• 한국멀티미디어학회논문지
• /
• 제4권6호
• /
• pp.524-533
• /
• 2001

연속음성인식 시스템의 실용화를 위해서 가장 중요한 것은 높은 인식 성능을 가지면서 동시에 실시간으로 인식되어야 한다. 이를 위하여 본 연구에서는 먼저 연속음성인식의 인식률 향상을 위하여 효과적인 음향모델을 구성하기 위하여 PDT-SSS(Phonetic Decision Tree-based Successive State Splitting) 알고리즘을 도입하여 HM-Net을 구성하고, 언어모델로서 반복학습을 이용하여 인식률 향상을 제고한다. 그리고, 기존의 연구에서 유효함이 입증된 프레임 단위 적응 프루닝 알고리즘을 연속음성에 적용하여 인식 속도를 개선하고자 한다. 제안된 방법의 유효성을 확인하기 위하여, 남성 4인이 항공편 예약 관련 음성에 대하여 인식 실험을 수행하였다. 그 결과 연속음성인식률 90.9%, 단어인식률 90.7%의 높은 인식성능을 얻었으며, 적응 프루닝 알고리즘을 적용한 경우 인식성능의 저하없이 약 1.2초(전체의 15%)의 인식시간을 줄일 수 있어 제안된 방법의 유효성을 확인할 수 있었다.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1992년도 학술논문발표회 논문집 제11권 1호
• /
• pp.61-65
• /
• 1992

This paper presents a study on Neural Networks for Phoneme Recognition and performs phoneme recognition using TDNN(Time Delay Neural Network). Also, this paper proposes new training algorithm for speech recognition using neural nets that proper to large scale TDNN. Because phoneme recognition is indispensable for continuous speech recognition, this paper uses TDNN to get accurate recognition result of phoneme. And this paper proposes new training algorithm that can converge TDNN to optimal state regardless of the number of phoneme to be recognized. The result of recognition on three phoneme classes shows recognition rate of 9.1%. And this paper proves that proposed algorithm is a efficient method for high performance and reducing convergence time.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 제11회 음성통신 및 신호처리 워크샵 논문집 (SCAS 11권 1호)
• /
• pp.358-363
• /
• 1994

This paper presents the improvement of the SCHMM using discrete VQ and One-Pass algorithm for keywords recognition in Korean continuous speech. The SCHMM using discrete VQ is a simple model that is composed of a variable mixture gaussian probability density function with dynamic mixture number. One-Pass algorithm is improved such that recognition rates are enhanced by fathoming any undesirable semisyllable with the low likelihood and the high duration penalty, and computation time is reduced by testing only the frame which is dissimilar to the previously testd frame. In recognition experiments for speaker-dependent case, the improved One-Pass algorithm has shown recognition rates as high as 99.7% and has reduced compution time by about 30% compared with the currently abailable one-pass algorithm.

• Journal of Information Processing Systems
• /
• 제7권3호
• /
• pp.425-434
• /
• 2011

The success of iris recognition depends mainly on two factors: image acquisition and an iris recognition algorithm. In this study, we present a system that considers both factors and focuses on the latter. The proposed algorithm aims to find out the most efficient wavelet family and its coefficients for encoding the iris template of the experiment samples. The algorithm implemented in software performs segmentation, normalization, feature encoding, data storage, and matching. By using the Haar and Biorthogonal wavelet families at various levels feature encoding is performed by decomposing the normalized iris image. The vertical coefficient is encoded into the iris template and is stored in the database. The performance of the system is evaluated by using the number of degrees of freedom, False Reject Rate (FRR), False Accept Rate (FAR), and Equal Error Rate (EER) and the metrics show that the proposed algorithm can be employed for an iris recognition system.

