• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.8
• /
• pp.29-37
• /
• 2019

Textile quality management targets the physical properties of fabrics and the subjective discriminations of color and fitting. Color is the most representative quality factor that consumers can use to evaluate quality levels without any instruments. For this reason, quantification using a color discrimination device has been used for statistical quality management in the textile industry. However, small and medium-sized domestic textile manufacturers use only visual inspection for color discrimination. As a result, color discrimination is different based on the inspectors' individual tendencies and work procedures. In this research, we want to develop a textile industry-friendly quality management method, evaluating the possibility of rapid color discrimination using a digital imaging device, which is one of the office-automation instruments. The results show that an imaging process-based color discrimination method is highly correlated with conventional color discrimination instruments ($R^2=0.969$), and is also applicable to field discrimination of the manufacturing process, or for different lots. Moreover, it is possible to recognize quality management factors by analyzing color components, ${\Delta}L$, ${\Delta}a$, ${\Delta}b$. We hope that our rapid discrimination method will be a substitute technique for conventional color discrimination instruments via elaboration and optimization.

• Journal of the Korean Electrochemical Society
• /
• v.22 no.1
• /
• pp.36-42
• /
• 2019

A macroporous Sn thick film as a high capacity negative electrode for a lithium ion secondary battery was prepared by using a chemical etching method using nitric acid for a Sn film having a thickness of $52{\mu}m$. The porous Sn thick film greatly reduced the over-voltage for the alloying reaction with lithium by the increased reaction area. At the same time. The porous structure of active Sn film plays a part in the buffer and reduces the damage by the volume change during cycles. Since the porous Sn thick film electrode does not require the use of the binder and the conductive carbon black, it has substantially larger energy density. As the concentration of nitric acid in etching solution increased, the degree of the etching increased. The etching of the Sn film effectively proceeded with nitric acid of 3 M concentration or more. The porous Sn film could not be recovered because the most of Sn was eluted within 60 seconds by the rapid etching rate in the 5 M nitric acid. In the case of etching with 4 M nitric acid for 60 seconds, the appropriate porous Sn film was formed with 48.9% of weight loss and 40.3% of thickness change during chemical acid etching process. As the degree of etching of Sn film increased, the electrochemical activity and the reversible capacity for the lithium storage of the Sn film electrode were increased. The highest reversible specific capacity of 650 mAh/g was achieved at the etching condition with 4 M nitric acid. The porous Sn film electrode showed better cycle performance than the conventional electrode using a Sn powder.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.1
• /
• pp.201-210
• /
• 2019

Recently, due to the efforts to mitigate traffic congestion and expansion of space efficiency, the construction of underground roads has been increased in big-scale cities. Since tunnels in the city have a higher chance for a fire leading to a great tragedy during a severe traffic jam than mountain tunnels, it is highly likely that it will be constructed as a tunnel, having a small cross section, for small vehicles. However, if they are constructed as such small-vehicle tunnels, it would be possible to reduce the design fire intensity while the concentration of harmful gases would increase due to a reduction in the small cross sectional area, led by a decrease in the tunnel height. In this study, behaviors of fire smoke by the installation interval and format of large-scale exhaust-gas ports were examined and compared in the analysis of temperatures and CO concentrations of a tunnel and its results were as the following. Although there were no significant differences in the smoke spreading distance between installation intervals, but in this study, 100 m was found to be the most effective installation interval. The smoke exhaustion performance was found to be excellent in the order of $4m{\times}3m$, $6m{\times}2m$, and $3m{\times}2m$ (2 lane) of the smoke spreading distance. Although there was no significant difference in the smoke spreading distance between formats of large-scale exhaust-gas ports, it was found that the smoke spreading distance was larger than other cases when it was $3m{\times}2m$ in the fire growing process. The analysis of smoke spreading distances by the aspect ratio showed that a smoke spreading distance was shorted when its the smoke spreading distance was found to be shorter when its traverse distance was relatively longer than its longitudinal distance.

• Korean Chemical Engineering Research
• /
• v.59 no.4
• /
• pp.574-583
• /
• 2021

