• 제목, 요약, 키워드: recognition

검색결과 18,545건 처리시간 0.074초

HMM을 기반으로 한 사전 확률의 문제점을 해결하기 위해 베이시안 기법 어휘 인식 모델에의 사후 확률을 융합한 잡음 제거 (Noise Removal using a Convergence of the posteriori probability of the Bayesian techniques vocabulary recognition model to solve the problems of the prior probability based on HMM)

  • 오상엽
    • 디지털융복합연구
    • /
    • v.13 no.8
    • /
    • pp.295-300
    • /
    • 2015
  • 사전 확률분포를 모델링하는 HMM을 사용하는 어휘 인식에서 인식 어휘의 모델들의 대한 인식 확률이 이산적인 분포를 나타내며 인식을 위한 계산량이 적은 장점이 있지만 인식률을 계산했을 때 상대적으로 낮은 단점이 있다. 이를 개선하기 위하여 베이시안 기법 어휘 인식 모델을 융합한 잡음 제거 인식률 향상을 제안한다. 본 논문은 베이시안 기법 어휘 인식을 위한 모델 구성을 베이시안 기법의 최적화한 인식 모델을 구성하였다. HMM을 기반으로 한 사전 확률 방법과 베이시안 기법인 사후확률을 융합하여 잡음을 제거하고 인식률을 향상시켰다. 본 논문에서 제안한 방법을 적용한 결과 어휘 인식률에서 98.1%의 인식률을 나타내었다.

MLHF 모델을 적용한 어휘 인식 탐색 최적화 시스템 (Vocabulary Recognition Retrieval Optimized System using MLHF Model)

  • 안찬식;오상엽
    • 한국컴퓨터정보학회논문지
    • /
    • v.14 no.10
    • /
    • pp.217-223
    • /
    • 2009
  • 모바일 단말기의 어휘 인식 시스템에서는 통계적 방법에 의한 어휘인식을 수행하고 N-gram을 이용한 통계적 문법 인식 시스템을 사용한다. 인식 대상이 되는 어휘의 수가 증가하면 어휘 인식 알고리즘이 복잡해지고 대규모의 탐색공간을 필요로 하게 되며 처리시간이 길어지므로 제한된 연산처리 능력과 메모리로는 처리하기가 불가능하다. 따라서 본 논문에서는 이러한 단점을 개선하고 어휘 인식을 최적화하기 위하여 MLHF 시스템을 제안한다. MLHF는 FLaVoR의 구조를 이용하여 음향학적 탐색과 언어적 탐색을 분리하여 음향학적 탐색에서는 HMM을 사용하고 언어적 탐색 단계에서는 Levenshtein distance 알고리즘을 사용한다. 시스템 성능 평가 결과 어휘 종속 인식률은 98.63%, 어휘 독립 인식률은 97.91%의 인식률을 나타냈으며 인식속도는 1.61초로 나타내었다.

  • PDF

영상 정규화 및 얼굴인식 알고리즘에 따른 거리별 얼굴인식 성능 분석 (Performance Analysis of Face Recognition by Distance according to Image Normalization and Face Recognition Algorithm)

  • 문해민;반성범
    • 정보보호학회논문지
    • /
    • v.23 no.4
    • /
    • pp.737-742
    • /
    • 2013
  • 최근 감시시스템은 휴먼인식 기술을 활용하여 스스로 판단하고 대처할 수 있는 지능형으로 발전하고 있다. 기존 얼굴인식 기술은 근거리에서 인식성능이 우수하지만 원거리로 갈수록 인식률이 떨어진다. 본 논문에서는 원거리 휴먼인식을 위해 거리별 얼굴영상을 학습으로 사용한 얼굴인식에서 보간법 및 얼굴인식 알고리즘에 따른 얼굴인식률의 성능을 분석한다. 영상 정규화에는 최근접 이웃, 양선형, 양3차회선, Lanczos3 보간법을 사용하고, 얼굴인식 알고리즘은 PCA와 LDA를 사용한다. 실험결과, 영상 정규화로 양선형 보간법과 얼굴인식 알고리즘으로 LDA를 사용했을 때 우수한 성능을 나타냄을 확인하였다.

