• Convergence Security Journal
• /
• v.6 no.4
• /
• pp.83-90
• /
• 2006

In this paper, we propose a new design method for a robust vector quantizer that is independent of the statistical characteristics of input images in the wavelet transformed image coding. The conventional vector quantizers have failed to get quality coding results because of the different statistical properties between the image to be quantized and the training sequence for a codebook of the vector quantizer. Therefore, in order to solve this problem, we used a pseudo image as a training sequence to generate a codebook of the vector quantizer; the pseudo image is created by adding correlation coefficient and edge components to uniformly distributed random numbers. We will clearly define the problem of the conventional vector quantizers, which use real images as a training sequence to generate a codebook used, by comparing the conventional methods with the proposed through computer simulation. Also, we will show the proposed vector quantizer yields better coding results.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.2
• /
• pp.529-535
• /
• 2000

In this paper, we propose a new method of designing the vector quantizer which is robustness to coding results and independent of statistical characteristics of an input image in wavelet transformed image coding processes. The most critical drawback of a conventional vector quantizer is the degradation of coding capability resulted from the discordance between quantizer objective image and statistical characteristics of training sequence which is for generating representing vector. In order to resolve the problem of conventional methods, we use independent random-variables and pseudo image to which image correlation and edge component were added, as a training sequence for generating representing vector. We have done a computer simulation in order to compare coding capability between a vector quantizer designed by the proposed method and one with the conventional method using real image as same as that is objective to coding of training sequence used in codebook generation. The results show the superiority of the proposed vector quantizer method at the aspect of coding capability compared to conventional one. They also clarify the problems of conventional methods.

• The Journal of Korean Physical Therapy
• /
• v.34 no.6
• /
• pp.298-303
• /
• 2022

Purpose: The purpose of this study was to figure out how PAP (Post-Activation Potentiation) phenomenon affects short-term performance efficiently. Methods: This study was conducted with 18 Taekwondo athletes and 16 healthy adults. By using isokinetic dynamometer, two different intervention, TDP (Top-down program) and BUP (Bottom-up program), were performed to measure isokinetic parameter; (peak torque: PT, total work: TW, average power: AP, and average torque: AT) of knee extensor for intragroup, intergroup comparison and two-way ANOVA. Results: The Taekwondo athletes group showed statistically significant differences in all isokinetic parameters PT, TW, AP, and AT after TDP (p<0.05). However, in the healthy adult group, the difference in isokinetic parameters according to the exercise sequence was not statistically significant. (p>0.05). PT and TW at TDP were statistically significant (p<0.05) when the rate of change in TDP and BUP was compared and analyzed considering the difference in physical ability between the Taekwondo athlete group and the healthy adult group. However, AP and AT were not statistically significant. Finally, when examining the interaction between the two groups and two exercise sequence according to isokinetic parameters, only TW (p<0.05) showed a statistically significant interaction, while PT (P=0.099), AP (P=0.103), and AT (P=0.096) did not. This study suggests that short-term performance can be improved through the PAP phenomenon when TDP is applied to the Taekwondo group. Conclusion: According to our result, for Taekwondo athletes, if the goal is to improve short-term performance just before the main game, we suggest a training program through TDP.

• Physical Therapy Korea
• /
• v.10 no.3
• /
• pp.17-27
• /
• 2003

Repetitive transcranial magnetic stimulation (rTMS) modulates cortical excitability beyond the duration of the rTMS trains themselves. Depending on rTMS parameters, a lasting inhibition or facilitation of cortical excitability can be induced. Therefore, rTMS of high or low frequency over motor cortex may change certain aspects of motor learning performance and cortical activation. This study investigated the effect of high and low frequency subthreshold rTMS applied to the motor cortex on motor learning of sequential finger movements and brain activation using functional MRI (fMRI). Three healthy right-handed subjects (mean age 23.3) were enrolled. All subjects were trained with sequences of seven-digit rapid sequential finger movements, 30 minutes per day for 5 consecutive days using their left hand. 10 Hz (high frequency) and 1 Hz (low frequency) trains of rTMS with 80% of resting motor threshold and sham stimulation were applied for each subject during the period of motor learning. rTMS was delivered on the scalp over the right primary motor cortex using a figure-eight shaped coil and a Rapid(R) stimulator with two Booster Modules (Magstim Co. Ltd, UK). Functional MRI (fMRI) was performed on a 3T ISOL Forte scanner before and after training in all subjects (35 slices per one brain volume TR/TE = 3000/30 ms, Flip angle $60^{\circ}$, FOV 220 mm, $64{\times}64$ matrix, slice thickness 4 mm). Response time (RT) and target scores (TS) of sequential finger movements were monitored during the training period and fMRl scanning. All subjects showed decreased RT and increased TS which reflecting learning effects over the training session. The subject who received high frequency rTMS showed better performance in TS and RT than those of the subjects with low frequency or sham stimulation of rTMS. In fMRI, the subject who received high frequency rTMS showed increased activation of primary motor cortex, premotor, and medial cerebellar areas after the motor sequence learning after the training, but the subject with low frequency rTMS showed decreased activation in above areas. High frequency subthreshold rTMS on the motor cortex may facilitate the excitability of motor cortex and improve the performance of motor sequence learning in normal subject.

