• 방송공학회논문지
• /
• 제26권7호
• /
• pp.844-854
• /
• 2021

Semi-supervised learning (SSL) and few-shot learning (FSL) have shown impressive performance even then the volume of labeled data is very limited. However, SSL and FSL can encounter a significant performance degradation if the diversity gap between the labeled and unlabeled data is high. To reduce this diversity gap, we propose a novel scheme that relies on an autoencoder for generating pseudo examples. Specifically, the autoencoder is trained on a specific class using the available labeled data and the decoder of the trained autoencoder is then used to generate N samples of that specific class based on N random noise, sampled from a standard normal distribution. The above process is repeated for all the classes. Consequently, the generated data reduces the diversity gap and enhances the model performance. Extensive experiments on MNIST and FashionMNIST datasets for SSL and FSL verify the effectiveness of the proposed approach in terms of classification accuracy and robustness against adversarial attacks.

• 공업화학
• /
• 제26권2호
• /
• pp.210-216
• /
• 2015

석탄계 입상 활성탄을 사용하여 수용액으로부터 brilliant blue FCF 염료의 흡착에 대해 조사하였다. 회분식 실험은 흡착제의 양, 초기농도와 접촉시간과 온도를 흡착변수로 사용하여 수행하였다. 흡착평형자료는 Langmuir, Freundlich 및 Temkin 식을 사용하여 해석한 결과 Freundlich 식이 가장 좋은 일치도를 나타냈다. 평가된 Freundlich 상수(1/n = 0.129~0.212)로부터 이 흡착공정이 적절한 처리방법이 될 수 있음을 알았다. 흡착속도실험자료를 유사일차 및 유사이차 반응속도식에 적용해 본 결과는 유사이차반응속도식에 잘 맞는 것으로 나타났다. 음수값의 Gibbs 자유에너지(-4.81~-10.33 kJ/mol)와 양수값의 엔탈피(+78.59 kJ/mol)는 흡착이 자발적이고 흡열공정으로 진행된다는 것을 나타냈다.

• Membrane and Water Treatment
• /
• 제8권3호
• /
• pp.259-277
• /
• 2017

Adsorption is a widely used technique for the removal of dyes from wastewaters by variety of adsorbents. In this work, the main focus is on the potential assessment of anion exchange membrane for the removal of different dyes using batch system and investigation of experimental data by applying various kinetic and thermodynamic models. The removal of anionic dyes i.e., Eosin-B, Eriochrome Black-T and Congo Red by anion exchange membrane BII from aqueous solution was carried out and effect of various parameters such as contact time, membrane dosage, temperature and ionic strength on the percentage removal of anionic dyes was studied. The experimental data was assessed by kinetic models namely pseudo-first-order, pseudo-second-order, Elovich liquid film diffusion, Bangham and the modified Freundlich models equation have been used to analyze the experimental data. These results indicate that the adsorption of these anionic dyes on BII follows pseudo-second-order kinetics with maximum values of regression coefficient (0.992-0.998) for all the systems. The adsorption of dyes was more suitable to be controlled by a liquid film diffusion mechanism. The adsorptive removal of dye Eosin-B and Eriochrome Black-T were decreased with temperature and thermodynamic parameters such as free energy (${\Delta}G^o$), enthalpy (${\Delta}H^o$) and entropy (${\Delta}S^o$) for adsorption of dyes on membrane BII were calculated at 298 K, 308 K and 318 K. The values of enthalpy and entropy were negative for EB and EBT representing that the adsorption of these dyes on BII is physiosorptive and exothermic in nature. Whereas the positive values of enthalpy and entropy for CR adsorption on BII, indicating that its adsorption is endothermic and spontaneous in nature. It is evident from this study that anion exchange membrane has shown good potential for the removal of dyes from aqueous solution and it can be used as adsorbent for dues removal on commercial levels.

