• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.139-156
• /
• 2021

The biggest reason for using a deep learning model in image classification is that it is possible to consider the relationship between each region by extracting each region's features from the overall information of the image. However, the CNN model may not be suitable for emotional image data without the image's regional features. To solve the difficulty of classifying emotion images, many researchers each year propose a CNN-based architecture suitable for emotion images. Studies on the relationship between color and human emotion were also conducted, and results were derived that different emotions are induced according to color. In studies using deep learning, there have been studies that apply color information to image subtraction classification. The case where the image's color information is additionally used than the case where the classification model is trained with only the image improves the accuracy of classifying image emotions. This study proposes two ways to increase the accuracy by incorporating the result value after the model classifies an image's emotion. Both methods improve accuracy by modifying the result value based on statistics using the color of the picture. When performing the test by finding the two-color combinations most distributed for all training data, the two-color combinations most distributed for each test data image were found. The result values were corrected according to the color combination distribution. This method weights the result value obtained after the model classifies an image's emotion by creating an expression based on the log function and the exponential function. Emotion6, classified into six emotions, and Artphoto classified into eight categories were used for the image data. Densenet169, Mnasnet, Resnet101, Resnet152, and Vgg19 architectures were used for the CNN model, and the performance evaluation was compared before and after applying the two-stage learning to the CNN model. Inspired by color psychology, which deals with the relationship between colors and emotions, when creating a model that classifies an image's sentiment, we studied how to improve accuracy by modifying the result values based on color. Sixteen colors were used: red, orange, yellow, green, blue, indigo, purple, turquoise, pink, magenta, brown, gray, silver, gold, white, and black. It has meaning. Using Scikit-learn's Clustering, the seven colors that are primarily distributed in the image are checked. Then, the RGB coordinate values of the colors from the image are compared with the RGB coordinate values of the 16 colors presented in the above data. That is, it was converted to the closest color. Suppose three or more color combinations are selected. In that case, too many color combinations occur, resulting in a problem in which the distribution is scattered, so a situation fewer influences the result value. Therefore, to solve this problem, two-color combinations were found and weighted to the model. Before training, the most distributed color combinations were found for all training data images. The distribution of color combinations for each class was stored in a Python dictionary format to be used during testing. During the test, the two-color combinations that are most distributed for each test data image are found. After that, we checked how the color combinations were distributed in the training data and corrected the result. We devised several equations to weight the result value from the model based on the extracted color as described above. The data set was randomly divided by 80:20, and the model was verified using 20% of the data as a test set. After splitting the remaining 80% of the data into five divisions to perform 5-fold cross-validation, the model was trained five times using different verification datasets. Finally, the performance was checked using the test dataset that was previously separated. Adam was used as the activation function, and the learning rate was set to 0.01. The training was performed as much as 20 epochs, and if the validation loss value did not decrease during five epochs of learning, the experiment was stopped. Early tapping was set to load the model with the best validation loss value. The classification accuracy was better when the extracted information using color properties was used together than the case using only the CNN architecture.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.1-22
• /
• 2021