• 전자공학회논문지B
• /
• 제32B권12호
• /
• pp.1635-1643
• /
• 1995

This paper proposes an algorithm for machine recognition of phonemes in continuously spoken Korean. The proposed algorithm is a static strategy neural network. The algorithm uses, at the stage of training neurons, features such as the rate of zero crossing, short-term energy, and either PARCOR or auditory-like perceptual linear prediction(PLP) but not both, covering a time of 171ms long. Numerical results show that the algorithm with PLP achieves approximately the frame-based phoneme recognition rate of 99% for small vocabulary recognition experiments. Based on this it is concluded that the proposed algorithm with PLP analysis is effective in phoneme recognition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4573-4584
• /
• 2015

The traditional feature extraction algorithms for recognition of communication signals can hardly realize the balance between computational complexity and signals' interclass gathered degrees. They can hardly achieve high recognition rate at low SNR conditions. To solve this problem, a novel feature extraction algorithm based on Holder coefficient was proposed, which has the advantages of low computational complexity and good interclass gathered degree even at low SNR conditions. In this research, the selection methods of parameters and distribution properties of the extracted features regarding Holder coefficient theory were firstly explored, and then interval gray relation algorithm with improved adaptive weight was adopted to verify the effectiveness of the extracted features. Compared with traditional algorithms, the proposed algorithm can more accurately recognize signals at low SNR conditions. Simulation results show that Holder coefficient based features are stable and have good interclass gathered degree, and interval gray relation classifier with adaptive weight can achieve the recognition rate up to 87% even at the SNR of -5dB.

• 전기학회논문지
• /
• 제64권3호
• /
• pp.465-472
• /
• 2015

This paper presents a visual touch recognition for NUI(Natural User Interface) using Voronoi-tessellation algorithm. The proposed algorithms are three parts as follows: hand region extraction, hand feature point extraction, visual-touch recognition. To improve the robustness of hand region extraction, we propose RGB/HSI color model, Canny edge detection algorithm, and use of spatial frequency information. In addition, to improve the accuracy of the recognition of hand feature point extraction, we propose the use of Douglas Peucker algorithm, Also, to recognize the visual touch, we propose the use of the Voronoi-tessellation algorithm. Finally, we demonstrate the feasibility and applicability of the proposed algorithms through some experiments.

• ETRI Journal
• /
• 제40권6호
• /
• pp.774-787
• /
• 2018

Arabic handwriting segmentation and recognition is an area of research that has not yet been fully understood. Dealing with Arabic ligature segmentation, where the Arabic characters are connected and unconstrained naturally, is one of the fundamental problems when dealing with the Arabic script. Arabic character-recognition techniques consider ligatures as new classes in addition to the classes of the Arabic characters. This paper introduces an enhanced technique for Arabic handwriting recognition using the deep belief network (DBN) and a new morphological algorithm for ligature segmentation. There are two main stages for the implementation of this technique. The first stage involves an enhanced technique of the Sari segmentation algorithm, where a new ligature segmentation algorithm is developed. The second stage involves the Arabic character recognition using DBNs and support vector machines (SVMs). The two stages are tested on the IFN/ENIT and HACDB databases, and the results obtained proved the effectiveness of the proposed algorithm compared with other existing systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권10호
• /
• pp.4987-5005
• /
• 2017

This paper introduces a new method to solve Similarity Recognition for Cloth Products, which is based on Fuzzy logic and Particle swarm optimization algorithm. For convenience, it is called the SRCPFP method hereafter. In this paper, the SRCPFP method combines Fuzzy Logic (FL) and Particle Swarm Optimization (PSO) algorithm to solve similarity recognition for cloth products. First, it establishes three features, length, thickness, and temperature resistance, respectively, for each cloth product. Subsequently, these three features are engaged to construct a Fuzzy Inference System (FIS) which can find out the similarity between a query cloth and each sampling cloth in the cloth database D. At the same time, the FIS integrated with the PSO algorithm can effectively search for near optimal parameters of membership functions in eight fuzzy rules of the FIS for the above similarities. Finally, experimental results represent that the SRCPFP method can realize a satisfying recognition performance and outperform other well-known methods for similarity recognition under considerations here.

• 한국정밀공학회지
• /
• 제22권4호
• /
• pp.121-127
• /
• 2005

There is a necessity for the communication between intelligent robots and human beings because of wide spread use of them. Gesture recognition is currently being studied in regards to better conversing. On the basis of previous research, however, the gesture recognition algorithms appear to require not only complicated algorisms but also separate training process for high recognition rates. This study suggests a gesture recognition algorithm based on computer vision system, which is relatively simple and more efficient in recognizing various human gestures. After tracing the hand gesture using a marker, direction changes of the gesture trajectory were analyzed to determine the simple gesture code that has minimal information to recognize. A map is developed to recognize the gestures that can be expressed with different gesture codes. Through the use of numerical and geometrical trajectory, the advantages and disadvantages of the suggested algorithm was determined.