• Korean Chemical Engineering Research
• /
• 제53권3호
• /
• pp.309-314
• /
• 2015

야자계 활성탄을 사용하여 수용액으로부터 brilliant blue FCF 염료의 흡착에 대해 조사하였다. 회분식 실험은 흡착제의 양, 초기농도와 접촉시간과 온도를 흡착변수로 사용하여 수행하였다. 흡착평형자료는 Langmuir와 Freundlich 식을 사용하여 해석하였으며, Freundlich 식이 더 좋은 일치도를 나타냈다. 평가된 Freundlich 상수(1/n=0.129~0.212)로부터 활성탄에 의한 brilliant blue FCF의 흡착조작이 적절한 처리방법이 될 수 있음을 알았다. 흡착속도실험자료를 유사일차반응속도식과 유사이차반응속도식에 적용해 본 결과, 흡착동력학은 유사이차반응속도식에 잘 맞는 것으로 나타났다. 음수값의 Gibbs 자유에너지(-4.81~-10.33 kJ/mol)와 양수값의 엔탈피(+78.59 kJ/mol)는 흡착이 자발적이고 흡열공정으로 진행된다는 것을 나타낸다.

• Korean Chemical Engineering Research
• /
• 제54권2호
• /
• pp.255-261
• /
• 2016

활성탄을 사용한 Acid Yellow 14 염료의 흡착 실험은 흡착제의 양, pH, 초기농도, 접촉시간과 온도를 흡착변수로 사용하여 수행하였다. 흡착평형자료는 Langmuir, Freundlich 및 Temkin 등온식을 사용하여 해석하였는데, Freundlich 식이 가장 좋은 일치도를 나타냈다. 평가된 Freundlich 상수(1/n=0.129~0.212)와 Langmuir 분리계수($R_L=0.202{\sim}0.243$)로부터 활성탄에 의한 Acid Yellow 14의 흡착조작은 적절한 처리방법이 될 수 있음을 알았다. Temkin의 흡착열관련상수(B)는 5.101~9.164 J/mol로 평가되어, 흡착공정이 물리흡착임을 알았다. 흡착속도실험자료를 유사일차반응속도식과 유사이차반응속도식에 적용해 본 결과, 흡착동력학은 유사이차반응속도식에 잘 맞는 것으로 나타났다. Gibbs 자유에너지(-4.81~-10.33 kJ/mol)와 엔탈피(+78.59 kJ/mol)는 흡착이 자발적이고 흡열공정으로 진행된다는 것을 나타낸다.

• 한국측량학회지
• /
• 제21권4호
• /
• pp.301-308
• /
• 2003

본 논문은 IKONOS 입체영상과 RPC(Rational Polynomial Coefficients) 모델을 이용하여 3차원 지형자료를 추출할 경우, 추출되는 3차원 지형자료의 정확도를 향상시키기 위하여 RPC모델의 위치정확도를 보정하는 연구를 수행하였다. 이를 위하여 지상기준점을 활용하여 RPC 모델의 보정을 시도할 경우, 지상기준점의 지역적인 분포 및 사용되는 개수가 보정되는 RPC 모델의 위치정확도 향상에 어떠한 영향을 미치는지를 검증하였다. 실험결과 사용되는 지상기준점의 개수보다는 지상기준점의 분포상태가 보정되는 RPC 모델의 위치정확도에 미치는 영향이 더 크다는 것을 확인할 수 있었다. 또한, 분포상태가 고른 지상기준점을 사용하여 본 연구에서 사용한 알고리즘을 적용할 경우에 안정적으로 위치정확도가 향상된 RPC 모델을 획득할 수 있음을 알 수 있었다. 수행된 연구결과를 토대로 지상기준점의 분포가 좋지 않거나 사용 가능한 지상기준점의 개수가 부족한 경우, 이를 극복하기 위해 의사지상기준점을 활용하는 알고리즘에 관한 연구도 수행하였다. 실험 결과에 따르면 지역적으로 좋지 않은 분포를 보이는 지상기준점들을 활용한 경우에도 의사지상기준점을 활용하면 원래의 RPC 모델보다 위치정확도가 향상된 RPC 모델을 얻을 수 있었다. 그리고 적은 수의 의사지상기준점을 활용할수록 즉, 지상기준점에 대한 가중치를 더 높일수록 정확도가 높은 RPC 모델을 획득할 수 있었다. 마지막으로, 본 연구에서 개발된 알고리즘들을 적용하여 RPC 모델의 위치정확도를 보정하고 이를 이용하여 3차원 지형좌표를 추출하였다. 정확도평가 결과 원래의 RPC 모델을 사용하여 추출된 3차원 지형좌표보다 정확도가 향상된 3차원 지형좌표의 취득이 가능하였다. 이는 본 연구에서 개발한 알고리즘들의 효용성을 입증하는 결과라 사료된다.