바디 제스처 인식을 위한 기초적 신체 모델 인코딩과 선택적 / 비동시적 입력을 갖는 병렬 상태 기계 (Primitive Body Model Encoding and Selective / Asynchronous Input-Parallel State Machine for Body Gesture Recognition)

  • 김주창;박정우;김우현;이원형;정명진
    • 로봇학회논문지
    • /
    • v.8 no.1
    • /
    • pp.1-7
    • /
    • 2013
  • Body gesture Recognition has been one of the interested research field for Human-Robot Interaction(HRI). Most of the conventional body gesture recognition algorithms used Hidden Markov Model(HMM) for modeling gestures which have spatio-temporal variabilities. However, HMM-based algorithms have difficulties excluding meaningless gestures. Besides, it is necessary for conventional body gesture recognition algorithms to perform gesture segmentation first, then sends the extracted gesture to the HMM for gesture recognition. This separated system causes time delay between two continuing gestures to be recognized, and it makes the system inappropriate for continuous gesture recognition. To overcome these two limitations, this paper suggests primitive body model encoding, which performs spatio/temporal quantization of motions from human body model and encodes them into predefined primitive codes for each link of a body model, and Selective/Asynchronous Input-Parallel State machine(SAI-PSM) for multiple-simultaneous gesture recognition. The experimental results showed that the proposed gesture recognition system using primitive body model encoding and SAI-PSM can exclude meaningless gestures well from the continuous body model data, while performing multiple-simultaneous gesture recognition without losing recognition rates compared to the previous HMM-based work.

Proposed Efficient Architectures and Design Choices in SoPC System for Speech Recognition

  • Trang, Hoang;Hoang, Tran Van
    • 전기전자학회논문지
    • /
    • v.17 no.3
    • /
    • pp.241-247
    • /
    • 2013
  • This paper presents the design of a System on Programmable Chip (SoPC) based on Field Programmable Gate Array (FPGA) for speech recognition in which Mel-Frequency Cepstral Coefficients (MFCC) for speech feature extraction and Vector Quantization for recognition are used. The implementing process of the speech recognition system undergoes the following steps: feature extraction, training codebook, recognition. In the first step of feature extraction, the input voice data will be transformed into spectral components and extracted to get the main features by using MFCC algorithm. In the recognition step, the obtained spectral features from the first step will be processed and compared with the trained components. The Vector Quantization (VQ) is applied in this step. In our experiment, Altera's DE2 board with Cyclone II FPGA is used to implement the recognition system which can recognize 64 words. The execution speed of the blocks in the speech recognition system is surveyed by calculating the number of clock cycles while executing each block. The recognition accuracies are also measured in different parameters of the system. These results in execution speed and recognition accuracy could help the designer to choose the best configurations in speech recognition on SoPC.

Low-Quality Banknote Serial Number Recognition Based on Deep Neural Network

  • Jang, Unsoo;Suh, Kun Ha;Lee, Eui Chul
    • Journal of Information Processing Systems
    • /
    • v.16 no.1
    • /
    • pp.224-237
    • /
    • 2020
  • Recognition of banknote serial number is one of the important functions for intelligent banknote counter implementation and can be used for various purposes. However, the previous character recognition method is limited to use due to the font type of the banknote serial number, the variation problem by the solid status, and the recognition speed issue. In this paper, we propose an aspect ratio based character region segmentation and a convolutional neural network (CNN) based banknote serial number recognition method. In order to detect the character region, the character area is determined based on the aspect ratio of each character in the serial number candidate area after the banknote area detection and de-skewing process is performed. Then, we designed and compared four types of CNN models and determined the best model for serial number recognition. Experimental results showed that the recognition accuracy of each character was 99.85%. In addition, it was confirmed that the recognition performance is improved as a result of performing data augmentation. The banknote used in the experiment is Indian rupee, which is badly soiled and the font of characters is unusual, therefore it can be regarded to have good performance. Recognition speed was also enough to run in real time on a device that counts 800 banknotes per minute.

다중 신경망을 이용한 인식단위 결합 기반의 인쇄체 문자인식 (Machine Printed Character Recognition Based on the Combination of Recognition Units Using Multiple Neural Networks)

  • 임길택;김호연;남윤석
    • 정보처리학회논문지B
    • /
    • v.10B no.7
    • /
    • pp.777-784
    • /
    • 2003
  • 본 논문에서는 다중 신경망을 이용한 인식단위 결합 기반의 인쇄체 문자인식 방법을 제안한다. 입력 문자영상은 한글 문자 형식 6가지와 한글 이외의 기타 문자 형식의 전체 7가지 형식으로 분류되어 인식된다. 한글 문자는 2단계의 MLP 신경망 인식기에 의해 인식된다. 첫째 단계에서는 한글 문자를 자소의 조합 형태에 따라 2개 또는 3개의 인식단위로 나누고, 각 인식단위에서 추출된 방향각도 특징 벡터를 입력으로 하는 MLP 신경망으로 1차 인식한다. 둘째 단계에서는 첫째 단계의 인식단위별 MLP 신경망 인식기의 인식양상 특징을 추출하고 다른 MLP 신경망에 입력하여 최종 한글 문자인식을 한다. 한글 이외의 기타 문자의 인식을 위해서는 단일 MLP 신경망을 사용한다. 인식 실험에서는 실제 우편물 50,000통 영상으로부터 추출한 문자영상 데이터베이스를 이용하였는데, 실험 결과 본 논문에서 제안한 방법이 매우 우수함을 알 수 있었다.

