• The Transactions of the Korea Information Processing Society
• /
• v.4 no.12
• /
• pp.3033-3044
• /
• 1997

This paper presents a system which recognizes dynamic hand gesture for virtual reality (VR). A dynamic hand gesture is a method of communication for human and computer who uses gestures, especially both hands and fingers. Since the human hands and fingers are not the same in physical dimension, the produced by two persons with their hands may not have the same numerical values where obtained through electronic sensors. To recognize meaningful gesture from continuous gestures which have no token of beginning and end, this system segments current motion states using the state automata. In this paper, we apply a fuzzy min-max neural network and feature analysis method using fuzzy logic for on-line pattern recognition.

• Journal of the Korea Convergence Society
• /
• v.12 no.12
• /
• pp.65-74
• /
• 2021

Recently, studies on Korean spelling correction have been actively conducted based on machine translation and automatic noise generation. These methods generate noise and use as train and data set. This has limitation in that it is difficult to accurately measure performance because it is unlikely that noise other than the noise used for learning is included in the test set In addition, there is no practical error type standard, so the type of error used in each study is different, making qualitative analysis difficult. This paper proposes new 'error type classification' for deep learning-based Korean spelling correction research, and error analysis perform on existing commercialized Korean spelling correctors (System A, B, C). As a result of analysis, it was found the three correction systems did not perform well in correcting other error types presented in this paper other than spacing, and hardly recognized errors in word order or tense.

• Journal of Internet Computing and Services
• /
• v.22 no.1
• /
• pp.23-30
• /
• 2021

This paper proposes a noise-tolerant image classification system using multiple autoencoders. The development of deep learning technology has dramatically improved the performance of image classifiers. However, if the images are contaminated by noise, the performance degrades rapidly. Noise added to the image is inevitably generated in the process of obtaining and transmitting the image. Therefore, in order to use the classifier in a real environment, we have to deal with the noise. On the other hand, the autoencoder is an artificial neural network model that is trained to have similar input and output values. If the input data is similar to the training data, the error between the input data and output data of the autoencoder will be small. However, if the input data is not similar to the training data, the error will be large. The proposed system uses the relationship between the input data and the output data of the autoencoder, and it has two phases to classify the images. In the first phase, the classes with the highest likelihood of classification are selected and subject to the procedure again in the second phase. For the performance analysis of the proposed system, classification accuracy was tested on a Gaussian noise-contaminated MNIST dataset. As a result of the experiment, it was confirmed that the proposed system in the noisy environment has higher accuracy than the CNN-based classification technique.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.10
• /
• pp.207-216
• /
• 2013

The aim of this study is to design a classifier for sleep stage analysis and select important feature set which shows sleep stage well based on physiological signals during sleep. Sleep has a significant effect on the quality of human life. When people undergo lack of sleep or sleep-related disease, they are likely to reduced concentration and cognitive impairment affects, etc. Therefore, there are a lot of research to analyze sleep stage. In this study, after acquisition physiological signals during sleep, we do pre-processing such as filtering for extracting features. The features are used input for the new combination algorithm using genetic algorithm(GA) and neural networks(NN). The algorithm selects features which have high weights to classify sleep stage. As the result of this study, accuracy of the algorithm is up to 90.26% with electroencephalography(EEG) signal and electrocardiography(ECG) signal, and selecting features are alpha and delta frequency band power of EEG signal and standard deviation of all normal RR intervals(SDNN) of ECG signal. We checked the selected features are well shown that they have important information to classify sleep stage as doing repeating the algorithm. This research could use for not only diagnose disease related to sleep but also make a guideline of sleep stage analysis.

• Journal of the Korean Society for Library and Information Science
• /
• v.56 no.3
• /
• pp.241-264
• /
• 2022

Bibliographic metadata can help researchers effectively utilize essential publications that they need and grasp academic trends of their own fields. With the manual creation of the metadata costly and time-consuming. it is nontrivial to effectively automatize the metadata construction using rule-based methods due to the immoderate variety of the article forms and styles according to publishers and academic societies. Therefore, this study proposes a two-step extraction process based on rules and deep neural networks for generating bibliographic metadata of scientific articlles to overcome the difficulties above. The extraction target areas in articles were identified by using a deep neural network-based model, and then the details in the areas were analyzed and sub-divided into relevant metadata elements. IThe proposed model also includes a model for generating reference summary information, which is able to separate the end of the text and the starting point of a reference, and to extract individual references by essential rule set, and to identify all the bibliographic items in each reference by a deep neural network. In addition, in order to confirm the possibility of a model that generates the bibliographic information of academic papers without pre- and post-processing, we conducted an in-depth comparative experiment with various settings and configurations. As a result of the experiment, the method proposed in this paper showed higher performance.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.3
• /
• pp.60-68
• /
• 2007