• Carbon letters
• /
• 제14권1호
• /
• pp.27-33
• /
• 2013

An increase in population initiating rapid industrialization was found to consequently increase the effluents and domestic wastewater into the aquatic ecosystem. In this research the potentialities of Sambucus nigra L. (SNL) plant in the remediation of water, contaminated with methylene blue (MB), a basic dye were investigated. SNL was chemically impregnated with $KHCO_3$. Operating variables studied were pH, amount of adsorbent and contact time. In general, pH did not have any significant effect on colour removal and the highest adsorption capacity was obtained in 0.035 g MB/g-activated carbon. The Langmuir, Freundlich, Temkin and Dubinin-Radushkevich adsorption models were applied to describe the equilibrium isotherms. The adsorption isotherm data were fitted to the Temkin isotherm. The mass transfer property of the sorption process was studied using Lagergren pseudo-first-order and chemisorption pseudo-second-order kinetic models. The sorption process obeyed the pseudo-second-order kinetic model. The surface area, pores volume and diameter were assessed by the Brunauer-Emmett-Teller and Barrett-Joyner-Halenda methods. The results were compared to those from activated carbon (Merck) and an actual sample. The results indicate that SNL can be employed as a natural and eco-friendly adsorbent material for the removal of dye MB from aqueous solutions.

• 한국환경과학회지
• /
• 제23권9호
• /
• pp.1609-1618
• /
• 2014

In this study, as a fundamental study for the remediation of the radionuclides-contaminated soil, the adsorption of cobalt, strontium, and cesium on natural soil and kaolin were experimently investigated and adsorption characteristics were evaluated by using several adsorption kinetic and isotherm models. The pseudo-first-order kinetic model (PFOM), pseudo-second-order kinetic model (PSOM), one-site mass transfer model (OSMTM), and two compartment first-order kinetic model (TCFOKM) were used to evaluate the kinetic data and the pseudo-second-order kinetic model was the best with good correlation. The adsorption equilibria of cobalt, strontium, and cesium on natural soil were fitted successfully by Redlich-Peterson and Sips models. For kaolin, the adsorption equilibria of cobalt, strontium, and cesium were fitted well by Redlich-Peterson, Freundlich, and Sips models, respectively. The amount of adsorbed radionuclides on natural soil and kaolin was in the order of cesium > strontium > cobalt. It is considered that these results could be useful to predicting the adsorption behaviors of radionuclides such as cobalt, strontium, and cesium in soil environments.

• 한국멀티미디어학회논문지
• /
• 제16권12호
• /
• pp.1465-1474
• /
• 2013

In this paper, air pollutant tracing models using perceptron neural network(PNN) and non-negative least square(NNLS) are proposed. When the measured values of the air pollution and the contribution concentration of each source by chemical transport modeling are given, they estimate and trace the amount of the air pollutants emission from each source. Two kinds of emissions data are used in the experiments : CH4 and N2O of Geumgo-dong landfill greenhouse gas, and PM10 of 17 areas in Northeast Asia and eight regions of the Korean Peninsula. Emission values were calculated using pseudo inverse method, PNN and NNLS. Pseudo inverse method could be used for the model, but it may have negative emission values. In order to deal with the problem, we used the PNN and NNLS methods. As a result, the estimation using the NNLS is closer to the measured values than that using PNN. The proposed tracing models have better utilization and generalization than those of conventional pseudo inverse model. It could be used more efficiently for air quality management and air pollution reduction.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제17권1호
• /
• pp.147-155
• /
• 2017

A new endurance test-pattern generation on NAND-flash memory is proposed to improve test cost. We mainly focus on the correlation between the data-pattern and the device error-rate during endurance testing. The novelty is the development of testing method using quasi-random pattern based on device architectures in order to increase the test efficiency during time-consuming endurance testing. It has been proven by the experiments using the commercial 32 nm NAND flash-memory. Using the proposed method, the error-rate increases up to 18.6% compared to that of the conventional method which uses pseudo-random pattern. Endurance testing time using the proposed quasi-random pattern is faster than that of using the conventional pseudo-random pattern since it is possible to reach the target error rate quickly using the proposed one. Accordingly, the proposed method provides more low-cost testing solutions compared to the previous pseudo-random testing patterns.