• Title, Summary, Keyword: pattern classification

Search Result 1,773, Processing Time 0.039 seconds

Application of Sensor Network Using Multivariate Gaussian Function to Hand Gesture Recognition (Multivariate Gaussian 함수를 이용한 센서 네트워크의 수화 인식에의 적용)

  • Kim Sung-Ho;Han Yun-Jong;Bogdana Diaconescu
    • Journal of Institute of Control, Robotics and Systems
    • /
    • v.11 no.12
    • /
    • pp.991-995
    • /
    • 2005
  • Sensor networks are the results of convergence of very important technologies such as wireless communication and micro electromechanical systems. In recent years, sensor networks found a wide applicability in various fields such as health, environment and habitat monitoring, military, etc. A very important step for these many applications is pattern classification and recognition of data collected by sensors installed or deployed in different ways. But, pattern classification and recognition are sometimes difficult to perform. Systematic approach to pattern classification based on modern teaming techniques like Multivariate Gaussian mixture models, can greatly simplify the process of developing and implementing real-time classification models. This paper proposes a new recognition system which is hierarchically composed of many sensor nodes haying the capability of simple processing and wireless communication. The proposed system is able to perform classification of sensed data using the Multivariate Gaussian function. In order to verify the usefulness of the proposed system, it was applied to hand gesture recognition system.

Import Vector Voting Model for Multi-pattern Classification (다중 패턴 분류를 위한 Import Vector Voting 모델)

  • Choi, Jun-Hyeog;Kim, Dae-Su;Rim, Kee-Wook
    • Journal of the Korean Institute of Intelligent Systems
    • /
    • v.13 no.6
    • /
    • pp.655-660
    • /
    • 2003
  • In general, Support Vector Machine has a good performance in binary classification, but it has the limitation on multi-pattern classification. So, we proposed an Import Vector Voting model for two or more labels classification. This model applied kernel bagging strategy to Import Vector Machine by Zhu. The proposed model used a voting strategy which averaged optimal kernel function from many kernel functions. In experiments, not only binary but multi-pattern classification problems, our proposed Import Vector Voting model showed good performance for given machine learning data.

Pattern Classification of the QRS-complexes Using Relational Correlation (관계상관식을 이용한 QRS 패턴분류)

  • Hwang, Seon-Cheol;Jeong, Hee-Kyo;Shin, Kun-Soo;Lee, Byung-Chae;Lee, Myoung-Ho
    • Proceedings of the KIEE Conference
    • /
    • /
    • pp.428-431
    • /
    • 1990
  • This paper describes a pattern classification algorithm of QRS-complexes using significant point detection for extracting features of signals. Significant point extraction was processed by zero-crossing method, and decision function based on relational spectrum was used for pattern classification of the QRS-complexes. The hierarchical AND/OR graph was obtained by decomposing the signal, and by use of this graph, QRS's patterns were classified. By using the proposed algorithm, the accuracy of pattern classification and the processing speed were improved.

  • PDF

ECG Pattern Classification Using Back Propagation Neural Network (역전달 신경회로망을 이용한 심전도 신호의 패턴분류에 관한 연구)

  • 이제석;이정환;권혁제;이명호
    • Journal of the Korean Institute of Telematics and Electronics B
    • /
    • v.30B no.6
    • /
    • pp.67-75
    • /
    • 1993
  • ECG pattern was classified using a back-propagation neural network. An improved feature extractor of ECG is proposed for better classification capability. It is consisted of preprocessing ECG signal by an FIR filter faster than conventional one by a factor of 5. QRS complex recognition by moving-window integration, and peak extraction by quadratic approximation. Since the FIR filter had a periodic frequency spectrum, only one-fifth of usual processing time was required. Also, segmentation of ECG signal followed by quadratic approximation of each segment enabled accurate detection of both P and T waves. When improtant features were extracted and fed into back-propagation neural network for pattern classification, the required number of nodes in hidden and input layers was reduced compared to using raw data as an input, also reducing the necessary time for study. Accurate pattern classification was possible by an appropriate feature selection.

  • PDF

Integrated GUI Environment of Parallel Fuzzy Inference System for Pattern Classification of Remote Sensing Images

  • Lee, Seong-Hoon;Lee, Sang-Gu;Son, Ki-Sung;Kim, Jong-Hyuk;Lee, Byung-Kwon
    • International Journal of Fuzzy Logic and Intelligent Systems
    • /
    • v.2 no.2
    • /
    • pp.133-138
    • /
    • 2002
  • In this paper, we propose an integrated GUI environment of parallel fuzzy inference system fur pattern classification of remote sensing data. In this, as 4 fuzzy variables in condition part and 104 fuzzy rules are used, a real time and parallel approach is required. For frost fuzzy computation, we use the scan line conversion algorithm to convert lines of each fuzzy linguistic term to the closest integer pixels. We design 4 fuzzy processor unit to be operated in parallel by using FPGA. As a GUI environment, PCI transmission, image data pre-processing, integer pixel mapping and fuzzy membership tuning are considered. This system can be used in a pattern classification system requiring a rapid inference time in a real-time.