얼굴 인식 Open API를 활용한 출입자 인식 시스템 개발 (Development of a Visitor Recognition System Using Open APIs for Face Recognition)

  • 옥기수;권동우;김현우;안동혁;주홍택
    • 정보처리학회논문지:컴퓨터 및 통신 시스템
    • /
    • v.6 no.4
    • /
    • pp.169-178
    • /
    • 2017
  • 최근 보안에 대한 관심과 필요성이 증가하면서 출입자 인식 시스템의 수요가 증대되고 있다. 출입자 인식 시스템은 출입자를 인식하기 위해서 다양한 생체인식 방법을 사용하고 있다. 본 논문에서는 다양한 특성과 강점을 가진 다수의 얼굴인식 Open API 서비스를 통합하고, 그 인식결과를 앙상블 함으로써 인식 성능을 개선하는 얼굴인식 기반 출입자 인식 시스템을 제안한다. 또한 다양한 얼굴 인식 Open API 서비스를 앙상블 하는 출입자 인식 시스템의 구조를 제안한다. 성능 측정은 약 5개월 간 수집한 얼굴 데이터를 이용하여 수행하였으며, 측정결과로 본 논문에서 제안하는 출입자 인식 시스템이 단일 얼굴인식 Open API 서비스를 사용했을 때보다 더 높은 얼굴인식률을 보임을 확인하였다.

도시 주부의 가정관리 능력의 제 영향 변인에 관한 연구 (A Study on Influential Variables Related to Home Management Ability of Urban Home Makers)

  • 이정우;오경희
    • 한국가정관리학회지
    • /
    • v.9 no.2
    • /
    • pp.1-18
    • /
    • 1991
  • The purpose of this study is to find out influential variables related to Home Management Ability of urban home makers. This study focuses on the following aspects; 1) to find out which variables of sociodemographic variables (ie. home maker's age, level of education-husband, wife, job-husband, wife, income, duration of marriage), of psychological variables (ie. degree of resourcefulness recognition, degree of stress recognition, degree of life level recognition) have significant effects on home management ability. 2) to find out which variables of sociodemographic variables have significant effects on degree of resourcefulness recognition, of stress recognition, and of life level recognition. 3) to identify the influence of significant variables related to home management ability. Data was analyzed by frequency. percentage, mean , F-test, t-test, Duncan's multiple range test. regression analysis , path analysis pearson's r. x2-test. Major findings are as follows; 1) The level of education (husband , wife)and occupation of husband were variables to have influences on home management ability. 2) a. The level of education (husband, wife) and income were variable to have influences on degree of resourcefulness recognition. b. The employment of home makers. income, and the form of family were variables to have influences on degree of stress recognition. c. The level of education (husband, wife) occupation of husband , income , and duration of marriage were variables to have influences on degree of life level recognition. 3) There were significant relationships between home management ability and degree of resourcefulness recognition and of stress recognition (r=0.13, r=-0.12, p<.05). a. The higher degree of resourcefulness recognition, the higher home management ability (x2=11.17. df=4. p<.05) b. The higher degree of stress recognition, the lower home n=management ability (x2=14.64. df=4. p<.01) 4) The education level of homemakers (β =0.15) and income (β=0.12) were variables to have indirect influences on home management ability through the medium of the degree of resourcefulness recognition (β =0.13) 5) The employment of home makers (β=-0.17) was a variable to have indirect influence on home management ability through the medium of the degree of stress recognition(β=-0.12) 6) the education level of husband (β=0.16) and income (β=0.32) were variables to have direct influence on degree of life level recognition. 7) The degree of life level recognition (β=0.13) and education level of home makers (β=0.17) were variables to have indirect influences on home management ability through the medium of the degree of resourcefulness recognition (β=0.13) 8)The degree of life level recognition (β=-0.22) the employment of home makers(β=-0.17) and the from of family(β=-0.10) were variables to have indirect influences on home management ability through the medium of the degree of stress recognition.

  • PDF

Improve Digit Recognition Capability of Backpropagation Neural Networks by Enhancing Image Preprocessing Technique

  • Feng, Xiongfeng;Kubik, K.Bogunia
    • 제어로봇시스템학회:학술대회논문집
    • /
    • /
    • pp.49.4-49
    • /
    • 2001
  • Digit recognition based on backpropagation neural networks, as an important application of pattern recognition, was attracted much attention. Although it has the advantages of parallel calculation, high error-tolerance, and learning capability, better recognition effects can only be achieved with some specific fixed format input of the digit image. Therefore, digit image preprocessing ability directly affects the accuracy of recognition. Here using Matlab software, the digit image was enhanced by resizing and neutral-rotating the extracted digit image, which improved the digit recognition capability of the backpropagation neural network under practical conditions. This method may also be helpful for recognition of other patterns with backpropagation neural networks.

  • PDF