위해 HTK를 이용하여 위와 동일한 과정을 수행하였다. 이렇게 구한 음성의 세그멘테이션 정보를 이용하여 연속분포형 HMM 기본모델과 HTK의 CHMM 기본모델을 각각 작성한 후, 국어공학센터(KLE) 단어 데이터에 대해 단어인식 성능을 평가하였다. 실험결과, KLE 452 남성과 여성에 대해, 본 연구실 인식 시스템은 화자독립 단어인식률 89.4%, 85.1%, HTK의 화자독립 단어인식률 85.1%, 81.9%를 각각 얻었다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.5
• /
• pp.1033-1039
• /
• 2012

In this paper, a simple and robust algorithm is proposed for detecting each steel plate from a image which contains several steel plates. Steel plate is characterized by line edge, so line detection is a fundamental task for analyzing and understanding of steel plate images. To detect the line edge, the proposed algorithm uses the small eigenvalue analysis. The proposed approach scans an input edge image from the top left corner to the bottom right corner with a moving mask. A covariance matrix of a set of edge pixels over a connected region within the mask is determined and then the statistical and geometrical properties of the small eigenvalue of the matrix are explored for the purpose of straight line detection. Using the detected line edges, each plate is determined based on the directional information and the distance information of the line edges. The results of the experiments emphasize that the proposed algorithm detects each steel plate from a image effectively.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.6
• /
• pp.995-1001
• /
• 1995

The structure of a typical pattern recognition consists a pre-processing, a feature extraction(algorithm) and classification or recognition. In classification, when widely varying patterns exist in same category, we need the clustering which organize the similar patterns. Clustering algorithm is two approaches. Firs, statistical approaches which are k-means, ISODATA algorithm. Second, neural network approach which is T. Kohonen's LVQ(Learning Vector Quantization). Nikhil R. Palet al proposed the GLVQ(Generalized LVQ, 1993). This paper suggest the efficient feature extraction methods of handwritten numerals in GLVQ clustering network. We use the handwritten numeral data from 21's authors(ie, 200 patterns) and compare the proportion of misclassified patterns for each feature extraction methods. As results, when we use the projection combination method, the classification ratio is 98.5%.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.10B
• /
• pp.1111-1116
• /
• 2009

In this paper, we discuss distortion-tolerant pattern recognition using computational integral imaging reconstruction. Three-dimensional object information is captured by the integral imaging pick-up process. The captured information is numerically reconstructed at arbitrary depth-levels by averaging the corresponding pixels. We apply Fisher linear discriminant analysis combined with principal component analysis to computationally reconstructed images for the distortion-tolerant recognition. Fisher linear discriminant analysis maximizes the discrimination capability between classes and principal component analysis reduces the dimensionality with the minimum mean squared errors between the original and the restored images. The presented methods provide the promising results for the classification of out-of-plane rotated objects.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.1649-1652
• /
• 2012

본 연구에서는 사용자의 성격패턴을 사용자의 대화음성 정보만으로 자동 분류할 수 있는 방법과 이를 기반으로 사용자의 성격 맞춤형 학습전략을 적용하는 애플리케이션을 개발하는 것을 목적으로 하였다. 음성대화 속의 발화된 말의 빠르기(speech rate)나 말소리의 크기, 기본주파수(fundamental frequency)의 값과 그들의 변화패턴, 그리고 묵음구간의 여러 가지 통계적 정보 같은 비언어적 단서를 활용하여 성격패턴을 최고 86.3% 까지 정확하게 인식해 낼 수 있었다. 또한 성격 별 영어단어 학습방법을 개발하여 사전 및 사후테스트를 기반으로 실험한 결과 약 24% 성적 향상을 보였다. 이 연구를 통해 확보되는 원천기술은 각종 에듀테인먼트 콘텐츠에는 물론 로봇과의 대화시스템, 치료나 재활을 위한 기능성 콘텐츠 등에 유용하게 사용될 것이다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.139-146
• /
• 2014

This paper proposes an efficient method for automatically distinguishing various facial expressions. To recognize the emotions from facial expressions, the facial images are obtained by digital cameras, and a number of feature points were extracted. The extracted feature points are then transformed to 49-dimensional feature vectors which are robust to scale and translational variations, and the facial emotions are recognized by statistical pattern classifiers such Naive Bayes, MLP (multi-layer perceptron), and SVM (support vector machine). Based on the experimental results with 5-fold cross validation, SVM was the best among the classifiers, whose performance was obtained by 50.8% for 6 emotion classification, and 78.0% for 3 emotions.

• Smart Media Journal
• /
• v.5 no.3
• /
• pp.35-41
• /
• 2016

In this paper, we propose hand-gesture signal recognition based on EPS(Electronic Potential Sensor) using Deep learning model. Extracted signals which from Electronic field based sensor, EPS have much of the noise, so it must remove in pre-processing. After the noise are removed with filter using frequency feature, the signals are reconstructed with dimensional transformation to overcome limit which have just one-dimension feature with voltage value for using convolution operation. Then, the reconstructed signal data is finally classified and recognized using multiple learning layers model based on deep learning. Since the statistical model based on probability is sensitive to initial parameters, the result can change after training in modeling phase. Deep learning model can overcome this problem because of several layers in training phase. In experiment, we used two different deep learning structures, Convolutional neural networks and Recurrent Neural Network and compared with statistical model algorithm with four kinds of gestures. The recognition result of method using convolutional neural network is better than other algorithms in EPS gesture signal recognition.