• Korean Chemical Engineering Research
• /
• v.57 no.2
• /
• pp.274-282
• /
• 2019

In recent years, big data analysis has been expanded to include automatic control through reinforcement learning as well as prediction through modeling. Research on the utilization of image data is actively carried out in various industrial fields such as chemical, manufacturing, agriculture, and bio-industry. In this paper, we applied NASNet, which is an AutoML reinforced learning algorithm, to DeepU-Net neural network that modified U-Net to improve image semantic segmentation performance. We used BRATS2015 MRI data for performance verification. Simulation results show that DeepU-Net has more performance than the U-Net neural network. In order to improve the image segmentation performance, remove dropouts that are typically applied to neural networks, when the number of kernels and filters obtained through reinforcement learning in DeepU-Net was selected as a hyperparameter of neural network. The results show that the training accuracy is 0.5% and the verification accuracy is 0.3% better than DeepU-Net. The results of this study can be applied to various fields such as MRI brain imaging diagnosis, thermal imaging camera abnormality diagnosis, Nondestructive inspection diagnosis, chemical leakage monitoring, and monitoring forest fire through CCTV.

• Journal of Broadcast Engineering
• /
• v.22 no.6
• /
• pp.693-701
• /
• 2017

In this paper, we propose a new method for music genre classification using spikegram and deep neural network. The human auditory system encodes the input sound in the time and frequency domain in order to maximize the amount of sound information delivered to the brain using minimum energy and resource. Spikegram is a method of analyzing waveform based on the encoding function of auditory system. In the proposed method, we analyze the signal using spikegram and extract a feature vector composed of key information for the genre classification, which is to be used as the input to the neural network. We measure the performance of music genre classification using the GTZAN dataset consisting of 10 music genres, and confirm that the proposed method provides good performance using a low-dimensional feature vector, compared to the current state-of-the-art methods.

• Journal of the Korean Society for Precision Engineering
• /
• v.30 no.5
• /
• pp.559-563
• /
• 2013

This study aims to quantify the effects of medication (Med) and deep brain stimulation (DBS) on resting rigidity in patients with Parkinson's disease. We tested 10 limbs of five patients under each of four treatment conditions: 1) baseline, 2) DBS, 3) Med, 4) DBS + Med. Rigidity at the wrist joint was assessed using the Unified Parkinson's Disease Rating Scale (UPDRS). The examiner randomly imposed flexion and extension movement on patient's wrist joint. Resistance to passive movement was quantified by viscoelastic properties. Not only rigidity score but also damping constant showed improvements in rigidity by DBS and Med treatments (p<0.05). This indicates that the viscosity can represent the change in rigidity due to DBS as well as Med, which was manifested by UPDRS score.

• Journal of Korean Society of Steel Construction
• /
• v.16 no.4 s.71
• /
• pp.443-452
• /
• 2004

Recently, the Artificial Neural Network(ANN) which can organize complex non-linear problems by effectively applying the parallel computational model that is similar to the human brain, was adopted in the wide department of technology and resulted in many successful applications. In this study, a more appropriate formal method is suggested for the preliminary structural design stage controlled merely by the designer's experience and intuition. To do so, this study proposes a multi-level ANN according to the general progressive structural design procedure, using Back-Propagation Algorithm (BP) and Genetic Algorithm (GA) for the ANN learning. The preliminary structural design of cable-stayed bridges was applied to illustrate the applicability of the study formulated as stated above, and the results of two different learning methods were compared.

