• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.5
• /
• pp.985-993
• /
• 2001

Neural networks are shown to be effective in being able to distinguish incomplete penetration-like weld defects by directly analyzing the plasma which is generated on each impingement of the laser on the materials. The performance is similar to that of existing methods based on extracted feature parameters. In each case around 93% of the defects in a database derived from 100 artificially produced defects of known types can be placed into one of two classes: incomplete penetration and bubbling. The present method based on classification using plasma is faster, and the speed is sufficient to allow on-line classification during data collection.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.1015-1020
• /
• 2003

Among the techniques to protect private information by adopting biometrics, speaker verification is expected to be widely used due to advantages in convenient usage and implementation cost. Speaker verification should achieve a high degree of the reliability in the verification score, the flexibility in speech text usage, and the efficiency in verification system complexity. Continuants have excellent speaker-discriminant power and the modest number of phonemes in the category, and multilayer perceptrons (MLPs) have superior recognition ability and fast operation speed. In consequence, the two provide viable ways for speaker verification system to obtain the above properties. This paper implements a system to which continuants and MLPs are applied, and evaluates the system using a Korean speech database. The results of the experiment prove that continuants and MLPs enable the system to acquire the three properties.

• Journal of KIISE:Software and Applications
• /
• v.34 no.10
• /
• pp.910-917
• /
• 2007

Of the possible interactions between human and robot, touch is an important means of providing human beings with emotional relief. However, most previous studies have focused on interactions based on voice and images. In this paper. a method of recognizing human touching behaviors is proposed for developing a robot that can naturally interact with humans through touch. In this method, the recognition process is divided into pre-process and recognition Phases. In the Pre-Process Phase, recognizable characteristics are calculated from the data generated by the touch detector which was fabricated using force sensors. The force sensor used an FSR (force sensing register). The recognition phase classifies human touching behaviors using a multi-layer perceptron which is a neural network model. Experimental data was generated by six men employing three types of human touching behaviors including 'hitting', 'stroking' and 'tickling'. As the experimental result of a recognizer being generated for each user and being evaluated as cross-validation, the average recognition rate was 82.9% while the result of a single recognizer for all users showed a 74.5% average recognition rate.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.535-538
• /
• 1990

The main problem in speech recognition is the enormous variability in acoustic signals due to complex but predictable contextual effects. Especially in plosive consonants it is very difficult to find invariant cue due to various contextual effects, but humans use these contextual effects as helpful information in plosive consonant recognition. In this paper we experimented on three artificial neural net models for the recognition of plosive consonants. Neural Net Model I used "Multi-layer Perceptron ". Model II used a variation of the "Self-organizing Feature Map Model". And Model III used "Interactive and Competitive Model" to experiment contextual effects. The recognition experiment was performed on 9 Korean plosive consonants. We used VCV speech chains for the experiment on contextual effects. The speech chain consists of Korean plosive consonants /g, d, b, K, T, P, k, t, p/ (/ㄱ, ㄷ, ㅂ, ㄲ, ㄸ, ㅃ, ㅋ, ㅌ, ㅍ/) and eight Korean monothongs. The inputs to Neural Net Models were several temporal cues - duration of the silence, transition and vot -, and the extent of the VC formant transitions to the presence of voicing energy during closure, burst intensity, presence of asperation, amount of low frequency energy present at voicing onset, and CV formant transition extent from the acoustic signals. Model I showed about 55 - 67 %, Model II showed about 60%, and Model III showed about 67% recognition rate.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.55 no.2
• /
• pp.45-51
• /
• 2006

In this paper, we propose a new architecture of Information Granulation based genetically optimized Self-Organizing Fuzzy Polynomial Neural Networks (IG_gSOFPNN) that is based on a genetically optimized multilayer perceptron with fuzzy polynomial neurons (FPNs) and discuss its comprehensive design methodology involving mechanisms of genetic optimization, especially information granulation and genetic algorithms. The proposed IG_gSOFPNN gives rise to a structurally optimized structure and comes with a substantial level of flexibility in comparison to the one we encounter in conventional SOFPNNs. The design procedure applied in the construction of each layer of a SOFPNN deals with its structural optimization involving the selection of preferred nodes (or FPNs) with specific local characteristics (such as the number of input variables, the order of the polynomial of the consequent part of fuzzy rules, and a collection of the specific subset of input variables) and addresses specific aspects of parametric optimization. In addition, the fuzzy rules used in the networks exploit the notion of information granules defined over system's variables and formed through the process of information granulation. That is, we determine the initial location (apexes) of membership functions and initial values of polynomial function being used in the premised and consequence part of the fuzzy rules respectively. This granulation is realized with the aid of the hard c-menas clustering method (HCM). To evaluate the performance of the IG_gSOFPNN, the model is experimented with using two time series data(gas furnace process and NOx process data).

• Proceedings of the Korea Society of Environmental Toocicology Conference
• /
• 2002.10a
• /
• pp.156-156
• /
• 2002

The goal of this study is to develop a biomarker used in monitoring abnormal behaviors of Japanese medaka (Oryzias latipes) as a model organism caused by hazardous chemicals that are toxic and persistent in the ecosystem. A widely used insecticide, diazinon (O, O-diethyl O- (2-isopropyl-4-methyl-6-pyrimidinyl) phosphorothioate), is highly neurotoxic to fish, and it is also well known that it causes vertebral malformation and behavioral changes of fish at relatively low concentrations. The fish behaviors were observed on a real time basis using an image processing and automatic data acquisition system. The genes potentially involved in the abnormal behaviors were cloned using suppression subtractive hybridization (SSH) technique. The untreated individuals showed common behavioral characteristics. When the test fish was affected by diazinon at a concentration of 0.1 and 1 ppm, some specific patterns were observed in its behavioral activity and locomotive tracks. The typical patterns were enhanced surfacing activity, opercular movement, erratic movement, tremors and convulsions as reported previously. The number of genes up-regulated tty diazinon treatment were 97 which includes 27 of unknown genes. The number of down-regulated genes were 99 including 60 of unknown genes. These gene expression patterns will be analyzed by the artificial neural networks such as self organization map (SOM) and multilayer perceptron (MLP), revealing the role of genes responsible for the behaviors. These results may provide molecular biological and neurobehavioral bases of a biomonitoring system for diazinon using a model organism such as fish.