Aspect Based Sentiment Analysis (ABSA), which analyzes sentiment based on aspects that appear in the text, is drawing attention because it can be used in various business industries. ABSA is a study that analyzes sentiment by aspects for multiple aspects that a text has. It is being studied in various forms depending on the purpose, such as analyzing all targets or just aspects and sentiments. Here, the aspect refers to the property of a target, and the target refers to the text that causes the sentiment. For example, for restaurant reviews, you could set the aspect into food taste, food price, quality of service, mood of the restaurant, etc. Also, if there is a review that says, "The pasta was delicious, but the salad was not," the words "steak" and "salad," which are directly mentioned in the sentence, become the "target." So far, in ABSA, most studies have analyzed sentiment only based on aspects or targets. However, even with the same aspects or targets, sentiment analysis may be inaccurate. Instances would be when aspects or sentiment are divided or when sentiment exists without a target. For example, sentences like, "Pizza and the salad were good, but the steak was disappointing." Although the aspect of this sentence is limited to "food," conflicting sentiments coexist. In addition, in the case of sentences such as "Shrimp was delicious, but the price was extravagant," although the target here is "shrimp," there are opposite sentiments coexisting that are dependent on the aspect. Finally, in sentences like "The food arrived too late and is cold now." there is no target (NULL), but it transmits a negative sentiment toward the aspect "service." Like this, failure to consider both aspects and targets - when sentiment or aspect is divided or when sentiment exists without a target - creates a dual dependency problem. To address this problem, this research analyzes sentiment by considering both aspects and targets (Target-Aspect-Sentiment Detection, hereby TASD). This study detected the limitations of existing research in the field of TASD: local contexts are not fully captured, and the number of epochs and batch size dramatically lowers the F1-score. The current model excels in spotting overall context and relations between each word. However, it struggles with phrases in the local context and is relatively slow when learning. Therefore, this study tries to improve the model's performance. To achieve the objective of this research, we additionally used auxiliary loss in aspect-sentiment classification by constructing CNN(Convolutional Neural Network) layers parallel to existing models. If existing models have analyzed aspect-sentiment through BERT encoding, Pooler, and Linear layers, this research added CNN layer-adaptive average pooling to existing models, and learning was progressed by adding additional loss values for aspect-sentiment to existing loss. In other words, when learning, the auxiliary loss, computed through CNN layers, allowed the local context to be captured more fitted. After learning, the model is designed to do aspect-sentiment analysis through the existing method. To evaluate the performance of this model, two datasets, SemEval-2015 task 12 and SemEval-2016 task 5, were used and the f1-score increased compared to the existing models. When the batch was 8 and epoch was 5, the difference was largest between the F1-score of existing models and this study with 29 and 45, respectively. Even when batch and epoch were adjusted, the F1-scores were higher than the existing models. It can be said that even when the batch and epoch numbers were small, they can be learned effectively compared to the existing models. Therefore, it can be useful in situations where resources are limited. Through this study, aspect-based sentiments can be more accurately analyzed. Through various uses in business, such as development or establishing marketing strategies, both consumers and sellers will be able to make efficient decisions. In addition, it is believed that the model can be fully learned and utilized by small businesses, those that do not have much data, given that they use a pre-training model and recorded a relatively high F1-score even with limited resources.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.6
• /
• pp.469-477
• /
• 2008

Purpose: The goal of this paper is to present the design and performance of a position encoding circuit for $16{\times}16$ array of position sensitive multi-anode photomultiplier tube for small animal PET scanners. This circuit which reduces the number of readout channels from 256 to 4 channels is based on a charge division method utilizing a resistor array. Materials and Methods: The position encoding circuit was simulated with PSpice before fabrication. The position encoding circuit reads out the signals from H9500 flat panel PMTs (Hamamatsu Photonics K.K., Japan) on which $1.5{\times}1.5{\times}7.0\;mm^3$ $L_{0.9}GSO$ ($Lu_{1.8}Gd_{0.2}SiO_{5}:Ce$) crystals were mounted. For coincidence detection, two different PET modules were used. One PET module consisted of a $29{\times}29\;L_{0.9}GSO$ crystal layer, and the other PET module two $28{\times}28$ and $29{\times}29\;L_{0.9}GSO$ crystal layers which have relative offsets by half a crystal pitch in x- and y-directions. The crystal mapping algorithm was also developed to identify crystals. Results: Each crystal was clearly visible in flood images. The crystal identification capability was enhanced further by changing the values of resistors near the edge of the resistor array. Energy resolutions of individual crystal were about 11.6%(SD 1.6). The flood images were segmented well with the proposed crystal mapping algorithm. Conclusion: The position encoding circuit resulted in a clear separation of crystals and sufficient energy resolutions with H9500 flat-panel PMT and $L_{0.9}GSO$ crystals. This circuit is good enough for use in small animal PET scanners.

• Journal of Food Hygiene and Safety
• /
• v.35 no.3
• /
• pp.225-233
• /
• 2020

By the standards and specifications for hygiene products, three test methods for formaldehyde are specified for each item type of hygiene product. After derivatization using acetylacetone and 2,4-dinitrophenylhydrazine (2,4-DNPH), formaldehyde is analyzed by spectrophotometer and high-performance liquid chromatography (HPLC). Validation of the three test methods was performed on tissue, diaper lining and waterproof layer, and panty liner products. The results of linearity (R²), limit of detection (LOD), limit of quantification (LOQ), recovery rate (%) and reproducibility (%), showed that all three methods are suitable for analyzing formaldehyde in hygiene products. After derivatization with 2,4-DNPH and cetylacetone, formaldehyde was analyzed at 0, 3, 6, 9, 24 and 48 hours by HPLC. Formaldehyde derivatized with 2,4-DNPH showed no statistically significant change in formaldehyde peak area over time (P>0.05). But, acetylacetone-derivatizated formaldehyde showed a negative correlation coefficient (r) over time (P<0.01). We investigated the residual amounts of formaldehyde in 205 hygiene products distributed in Busan. Among 74 disposable diaper products tested, 73 had low concentrations of formaldehyde (0.13-29.87 mg/kg). Moreover, formaldehyde was not detected in any of 78 tissue, 27 disposable paper towel, 12 disposable dishcloth, 7 paper cup, one brand of paper straw and 6 disposable napkin products.