• Journal of Sericultural and Entomological Science
• /
• v.51 no.2
• /
• pp.197-200
• /
• 2013

Silk fibroin is a natural biomaterial that has the biocompatibility and other many advantages. But as a silk fibroin membrane thickness increases, the transparency becomes more opaque. Because the transparency of membranes tissue such as the cornea and dura mater are necessary, transparent membrane is required to replace these transparent membranes. In this study, we fabricated blending silk fibroin membranes that made by mixing the various inorganic salts or polymer in an aqueous solution of silk fibroin. The transparency of the membranes were analyzed. the transparency of these membranes is very different, depending on the mixed materials. Inorganic salts mixed silk membrane was more transparent than the polymer mixed one. Especially, the silk fibroin membrane with calcium chloride was very transparent. We showed the possibility of blending silk fibroin membrane, which can be used in perfect transparent membrane such as the cornea. In the future, we expect that the transparent blending silk fibroin membrane can be used in various medical applications.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.9
• /
• pp.179-185
• /
• 2013

In this paper, the RTA-ERP(Response Time Aligned-Evoked Relative Potential) was modelled to analyze the effect of motor activation pattern in response to visual sensory stimuli. Simon effect was analysed using the amplitude response of RTA-ERP and measured response time. The 'odd number' experiments, which identify an odd number mixed with same numbers, was performed with 15 healthy adult participants(9 males and 6 females, whose mean age of 31) for 7 minutes for each participant. Throughout experimentation, we observed that the proposed RTA-ERP can compensate the timing variation due to different neural processing procedures in the brain, and shows enhanced LRP(Lateralized Readiness Potential) and Pe(Error Related Positivity). Regarding to 'congruence' and 'incongruence' testing patterns, the amplitude of RTA-ERP and the response time for the 'congruence' are $0.03{\mu}V$ larger, and 43 ms faster than those for the 'incongruence', respectively. The amplitude characteristics of RTA-ERP, obtained by synchronizing the onset times with respect to response time, corresponds more likely to that of P300 in the ERP pattern (the characteristics of the Simon effect).

• Journal of Broadcast Engineering
• /
• v.5 no.1
• /
• pp.10-22
• /
• 2000

In this paper, we present a texture description method as a standardization of multimedia contents description. Like color, shape, object and camera motion information, texture is one of very important information in the visual part of international standard (MPEG-7) in multimedia contents description. Current MPEG-7 texture descriptor has been designed to fit human visual system. Many psychophysical experiments give evidence that the brain decomposes the spectra into perceptual channels that are bands in spatial frequency. The MPEG-7 texture description method has employed Radon transform that fits with HVS behavior. By taking average energy and energy deviation of HVS channels, the texture descriptor is generated. To test the performance of current texture descriptor, experiments with MPEG-7 Texture data sets of T1 to T7 are performed. Results show that the current MPEG-7 texture descriptor gives better retrieval rate and fast and fast extraction time for texture feature.

