• 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.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.10a
• /
• pp.565-568
• /
• 2008

In this paper, we present bit error rate(BER) performance of an adaptive decision feedback equalizer(DFE) with experimental data. The experiment was performed at the shore of Geoje in November 2007. The BER performance of the adaptive DFE whose tap weight is updated by RLS is described with change of feedforward tap number, feedback tap number, traning seqence length and delay, which shows that the uncoded average BER is $4{\times}10^{-2}\;and\;1.5{\times}10^{-2}$ with transmission range 9.7km and 4km, respectively.

• Korean Journal of Materials Research
• /
• v.8 no.3
• /
• pp.252-255
• /
• 1998

In this paper, we investigate the magneto-impedance(M1) effect of the FeCoSiB amorphous magnktic films. The amorphous magnetic film having near zero magnetostriction is fabricated by using the sputtering method, and then annealed in magnetic field. When the external magnetic field is directly applied to the fabricated film, the voltage amplitude between both side of the magnetic film varies about 76.2% at 120[MHzl and the impedance varies about 2.1%/0e. Thus, we find that the fabricated magnetic film has the characteristics of good sensor element.

• Journal of Oriental Neuropsychiatry
• /
• v.21 no.3
• /
• pp.151-161
• /
• 2010

Objectives : Postpartum depression is a kind of serious problem which influences on the parturients, her family. In this case we tried psychotherapy to postpartum depression patient. Methods : We treated the patient with Oriental psychotherapy. And then we compared postpartum depression before treatment with on after treatment by Korean version of Edinburgh Postnatal Depression Scale(K-EPDS). Results : After treatments, the clinical symptoms were improved and K-EPDS was reduced 16 to 6. Conclusions : Oriental psychotherapy is effective on postpartum depression.

• Anxiety and mood
• /
• v.6 no.1
• /
• pp.24-30
• /
• 2010

Objectives : Tic disorder is known to be a chronic neuro-behavioral disease, cognitive behavioral treatment (CBT) strategies, like habit reversal training (HRT), are introducing recently. We report the effectiveness of HRT in Tourette disorder, which are very common in clinical settings. Methods : The DSM-IV clinical diagnosis applied by child psychiatrist. YGTSS, Kovac's children's depression inventory, Spielberger State-Trait Anxiety Inventory, Abbreviated Conners' Rating Scales, Dupaul ADHD Rating Scales are used. Ultimately, totally 10 children were evaluated. Subject group are consist of 6 boys and 4 girls, and the mean age was $10.90{\pm}1.73$ years old. This study is treatment-refractory 10 patients (from 9-14 years old) though 1 years drug treatment and psychiatric consultation were taken. We administered 5 times of HRT for 4 weeks. Results and Conclusion : There were improvement of scores in Yale Global Tic Severity Scale, Clinical Global Improvement. Our observations indicate that HRT might be effective in the treatment of Tourette disorder.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.382-384
• /
• 2019

In this study, we analyzed the angle of shooting based on the footage of shooting training of basketball student athletes taken by the basketball team leaders. In the shooting video, the shot height of the player and the trajectory of the ball were drawn through a solid line, and the player was shown to guide the player to adjust the angle of projection to a situation where the success rate was higher. Through this study, the video shows that efficient guidance can be made available to basketball players.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.21 no.6
• /
• pp.153-160
• /
• 2011

Recently, the use of security surveillance system including CCTV is increasing due to the increase of terrors and crimes. At the same time, interest of face recognition at a distance using surveillance cameras has been increasing. Accordingly, we analyzed the performance of face recognition according to distance using PCA-based face recognition and interpolation. In this paper, we used Nearest, Bilinear, Bicubic, Lanczos3 interpolations to interpolate face image. As a result, we confirmed that existing interpolation have an few effect on performance of PCA-based face recognition and performance of PCA-based face recognition is improved by including face image according to distance in traning data.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.5 no.2
• /
• pp.183-188
• /
• 2002