Waste air containing ethanol (100 ppmv) and hydrogen sulfide (10 ppmv) was continuously treated by waste air-treating system composed of two annular photocatalytic reactors (effective volume: 1.5 L) packed with porous SiO₂ media carrying TiO₂-anatase photocatalyst, one of which was alternately operated for 32 d/run while the other was regenerated by 100 ℃ hot air with 15 W UV(-A)-light on. As its elimination-behavior of ethanol, the removal efficiencies of ethanol at 1^st, 2^nd and 3^rd operation of the photocatalytic reactor system(A), turned out to be ca. 60, 55 and 54%, respectively, at their steady state condition. Unlike the elimination-behavior of ethanol, its hydrogen sulfide-elimination behavior showed repeated decrease of hydrogen sulfide removal efficiency by its resultant arrival at a lower level of steady state condition. Nevertheless, the removal efficiencies of hydrogen sulfide at 1^st, 2^nd and 3^rd operation of the photocatalytic reactor system, turned out to be ca. 80, 75 and 73%, respectively, at their final steady state condition, higher by ca. 20, 20 and 19% than those of ethanol, respectively. Therefore, assuming that adsorption on porous SiO₂-photocatalyst carrier was regarded to belong to a reversible deactivation and that decreased % of removal efficiency due to the reversible deactivation of photocatalyst including the adsorption was independent of the number of its use upon regeneration, the increments of the decreased % of removal efficiency of ethanol and hydrogen sulfide, due to an irreversible deactivation of photocatalyst, for the 3^rd use of regenerated photocatalyst, compared with the 2^nd use of regenerated photocatalyst, were ca. 1 and 2%, respectively, which was insignificant or the less than those of ca. 5 and 5%, respectively, for the 2^nd use of regenerated photocatalyst compared with the 1^st use of virgin photocatalyst. This trend of the photocatalytic reactor system was observed to be similar to that of the other alternately-operating photocatalytic reactor system.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Microbiology and Biotechnology Letters
• /
• v.49 no.1
• /
• pp.111-119
• /
• 2021

A bacterial strain isolated from a Malva verticillata leaf was identified as Bacillus velezensis MV2 based on the 16S rRNA sequencing results. Complete genome sequencing revealed that B. velezensis MV2 possessed a single 4,191,702-bp contig with 45.57% GC content. Generally, Bacillus spp. are known to produce diverse antimicrobial compounds including bacteriocins, polyketides, and non-ribosomal peptides. Antimicrobial compounds in the B. velezensis MV2 were extracted from culture supernatants using hydrophobic interaction chromatography. The crude extracts showed antimicrobial activity against both gram-positive bacteria and gram-negative bacteria; however, they were more effective against gram-positive bacteria. The extracts also showed antifungal activity against phytopathogenic fungi such as Fusarium fujikuroi and F. graminearum. In time-kill assays, these antimicrobial compounds showed bactericidal activity against Bacillus cereus, used as indicator strain. To predict the type of antimicrobial compounds produced by this strain, we used the antiSMASH algorithm. Forty-seven secondary metabolites were predicted to be synthesized in MV2, and among them, fourteen were identified with a similarity of 80% or more with those previously identified. Based on the antimicrobial properties, the antimicrobial compounds may be nonribosomal peptides or polyketides. These compounds possess the potential to be used as biopesticides in the food and agricultural industry as an alternative to antibiotics.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.66 no.4
• /
• pp.307-317
• /
• 2021

In recent years, global warming has led to frequent climate change-related problems, and elevated temperatures, among adverse climatic factors, represent a critical problem negatively affecting crop growth and yield. In this context, the present study examined the physiological traits of wheat plants grown under high temperatures. Specifically, the effects of elevated temperatures on seed development after heading were evaluated, and the vegetation indices of different organs were assessed using hyperspectral analysis. Among physiological traits, leaf greenness and OJIP parameters were higher in the high-temperature treatment than in the control treatment. Similarly, the leaf photosynthetic rate during seed development was higher in the high-temperature treatment than in the control treatment. Moreover, temperature by organ was higher in the high-temperature treatment than in the control treatment; consequently, the leaf transpiration rate and stomatal conductance were higher in the control treatment than in the high-temperature treatment. On all measuring dates, the weight of spikes and seeds corresponding to the sink organs was greater in the high-temperature treatment than in the control treatment. Additionally, the seed growth rate was higher in the high-temperature treatment than in the control treatment 14 days after heading, which may be attributed to the higher redistribution of photosynthates at the early stage of seed development in the former. In hyperspectral analysis, the vegetation indices related to leaf chlorophyll content and nitrogen state were higher in the high-temperature treatment than in the control treatment after heading. Our results suggest that elevated temperatures after the booting stage positively affect wheat growth and yield.

