• The Journal of the Acoustical Society of Korea
• /
• v.18 no.3
• /
• pp.62-67
• /
• 1999

In this paper, we investigates a reliable model of the Predictive Recurrent Neural Network for the speech recognition. Predictive Neural Networks are modeled by syllable units. For the given input syllable, then a model which gives the minimum prediction error is taken as the recognition result. The Predictive Neural Network which has the structure of recurrent network was composed to give the dynamic feature of the speech pattern into the network. We have compared with the recognition ability of the Recurrent Network proposed by Elman and Jordan. ETRI's SAMDORI has been used for the speech DB. In order to find a reliable model of neural networks, the changes of two recognition rates were compared one another in conditions of: (1) changing prediction order and the number of hidden units: and (2) accumulating previous values with self-loop coefficient in its context. The result shows that the optimum prediction order, the number of hidden units, and self-loop coefficient have differently responded according to the structure of neural network used. However, in general, the Jordan's recurrent network shows relatively higher recognition rate than Elman's. The effects of recognition rate on the self-loop coefficient were variable according to the structures of neural network and their values.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.3
• /
• pp.99-103
• /
• 2021

Recently, convolutional neural networks (CNNs) have been widely utilized due to their outstanding performance in various computer vision fields. However, due to their computational-intensive and high memory requirements, it is difficult to deploy CNNs on hardware platforms that have limited resources, such as mobile devices and IoT devices. To address these limitations, a neural network compression research is underway to reduce the size of neural networks while maintaining their performance. This paper proposes a CNN compression technique that dynamically adjusts the thresholds of pruning, one of the neural network compression techniques. Unlike the conventional pruning that experimentally or heuristically sets the thresholds that determine the weights to be pruned, the proposed technique can dynamically find the optimal thresholds that prevent accuracy degradation and output the light-weight neural network in less time. To validate the performance of the proposed technique, the LeNet was trained using the MNIST dataset and the light-weight LeNet could be automatically obtained 1.3 to 3 times faster without loss of accuracy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.2
• /
• pp.397-402
• /
• 1999

In this paper, we derive dynamic equation of a simulated helicopter and propose the control method base on a neural network to control its attitude. The coupling coefficients are adjusted to minimize the error between the output of a control system and the reference value. The gain of the proposed controller are automatically adjusted by the back propagation of a neural network. Simulation results using MATLAB are given in this paper.

• Journal of KSNVE
• /
• v.9 no.4
• /
• pp.849-855
• /
• 1999

Neural Networks have shown potential to become an attractive alternative to classic methods for identification and control of non-linear dynamic systems. The purpose of this paper is to present an application of neural networks, that is a neural reconstruction of the input signal of a non-linear unknown system. This basic methodology could be used for practical purpose in several engineering fields. Clearly applications of the proposed scheme can be of interest for physical systems where a complete network of sensors measuring system inputs is not available. It should also be emphasized that the application of the reconstruction scheme is of little or no interest when the analyzed system works and operates at nominal conditions. In fact, only when failures and/or system anomailes occur, leasing to performance degradation and/or shutdown, the application of this scheme is of interest. The paper presents the results of the methodology applied to unknown non-linear dynamic systems and the robustness of the scheme to white and colored system noise was evaluated.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.105-106
• /
• 2017

본 논문에서는 최대 우도 기반 파라미터 생성 알고리즘을 적용하여 인공 신경망의 출력인 음향 파라미터 열의 정확성 및 자연성을 향상시키는 방법을 제안하였다. 인공 신경망의 출력으로 정적 특징벡터 뿐 만 아니라 동적 특징벡터도 함께 사용하였고, 미리 계산된 파라미터 분산을 파라미터 생성에 사용하였다. 추정된 정적, 동적 특징벡터의 평균, 분산을 EM 알고리즘에 적용하여 최대 우도 기준 파라미터를 추정할 수 있다. 제안된 알고리즘은 파라미터 생성 시 동적 특징벡터 및 분산을 함께 적용하여 시간축에서의 자연성을 향상시켰다. 제안된 알고리즘의 객관적 평가로 MCD, F0 의 RMSE 를 측정하였고, 주관적평가로 선호도 평가를 실시하였다. 그 결과 기존 알고리즘 대비 객관적, 주관적 성능이 향상되는 것을 검증하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.592-599
• /
• 2020