• Journal of KIISE
• /
• v.44 no.7
• /
• pp.674-679
• /
• 2017

In big-data environments wherein large amounts of text documents are produced daily, titles are very important clues that enable a prompt catching of the key ideas in documents; however, titles are absent for numerous document types such as blog articles and social-media messages. In this paper, a title-generation model for which sequence-to-sequence RNNs with attention and copying mechanisms are employed is proposed. For the proposed model, input sentences are encoded based on bi-directional GRU (gated recurrent unit) networks, and the title words are generated through a decoding of the encoded sentences with keywords that are automatically selected from the input sentences. Regarding the experiments with 93631 training-data documents and 500 test-data documents, the attention-mechanism performances are more effective (ROUGE-1: 0.1935, ROUGE-2: 0.0364, ROUGE-L: 0.1555) than those of the copying mechanism; in addition, the qualitative-evaluation radiative performance of the former is higher.

• Smart Media Journal
• /
• v.7 no.4
• /
• pp.17-23
• /
• 2018

We proposed a LSTM-based RNN model that can effectively perform the automatic spacing characteristics. For those long or noisy sentences which are known to be difficult to handle within Neural Network Learning, we defined a proper input data format and decoding data format, and added dropout, bidirectional multi-layer LSTM, layer normalization, and attention mechanism to improve the performance. Despite of the fact that Sejong corpus contains some spacing errors, a noise-robust learning model developed in this study with no overfitting through a dropout method helped training and returned meaningful results of Korean word spacing and its patterns. The experimental results showed that the performance of LSTM sequence-to-sequence model is 0.94 in F1-measure, which is better than the rule-based deep-learning method of GRU-CRF.

• Journal of information and communication convergence engineering
• /
• v.21 no.2
• /
• pp.159-166
• /
• 2023

The COI gene is a sequence of approximately 650 bp at the 5' terminal of the mitochondrial Cytochrome c Oxidase subunit I (COI) gene. As an effective DeoxyriboNucleic Acid (DNA) barcode, it is widely used for the taxonomic identification and evolutionary analysis of species. We created a CNN-LSTM hybrid model by combining the gene features partially extracted by the Long Short-Term Memory ( LSTM ) network with the feature maps obtained by the CNN. Compared to K-Means Clustering, Support Vector Machines (SVM), and a single CNN classification model, after training 278 samples in a training set that included 15 genera from two orders, the CNN-LSTM hybrid model achieved 94% accuracy in the test set, which contained 118 samples. We augmented the training set samples and four genera into four orders, and the classification accuracy of the test set reached 100%. This study also proposes calculating the cosine similarity between the training and test sets to initially assess the reliability of the predicted results and discover new species.

• Bioinformatics and Biosystems
• /
• v.1 no.1
• /
• pp.82-87
• /
• 2006

The most popular protein structure prediction method is comparative modeling. To guarantee accurate comparative modeling, the sequence alignment between a query protein and a template should be accurate. Although choosing the best template based on the protein sequence alignments is most critical to perform more accurate fold-recognition in comparative modeling, even more critical is the sequence alignment quality. Contrast to a lot of attention to developing a method for choosing the best template, prediction of alignment accuracy has not gained much interest. Here, we develop a method for prediction of the shift score, a recently proposed measure for alignment quality. We apply support vector regression (SVR) to predict shift score. The alignment between a query protein and a template protein of length n in our own library is transformed into an input vector of length n +2. Structural alignments are assumed to be the best alignment, and SVR is trained to predict the shift score between structural alignment and profile-profile alignment of a query protein to a template protein. The performance is assessed by Pearson correlation coefficient. The trained SVR predicts shift score with the correlation between observed and predicted shift score of 0.80.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.1 s.343
• /
• pp.121-126
• /
• 2006

This paper report on the design and implementation of a 9.6kbps DSSS CDMA modem for a medium speed wireless link. The proposed modem provides a general purpose I/O interface with a microprocessor. The I/O interface consists of 8-bit data bus, chip enable, read/write, and interrupt pins. In transmit block, the 8-bit data delivered from the I/O interface buffer is converted to 9.6kbps serial data, which are spreaded into 76.8kcps with 8-bit PN code generated inside the modem by direct sequence method. An 8-bit training sequence is preceded in the data frame for data synchronization in receiver. In receiver block the PN code is synchronized from the received data spreaded to 76.8kcps and find the data timing from the 8-bit training sequence. We have used the Early-and-Late integration method. The modem has been implemented and verified using a Xilix FPGA board and has been fabricated as an ASIC CHIP through Hynir $0.25{\mu}m$ CMOS. The multiple accessing method is DSSS CDMA.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.59 no.1
• /
• pp.1-9
• /
• 2010

This paper summarizes an education and training system for the auto-reclosing of power transmission system using a real time digital simulator. The system is developed to understand the principle of reclosing and the sequence of automatic reclosing schemes, and practice the effects of reclosing actions to power system in real-time simulator. This study is concentrated into the following two parts. One is the development of real time education and training system of automatic reclosing schemes. For this, we use the RTDS(real time digital simulator) and the actual digital protective relay. The mathematical relay model of RTDS and the actual distance relay which is equipped automatic reclosing function are also used. The other is the user friendly interface between trainee and trainer. The various interface displays are used for user handing and result display. The conditions of automatic reclosing which is a number of reclosing, reclosing dead time, reset time, and so on, can be changed by the user interface panel. A number of scenario cases are reserved for the education and training. Through the test, we verified that the proposed system can be effectively used to accomplish the education and training of automatic reclosing.