Anti-fungal potential of 5 plant extracts viz., Eucalyptus citriodora, Allium sativum, Cassia sophera, Chromolaena odorata and Datura metel on the growth of mulberry brown leaf spot pathogen Myrothecium roridum were examined. Except fur the aqueous extract of Allium bulb, ethanolic leaf extract of all other plants more efficiently reduced the colony growth of the fungus on potato-dextrose-agar, Of which, Allium and Eucalyptus extracts were more effective. Initiation of radial growth of M. roridum on solid media was deferred maximum 6 days by ethanolic Eucalyptus extract and 4 days by aqueous Allium extract at $0.4 mg.ml^{-1}$. In the liquid media amended with Eucalyptus extract ($0.4 mg.ml^{-1}$) complete inhibition of sporulation was noticed upto 8 days, and initial inhibition of mycelial bio-mass generation was considerably diminished with time and reduction was 1.3 fold 14 days after application. While, complete inhibition of mycelial growth for 6-14 days was recorded with $\geq$0.1 mg.ml$^{-1}$ commercial eucalyptus oil. However, rejuvenation of growth appeared when fungus was re-inoculated in fresh media. Post-inoculate application of different doses Of Eucalyptus and Allium extracts significantly (p < 0.05) reduced the disease severity in pot-ted mulberry. However, persistence of the effect up to 28 days was apparent at $\geq$ 1.0 mg.ml$^{-1}$ and effectively was on par with carbendazim (1 mg.ml$^{-1}$ ). Almost equal control ability of 1.0 mg.ml$^{-1}$ Eucalyptus extracts can be achieved by ca. 10 times lowered dose of commercial eucalyptus oil. It seems, the toxic principle of E. citrodora to M. roridum is fungistatic in nature and may have essential oil based origin.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.2
• /
• pp.163-171
• /
• 1997

Function approximation from a set of input-output pairs has numerous applications in scientific and engineering areas. Multilayer feedforward neural networks have been proposed as a good approximator of nonlinear function. The back propagation(BP) algorithm allows multilayer feedforward neural networks to learn input-output mappings from training samples. It iteratively adjusts the network parameters(weights) to minimize the sum of squared approximation errors using a gradient descent technique. However, the mapping acquired through the BP algorithm may be corrupt when errorneous training data we employed. When errorneous traning data are employed, the learned mapping can oscillate badly between data points. In this paper we propose a robust BP learning algorithm that is resistant to the errorneous data and is capable of rejecting gross errors during the approximation process, that is stable under small noise perturbation and robust against gross errors.

• Journal of Biomedical Engineering Research
• /
• v.36 no.5
• /
• pp.228-234
• /
• 2015

The balance ability significantly decreased in the elderly because of deterioration of the neural musculature regulatory mechanisms. Several studies have investigated methods of improving balance ability using real-time systems, but it is limited by the expensive test equipment and specialized resources. Recently, Kinect system based on depth data has been applied to address these limitations. Little information about accuracy/inaccuracy of Kinect system is, however, available, particular in motion analysis for evaluation of effectiveness in rehabilitation training. Therefore, the aim of the current study was to evaluate accuracy/inaccuracy of Kinect system in specific rotational movement for balance rehabilitation training. Six healthy male adults with no musculoskeletal disorder were selected to participate in the experiment. Movements of the participants were induced by controlling the base plane of the balance training equipment in directions of AP (anterior-posterior), ML (medial-lateral), right and left diagonal direction. The dynamic motions of the subjects were measured using two Kinect depth sensor systems and a three-dimensional motion capture system with eight infrared cameras for comparative evaluation. The results of the error rate for hip and knee joint alteration of Kinect system comparison with infrared camera based motion capture system occurred smaller values in the ML direction (Hip joint: 10.9~57.3%, Knee joint: 26.0~74.8%). Therefore, the accuracy of Kinect system for measuring balance rehabilitation traning could improve by using adapted algorithm which is based on hip joint movement in medial-lateral direction.