This paper proposes a Convolutional Recurrent Neural Net (CRNN) structure that can simultaneously reflect both static and dynamic characteristics of seismic waveforms for various earthquake events classification. Addressing various earthquake events, including not only micro-earthquakes and artificial-earthquakes but also macro-earthquakes, requires both effective feature extraction and a classifier that can discriminate seismic waveform under noisy environment. First, we extract the static characteristics of seismic waveform through an attention-based convolution layer. Then, the extracted feature-map is sequentially injected as input to a multi-input single-output Long Short-Term Memory (LSTM) network structure to extract the dynamic characteristic for various seismic event classifications. Subsequently, we perform earthquake events classification through two fully connected layers and softmax function. Representative experimental results using domestic and foreign earthquake database show that the proposed model provides an effective structure for various earthquake events classification.

• Journal of Intelligence and Information Systems
• /
• v.16 no.2
• /
• pp.95-108
• /
• 2010

In this paper, we present a hybrid neural network model for dynamic hand gesture recognition. The model consists of two modules, feature extraction module and pattern classification module. We first propose a modified CNN(convolutional Neural Network) a pattern recognition model for the feature extraction module. Then we introduce a weighted fuzzy min-max(WFMM) neural network for the pattern classification module. The data representation proposed in this research is a spatiotemporal template which is based on the motion information of the target object. To minimize the influence caused by the spatial and temporal variation of the feature points, we extend the receptive field of the CNN model to a three-dimensional structure. We discuss the learning capability of the WFMM neural networks in which the weight concept is added to represent the frequency factor in training pattern set. The model can overcome the performance degradation which may be caused by the hyperbox contraction process of conventional FMM neural networks. From the experimental results of human action recognition and dynamic hand gesture recognition for remote-control electric home appliances, the validity of the proposed models is discussed.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.5
• /
• pp.1393-1402
• /
• 1999

This paper proposes an efficient principal feature extraction of the image data using neural networks of a new learning algorithm. The proposed learning algorithm is a backpropagation(BP) algorithm based on the steepest descent and dynamic tunneling. The BP algorithm based on the steepest descent is applied for high-speed optimization, and the BP algorithm based on the dynamic tunneling is also applied for global optimization. Converging to the local minimum by the BP algorithm of steepest descent, the new initial weights for escaping the local minimum is estimated by the BP algorithm of dynamic tunneling. The proposed algorithm has been applied to the 3 image data of 12${\times}$12pixels and the Lenna image of 128${\times}$128 pixels respectively. The simulation results shows that the proposed algorithm has better performances of the convergence and the feature extraction, in comparison with those using the Sanger method and the Foldiak method for single-layer neural networks and the BP algorithm for multilayer neural network.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.1
• /
• pp.54-60
• /
• 2002

In this paper, we proposed a hybrid neural network-fuzzy controller which compensate a output of neural network controller. Even if learn by neural network controller, it can occur an bad results from disturbance or load variations. So in order to adjust above case, we used the fuzzy compensator to get an expected results. And the weight of main neural network can be changed with the result of loaming a inverse model neural network of Plant, so a expected dynamic characteristics of plant can be got. As the results of simulation through the second order plant, we confirmed that the proposed speed controller get a good response compare with a neural network controller. We implemented the controller using the DSP processor and applied in a hydraulic servo system. And then we observed an experimental results.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2006.06a
• /
• pp.336-343
• /
• 2006