• The Journal of Korean Physical Therapy
• /
• v.27 no.4
• /
• pp.214-220
• /
• 2015

Purpose: The purpose of this study was to investigate the availability of software for rehabilitation with the Kinect sensor by presenting an efficient algorithm based on machine learning when classifying the motion data of the PNF pattern if the subjects were wearing a patient gown. Methods: The motion data of the PNF pattern for upper extremities were collected by Kinect sensor. The data were obtained from 8 normal university students without the limitation of upper extremities. The subjects, wearing a T-shirt, performed the PNF patterns, D1 and D2 flexion, extensions, 30 times; the same protocol was repeated while wearing a patient gown to compare the classification performance of algorithms. For comparison of performance, we chose four algorithms, Naive Bayes Classifier, C4.5, Multilayer Perceptron, and Hidden Markov Model. The motion data for wearing a T-shirt were used for the training set, and 10 fold cross-validation test was performed. The motion data for wearing a gown were used for the test set. Results: The results showed that all of the algorithms performed well with 10 fold cross-validation test. However, when classifying the data with a hospital gown, Hidden Markov model (HMM) was the best algorithm for classifying the motion of PNF. Conclusion: We showed that HMM is the most efficient algorithm that could handle the sequence data related to time. Thus, we suggested that the algorithm which considered the sequence of motion, such as HMM, would be selected when developing software for rehabilitation which required determining the correctness of the motion.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.20 no.11
• /
• pp.1009-1017
• /
• 2010

As rotating machines play an important role in industrial applications such as aeronautical, naval and automotive industries, many researchers have developed various condition monitoring system and fault diagnosis system by applying various techniques such as signal processing and pattern recognition. Recently, fault diagnosis systems using artificial neural network have been proposed. For effective fault diagnosis, this paper used MLP(multi-layer perceptron) network which is widely used in pattern classification. Since using obtained signals without preprocessing as inputs of neural network can decrease performance of fault classification, it is very important to extract significant features of captured signals and to apply suitable features into diagnosis system according to the kinds of obtained signals. Therefore, this paper proposes the decision method of the proper feature vectors about each fault signal for neural-network-based fault diagnosis system. We applied LPC coefficients, maximum magnitudes of each spectral section in FFT and RMS(root mean square) and variance of wavelet coefficients as feature vectors and selected appropriate feature vectors as comparing error ratios of fault diagnosis for sound, vibration and current fault signals. From experiment results, LPC coefficients and maximum magnitudes of each spectral section showed 100 % diagnosis ratios for each fault and the method using wavelet coefficients had noise-robust characteristic.

• Journal of Korean Society of Water and Wastewater
• /
• v.25 no.4
• /
• pp.581-589
• /
• 2011

Generally, treated water or raw water is transported into storage reservoirs which are receiving facilities of local governments from multi-regional water supply systems. A water supply control and operation center is operated not only to manage the water facilities more economically and efficiently but also to mitigate the shortage of water resources due to the increase in water consumption. To achieve the goal, important information such as the flow-rate in the systems, water levels of storage reservoirs or tanks, and pump-operation schedule should be considered based on the resonable water demand forecasting. However, it is difficult to acquire the pattern of water demand used in local government, since the operating information is not shared between multi-regional and local water systems. The pattern of water demand is irregular and unpredictable. Also, additional changes such as an abrupt accident and frequent changes of electric power rates could occur. Consequently, it is not easy to forecast accurate water demands. Therefore, it is necessary to introduce a short-term water demands forecasting and to develop an application of the forecasting models. In this study, the forecasting simulator for water demand is developed based on mathematical and neural network methods as linear and non-linear models to implement the optimal water demands forecasting. It is shown that MLP(Multi-Layered Perceptron) and ANFIS(Adaptive Neuro-Fuzzy Inference System) can be applied to obtain better forecasting results in multi-regional water supply systems with a large scale and local water supply systems with small or medium scale than conventional methods, respectively.

• Journal of Wetlands Research
• /
• v.12 no.3
• /
• pp.79-97
• /
• 2010

The purpose of this research is to apply the neural networks models for the hydrologic disaggregation of the yearly pan evaporation(PE) data in Republic of Korea. The neural networks models consist of multilayer perceptron neural networks model(MLP-NNM) and support vector machine neural networks model(SVM-NNM), respectively. And, for the evaluation of the neural networks models, they are composed of training and test performances, respectively. The three types of data such as the historic, the generated, and the mixed data are used for the training performance. The only historic data, however, is used for the testing performance. The application of MLP-NNM and SVM-NNM for the hydrologic disaggregation of nonlinear time series data is evaluated from results of this research. Four kinds of the statistical index for the evaluation are suggested; CC, RMSE, E, and AARE, respectively. Homogeneity test using ANOVA and Mann-Whitney U test, furthermore, is carried out for the observed and calculated monthly PE data. We can construct the credible monthly PE data from the hydrologic disaggregation of the yearly PE data, and the available data for the evaluation of irrigation and drainage networks system can be suggested.