Frequent Pattern Bayesian Classification for ECG Pattern Diagnosis (심전도 패턴 판별을 위한 빈발 패턴 베이지안 분류)

  • Noh, Gi-Yeong;Kim, Wuon-Shik;Lee, Hun-Gyu;Lee, Sang-Tae;Ryu, Keun-Ho
    • The KIPS Transactions:PartD
    • /
    • v.11D no.5
    • /
    • pp.1031-1040
    • /
    • 2004
  • Electrocardiogram being the recording of the heart's electrical activity provides valuable clinical information about heart's status. Many re-searches have been pursued for heart disease diagnosis using ECG so far. However, electrocardio-graph uses foreign diagnosis algorithm due to inaccuracy of diagnosis results for a heart disease. This paper suggests ECG data collection, data preprocessing and heart disease pattern classification using data mining. This classification technique is the FB(Frequent pattern Bayesian) classifier and is a combination of two data mining problems, naive bayesian and frequent pattern mining. FB uses Product Approximation construction that uses the discovered frequent patterns. Therefore, this method overcomes weakness of naive bayesian which makes the assumption of class conditional independence.

Implementation of simple statistical pattern recognition methods for harmful gases classification using gas sensor array fabricated by MEMS technology (MEMS 기술로 제작된 가스 센서 어레이를 이용한 유해가스 분류를 위한 간단한 통계적 패턴인식방법의 구현)

  • Byun, Hyung-Gi;Shin, Jeong-Suk;Lee, Ho-Jun;Lee, Won-Bae
    • Journal of Sensor Science and Technology
    • /
    • v.17 no.6
    • /
    • pp.406-413
    • /
    • 2008
  • We have been implemented simple statistical pattern recognition methods for harmful gases classification using gas sensors array fabricated by MEMS (Micro Electro Mechanical System) technology. The performance of pattern recognition method as a gas classifier is highly dependent on the choice of pre-processing techniques for sensor and sensors array signals and optimal classification algorithms among the various classification techniques. We carried out pre-processing for each sensor's signal as well as sensors array signals to extract features for each gas. We adapted simple statistical pattern recognition algorithms, which were PCA (Principal Component Analysis) for visualization of patterns clustering and MLR (Multi-Linear Regression) for real-time system implementation, to classify harmful gases. Experimental results of adapted pattern recognition methods with pre-processing techniques have been shown good clustering performance and expected easy implementation for real-time sensing system.

A Study of Efficient Pattern Classification on Texture Feature Representation Coordinate System (텍스처 특징 표현 좌표체계에서의 효율적인 패턴 분류 방법에 대한 연구)

  • Woo, Kyeong-Deok;Kim, Sung-Gook;Baik, Sung-Wook
    • Journal of Korea Multimedia Society
    • /
    • v.13 no.2
    • /
    • pp.237-248
    • /
    • 2010
  • When scenes in the real world are perceived for the purpose of computer/robot vision fields, there are great deals of texture based patterns in them. This paper introduces a texture feature representation on a coordinate system in which many different patterns can be represented with a mathematical model (Gabor function). The representation of texture features of each pattern on the coordinate system results in the high performance/competence of texture pattern classification. A decision tree algorithm is used to classify pattern data represented on the proposed coordinate system. The experimental results for the texture pattern classification show that the proposed method is better than previous researches.

Design of ECG Pattern Classification System Using Fuzzy-Neural Network (퍼지-뉴럴 네트워크를 이용한 심전도 패턴 분류시스템 설계)

  • 김민수;이승로;서희돈
    • Proceedings of the IEEK Conference
    • /
    • /
    • pp.273-276
    • /
    • 2002
  • This paper has design of ECG pattern classification system using decision of fuzzy IF-THEN rules and neural network. each fuzzy IF-THEN rule in our classification system has antecedent lingustic values and a single consequent class. we use a fuzzy reasoning method based on a single winner rule in the classification phase. this paper in, the MIT/BIH arrhythmia database for the source of input signal is used in order to evaluate the performance of the proposed system. From the simulation results, we can effectively pattern classification by application of learned from neural networks.

  • PDF

Development of An Operation Monitoring System for Intelligent Dust Collector By Using Multivariate Gaussian Function (Multivariate Gaussian Function을 이용한 지능형 집진기 운전상황 모니터링 시스템 개발)

  • Han, Yun-Jong;Kim, Sung-Ho
    • Proceedings of the KIEE Conference
    • /
    • /
    • pp.470-472
    • /
    • 2006
  • Sensor networks are the results of convergence of very important technologies such as wireless communication and micro electromechanical systems. In recent years, sensor networks found a wide applicability in various fields such as environment and health, industry scene system monitoring, etc. A very important step for these many applications is pattern classification and recognition of data collected by sensors installed or deployed in different ways. But, pattern classification and recognition are sometimes difficult to perform. Systematic approach to pattern classification based on modem learning techniques like Multivariate Gaussian mixture models, can greatly simplify the process of developing and implementing real-time classification models. This paper proposes a new recognition system which is hierarchically composed of many sensor nodes having the capability of simple processing and wireless communication. The proposed system is able to perform context classification of sensed data using the Multivariate Gaussian function. In order to verify the usefulness of the proposed system, it was applied to intelligent dust collecting system.

  • PDF