• Economic and Environmental Geology
• /
• v.54 no.6
• /
• pp.689-698
• /
• 2021

Since the mine reclamation scheme was implemented from 2007 in Korea, various remediation programs have been decontaminated the pollution associated with mining and 254 mines were managed to reclamation from 2011 to 2015. However, as the total amount of contaminated mine drainage has been increased due to the discovery of potential hazards and contaminated zone, more efficient and economical treatment technology is required. Therefore, in this study, the adsorption properties of arsenic was evaluated according to the adsorbents which were derived from water treatment sludge(Alum based adsorbent, ABA-500) and granular ferric hydroxide(GFH), already commercialized. The alum sludge and GFH adsorbents consisted of aluminum, silica materials and amorphous iron hydroxide, respectively. The point of zero charge of ABA-500 and GFH were 5.27 and 6.72, respectively. The result of the analysis of BET revealed that the specific surface area of GFH(257 m²·g^-1) was larger than ABA-500(126~136 m²·g^-1) and all the adsorbents were mesoporous materials inferred from N₂ adsorption-desorption isotherm. The adsorption capacity of adsorbents was compared with the batch experiments that were performed at different reaction times, pH, temperature and initial concentrations of arsenic. As a result of kinetic study, it was confirmed that arsenic was adsorbed rapidly in the order of GFH, ABA-500(granule) and ABA-500(3mm). The adsorption kinetics were fitted to the pseudo-second-order kinetic model for all three adsorbents. The amount of adsorbed arsenic was increased with low pH and high temperature regardless of adsorbents. When the adsorbents reacted at different initial concentrations of arsenic in an hour, ABA-500(granule) and GFH could remove the arsenic below the standard of drinking water if the concentration was below 0.2 mg·g^-1 and 1 mg·g^-1, respectively. The results suggested that the ABA-500(granule), a low-cost adsorbent, had the potential to field application at low contaminated mine drainage.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1723-1735
• /
• 2022

Landslides are one of the most prevalent natural disasters, threating both humans and property. Also landslides can cause damage at the national level, so effective prediction and prevention are essential. Research to produce a landslide susceptibility map with high accuracy is steadily being conducted, and various models have been applied to landslide susceptibility analysis. Pixel-based machine learning models such as frequency ratio models, logistic regression models, ensembles models, and Artificial Neural Networks have been mainly applied. Recent studies have shown that the kernel-based convolutional neural network (CNN) technique is effective and that the spatial characteristics of input data have a significant effect on the accuracy of landslide susceptibility mapping. For this reason, the purpose of this study is to analyze landslide vulnerability using a pixel-based deep neural network model and a patch-based convolutional neural network model. The research area was set up in Gangwon-do, including Inje, Gangneung, and Pyeongchang, where landslides occurred frequently and damaged. Landslide-related factors include slope, curvature, stream power index (SPI), topographic wetness index (TWI), topographic position index (TPI), timber diameter, timber age, lithology, land use, soil depth, soil parent material, lineament density, fault density, normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) were used. Landslide-related factors were built into a spatial database through data preprocessing, and landslide susceptibility map was predicted using deep neural network (DNN) and CNN models. The model and landslide susceptibility map were verified through average precision (AP) and root mean square errors (RMSE), and as a result of the verification, the patch-based CNN model showed 3.4% improved performance compared to the pixel-based DNN model. The results of this study can be used to predict landslides and are expected to serve as a scientific basis for establishing land use policies and landslide management policies.

• Journal of Korea Water Resources Association
• /
• v.54 no.spc1
• /
• pp.1107-1118
• /
• 2021

Climate change brought on by global warming increased the frequency of flood and drought on the Korean Peninsula, along with the casualties and physical damage resulting therefrom. Preparation and response to these water disasters requires national-level planning for water resource management. In addition, watershed-level management of water resources requires flow duration curves (FDC) derived from continuous data based on long-term observations. Traditionally, in water resource studies, physical rainfall-runoff models are widely used to generate duration curves. However, a number of recent studies explored the use of data-based deep learning techniques for runoff prediction. Physical models produce hydraulically and hydrologically reliable results. However, these models require a high level of understanding and may also take longer to operate. On the other hand, data-based deep-learning techniques offer the benefit if less input data requirement and shorter operation time. However, the relationship between input and output data is processed in a black box, making it impossible to consider hydraulic and hydrological characteristics. This study chose one from each category. For the physical model, this study calculated long-term data without missing data using parameter calibration of the Soil Water Assessment Tool (SWAT), a physical model tested for its applicability in Korea and other countries. The data was used as training data for the Long Short-Term Memory (LSTM) data-based deep learning technique. An anlysis of the time-series data fond that, during the calibration period (2017-18), the Nash-Sutcliffe Efficiency (NSE) and the determinanation coefficient for fit comparison were high at 0.04 and 0.03, respectively, indicating that the SWAT results are superior to the LSTM results. In addition, the annual time-series data from the models were sorted in the descending order, and the resulting flow duration curves were compared with the duration curves based on the observed flow, and the NSE for the SWAT and the LSTM models were 0.95 and 0.91, respectively, and the determination coefficients were 0.96 and 0.92, respectively. The findings indicate that both models yield good performance. Even though the LSTM requires improved simulation accuracy in the low flow sections, the LSTM appears to be widely applicable to calculating flow duration curves for large basins that require longer time for model development and operation due to vast data input, and non-measured basins with insufficient input data.