본 논문에서는 여권 영 상에서 PCA 알고리즘을 이용한 얼굴 인증과 개선된 퍼지 신경망을 이용한 여권 코드 인식 방법을 제안한다. 본 논문에서는 여권영상에 대해 소벨 연산자를 이용하여 에지를 추출하고 에지가 추출된 영상을 수평 스미어링하여 여권코드 영역을 추출한다. 추출된 여권 코드 영역의 기울기를 검사하여 기울기 보정을 하고, 여권 코드 영역을 이진화 한다. 이진화된 여권 코드 영역에 대하여 8방향윤곽선 추적 알고리즘을 적용하여 여권 코드를 추출한다. 추출된 여권 코드는 퍼지 신경망을 개선하여 여권 코드 인식에 적용한다. 개선된 퍼지 신경 망은 입력층과 중간층 사이의 학습 구조로는 FCM 클러스터링 알고리즘을 적용하고 중간층과 출력층 사이의 학습은 일반화된 델타학습 방법을 적용한다. 그리고 학습 성능을 개선하기 위하여 중간층과 출력층의 가중치 조정에 적용되는 학습률을 동적으로 조정하기 위해 퍼지 제어 시스템을 적용한다. 제안된 퍼지 신경망은 목표값과 출력값의 차이에 대한 절대값이 ${\epsilon}$ 보다 적거나 같으면 정확으로 분류하고 크면 부정확으로 분류하여 정확의 총 개수를 퍼지 제어 시스템에 적용하여 학습률과 모멘텀을 동적으로 조정한다. 여권의 주어진 규격에 근거하여 사진 영역을 추출하고 추출된 사진 영역에 대하여 YCbCr와 RGB 정보를 이용하여 얼굴영역을 추출한다. 추출된 얼굴 영역을 PCA 알고리즘과 스냅샷(Snap-Shot) 방법을 적용하여 얼굴 영역의 위조를 판별한다. 제안된 방법의 여권 코드 인식과 얼굴 인증의 성능을 평가하기 위하여 실제 여권 영상에 적용한 결과, 기존의 방법보다 여권 코드 인식과 얼굴 인증에 있어서 효율적인 것을 확인하였다.s, whereas AVs provide much better security.크는 기준년도부터 2031년까지 5년 단위로 계획된 장래도로를 반영하여 구축된다. 교통주제도 및 교통분석용 네트워크는 국가교통DB구축사업을 통해 구축된 자료로서 교통체계효율화법 제9조의4에 따라 공공기관이 교통정책 및 계획수립 등에 활용할 수 있도록 제공하고 있다. 건설교통부의 승인절차를 거쳐 제공하며 활용 후에는 갱신자료 및 활용결과를 통보하는 과정을 거치도록 되어있다. 교통주제도는 국가의 교통정책결정과 관련분야의 기초자료로서 다양하게 활용되고 있으며, 특히 ITS 노드/링크 기본지도로 활용되는 등 교통 분야의 중요한 지리정보로서 구축되고 있다..20{\pm}0.37L$, 72시간에 $1.33{\pm}0.33L$로 유의한 차이를 보였으므로(F=6.153, P=0.004), 술 후 폐환기능 회복에 효과가 있다. 4) 실험군과 대조군의 수술 후 노력성 폐활량은 수술 후 72시간에서 실험군이 $1.90{\pm}0.61L$, 대조군이 $1.51{\pm}0.38L$로 유의한 차이를 보였다(t=2.620, P=0.013). 5) 실험군과 대조군의 수술 후 일초 노력성 호기량은 수술 후 24시간에서 $1.33{\pm}0.56L,\;1.00{\ge}0.28L$로 유의한 차이를 보였고(t=2.530, P=0.017), 술 후 72시간에서 $1.72{\pm}0.65L,\;1.33{\pm}0.3L$로 유의한 차이를 보였다(t=2.540, P=0.016). 6) 대상자의 술 후 폐환기능에 영향을 미치는 요인은 성별로 나타났다. 이에 따