• The Magazine of the IEIE
• /
• v.28 no.2
• /
• pp.90-104
• /
• 2001

최근까지 동물 또는 사람이 극저주파 전자기장에 평생 또는 여러 세대에 걸쳐 노출되었을 경우, 나타나는 생체영향에 관한 연구는 거의 없다. 본 연구에서는 마우스에 60Hz 전자파를 1세대부터 3세대까지 지속적으로 노출시켜 나타나는 영향을 실험하였다. 실험동물은 5주령인 BALB/c 마우스를 1주일간 적응시킨 후 사용하였다. 실험군은 5kV/m, 30kV/m, 0.5mT 그리고 1.5mT의 4개군으로 나누었으며, 대조군은 1군으로 실험하였다. 생후 6주부터는 위에서 정해진 양의 전자파를 20-22주간 지속적으로 실험동물에 조사하고 동일조건의 암수 마우스를 교미시켰으며, 임신 후에도 사망 또는 부검시까지 동일한 조건으로 계속 조사하였다. 2세대와 3세대는 임신적부터 사망 또는 부검시까지 동일한 조건으로 계속 조사하였다. 1, 2 그리고 3세대 마우스들은 질병에 의한 사망 직전 또는 생후 46주, 66주 그리고 생후 49주에 부검한 뒤, 혈액학적 및 생화학적 검사 그리고 조직병리학적 검사를 실시하였다. 2세대 태아에서는 조기사망(early fetal death), 성장기사망(late fetal death) 그리고 뇌노출(excencephaly) 및 선천성 심장기형을 포함하는 선천이상이 발견되었는데, 이는 대조군에 비해 2-4배 높았다. 1, 2세대에서는 생식기인 고환(testis)과 난소(ovary)의 무게가 감소하였으나 2세대에서는 아무런 변화를 보이지 않았다. 실험군인 30kV/m, 0.5mT 그리고 1.5mT 전 실험군인 30kV/m, 0.5mT 그리고 1.5mT 전자파에 노출된 1세대와 2세대 마우스에서는 프종(lymphoma), 선암종(adenocarcinoma), 기저상피세포증(basal cell epithelioma), 편평상 피두유종(squamous papilloma) 그리고 선종(adenoma) 등이 발견되었으나, 3세대에서는 발견되지 않았다. 60Hz 전자파는 태아 및 생식기에 영향을 미치고, 또한 종양을 유발할 가능성이 있다. 그러나 3세대는 전자파 환경에 점차 적응을 하는 것으로 보인다. 그러나 몇몇 국제기구에서 정하여 놓은 안전한계치의 전자파가 생체에 장기간 노출되었을 경우에 나타날 수 있는 생체영향을 확인하기 위해서는 많은 연구가 필요하다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.12
• /
• pp.197-203
• /
• 2014

When the lightning current is injected to the ground system of information and communication facilities, analysis of the transient potential rise in the ground system is one of main factors to effectively design the ground system. The performance of grounding systems is normally estimated with the grounding impedance and the transient potential rise which represents the electrical characteristics of the grounding system. The method for calculating the grounding impedance depending on the injection point of the lightning current was proposed. The delta-gap source model was proposed to calculate the grounding impedance in the case that the lightning current is injected to the center of the horizontal ground electrode. A new program which is possible to apply the frequency-dependent soil parameters using the Debye model was developed, because a commercial program for analyzing the performance of the grounding system can not apply to the frequency-dependent soil parameters. The experiment was carried out to confirm the availability of the simulation results with the same condition. Finally, the transient potential rises of a horizontal ground electrode depending on the lightning current waveforms were analyzed by using the results of the grounding impedance which is associated with the frequency-dependent soil parameters.

• Journal of Biomedical Engineering Research
• /
• v.42 no.6
• /
• pp.268-276
• /
• 2021

Functional near-infrared spectroscopy-based brain-computer interface (fNIRS-based BCI) has been receiving much attention. However, we are practically constrained to obtain a lot of fNIRS data by inherent hemodynamic delay. For this reason, when employing machine learning techniques, a problem due to the high-dimensional feature vector may be encountered, such as deteriorated classification accuracy. In this study, we employ an elastic net-based feature selection which is one of the embedded methods and demonstrate the utility of which by analyzing the results. Using the fNIRS dataset obtained from 18 participants for classifying brain activation induced by mental arithmetic and idle state, we calculated classification accuracies after performing feature selection while changing the parameter α (weight of lasso vs. ridge regularization). Grand averages of classification accuracy are 80.0 ± 9.4%, 79.3 ± 9.6%, 79.0 ± 9.2%, 79.7 ± 10.1%, 77.6 ± 10.3%, 79.2 ± 8.9%, and 80.0 ± 7.8% for the various values of α = 0.001, 0.005, 0.01, 0.05, 0.1, 0.2, and 0.5, respectively, and are not statistically different from the grand average of classification accuracy estimated with all features (80.1 ± 9.5%). As a result, no difference in classification accuracy is revealed for all considered parameter α values. Especially for α = 0.5, we are able to achieve the statistically same level of classification accuracy with even 16.4% features of the total features. Since elastic net-based feature selection can be easily applied to other cases without complicated initialization and parameter fine-tuning, we can be looking forward to seeing that the elastic-based feature selection can be actively applied to fNIRS data.