In this paper, we propose an automatic facial expressions recognition system to analyze facial expressions (happiness, disgust, surprise and neutral) using tree structures based on heuristic rules. The facial region is first obtained using skin-color model and connected-component analysis (CCs). Thereafter the origins of user's eyes are localized using neural network (NN)-based texture classifier, then the facial features using some heuristics are localized. After detection of facial features, the facial expression recognition are performed using decision tree. To assess the validity of the proposed system, we tested the proposed system using 180 facial image in the MMI, JAFFE, VAK DB. The results show that our system have the accuracy of 93%.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.6
• /
• pp.518-523
• /
• 2015

This paper proposes a gait phase classifier using a Recurrent Neural Network (RNN). Walking is a type of dynamic system, and as such it seems that the classifier made by using a general feed forward neural network structure is not appropriate. It is known that an RNN is suitable to model a dynamic system. Because the proposed RNN is simple, we use a back propagation algorithm to train the weights of the network. The input data of the RNN is the lower body's joint angles and angular velocities which are acquired by using the lower limb exoskeleton robot, ROBIN-H1. The classifier categorizes a gait cycle as two phases, swing and stance. In the experiment for performance verification, we compared the proposed method and general feed forward neural network based method and showed that the proposed method is superior.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.2
• /
• pp.94-102
• /
• 2000

This paper presents a new FDI scheme based on dynamic fuzzy model(DFM) for the unknown nonlinear system, which can detect and isolate process faults continuously over all ranges of operating condition. The dynamic behavior of a nonlinear process is represented by a set of local linear models. The parameters of the DFM are identified by an on-line methods. The residual vector of the FDI system is consisted of the parameter deviations from nominal model and the set of grade of membership values indicating the operating condition of the nonlinear process. The detection and isolation of faults are performed via a neural network classifier that are learned the relationship between the residual vector and fault type. We apply the proposed FDI scheme to the FDI system design for a two-tank system and show the usefulness of the proposed scheme.

• The Journal of the Korea Contents Association
• /
• v.17 no.12
• /
• pp.281-288
• /
• 2017

Financial service commission notified that fingerprint information of their documents should be deleted till 2019 to the financial industry and the public institution. Business solutions for fingerprint detection and masking in document images are introduced. In this paper, a fingerprint information masking algorithm is proposed by using the multiple LBP features to extract fingerprint's intrinsic characteristics for artificial neural network decision whether the candidate is a true fingerprint or not after segmentation of versatile fingerprint candidates from a document image. The experimental results of the proposed fingerprint masking algorithm for 3,497 document images that are saved in a financial industry show that 96.4% of fingerprint information is masked, hence this fingerprint masking algorithm can be used efficiently in real fingerprint masking tasks.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.27 no.8
• /
• pp.1239-1249
• /
• 1990

In this paper, using the neural network design techniques, an adaptive Mahalanobis distance classifier(AMDC) is designed. This classifier has three layers: input layer, internal layer and output layer. The connection from input layer to internal layer is fully connected, and that from internal to output layer has partial connection that might be thought as an Oring. If two ormore clusters of patterns of one class are laid apart in the feature space, the network adaptively generate the internal nodes, whhch are corresponding to the subclusters of that class. The number of the output nodes in just same as the number of the classes to classify, on the other hand, the number of the internal nodes is defined by the number of the subclusters, and can be optimized by itself. Using the method of making the subclasses, the different patterns that are of the same class can easily be distinguished from other classes. If additional training is needed after the completion of the traning, the AMDC does not have to repeat the trainging that has already done. To test the performance of the AMDC, the experiments of classifying 500 Hangeuls were done. In experiment, 20 print font sets of Hangeul characters(10,000 cahracters) were used for training, and with 3 sets(1,500 characters), the AMDC was tested for various initial variance \ulcornerand threshold \ulcorner and compared with other statistical or neural classifiers.