• Korean Journal of Food Science and Technology
• /
• v.6 no.2
• /
• pp.98-108
• /
• 1974

Fourteen varieties of Korean chestnut were subjected to the test of adaptability for processing and the possibilities of colored products development. The fruit size of Korean chestnut ranged $11{\sim}18g$ and these small fruits could not be expected to be utilized for the processing of Marron glaces which requires larger size as $25{\sim}30g$. As the storge period is extended the discoloration tendency of chestnut flesh was increasingly developed, however, the fresh chestnut has shown layer-separating phenomenon and ragged surface of fruit which delivers disagreeable appearance to the finished product. The principal factors of discoloration occurred during processing were the behavior of tannin and darkening rate shown on flesh differed each other among varieties; the Chukpa and Yuma variety have exhibited the most serious discoloration and the Taab-b variety, the lightest. Taab-b variety in this connection could be expected to be available for Kanroni processing. For the industrialization of chestnut processing the flame-scorched peeling method is advisable. The capacity of this method is proportional to the square of scorching radius and highly flexible in controlling its performance. As for the processing of colored product, the sugar dehydration and coating and the sugar penetration method demand further study in basical views; however, the canned product of chestnut-redbean has shown the possibility of being utilized as a substitute for or paralleled use with the sugar-syruped canned product which so far has been considered as the only item of export to Japan.

• Korean Journal of Poultry Science
• /
• v.36 no.2
• /
• pp.103-110
• /
• 2009

This experiment was conducted using 350 Lohmann Brown layers (67 weeks of age) to evaluate the dietary effect of organic or inorganic Se and Cu on their contents in chicken eggs. The layers were divided into seven groups and fed a commercial diet or one of the six experimental diets containing different levels of Se and Cu (T1, 0.3ppm organic Se; T2, 1.0ppm organic Se; T3, 1.0ppm inorganic Se; T4, 125ppm organic Cu; T5, 250ppm organic Cu; and T6, 250ppm inorganic Cu) for 5 weeks. No significant differences were observed in egg production and its qualities among groups. The contents of blood cholesterol in the birds fed various levels of Se and Cu were not significantly different as compared to control. Se contents in eggs were linearly increased as dietary Se levels increased for both sources, but Se contents from the groups fed organic Se were slightly higher than those fed inorganic Se. Sensory characteristics of eggs were not influenced by dietary treatments. In conclusion, Se contents in eggs were linearly increased as dietary levels of organic or inorganic Se increased, but eggs in layers fed the diet containing organic Se did not show higher Se contents than those in birds fed dietary inorganic Se.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.40 no.10
• /
• pp.1397-1403
• /
• 2011

Epimedium koreanum Nakai is a wild medicinal plant commonly consumed in South Korea due to its health beneficial effects. In the present study, the antioxidative, antimutagenic and immunological activities of E. koreanum Nakai extracts were investigated for their use in food. The yields of icariin compounds from the ethanol extract as well as the ethyl acetate, butanol, hexane, water, and chloroform fractions of E. koreanum were 27.9, 2.5, 1.7, 1.4, and 1.3 ${\mu}g/g$, respectively. The icariin components (295.5 ${\mu}g/g$) were collected from the ethyl acetate fraction by thin layer chromatography (TLC) and analyzed via high performance liquid chromatography (HPLC). The antioxidant activities of each fraction were as follows: ethyl acetate (49.0 ${\mu}g/mL$), butanol (59.2 ${\mu}g/mL$), hexane (119.8 ${\mu}g/mL$), water (122.0 ${\mu}g/mL$), and chloroform (138.5 ${\mu}g/mL$), based on $RC_{50}$ ${\mu}g/mL$. Icariin, isolated and identified as the main component, showed strong antioxidant activity with a $RC_{50}$ value of 15.3 ${\mu}g/mL$, which was higher than those of ascorbic acid (19.5 ${\mu}g/mL$) and ${\alpha}$-tocopherol (18.2 ${\mu}g/mL$). In an Ames test, none of the fractions produced mutagenic effects on Salmonella Typhimurium TA98 and TA100. In an immunomodulating activity test, the effects of E. koreanum Nakai on B cells (Rhamos) and T cells (Jurkat) were investigated. These results show that the growth and viability of B and T cells were increased by isolated icariin components for 1.27 and 1.28 fold, respectively. These results also provide preliminary data for the development of E. koreanum Nakai as an edible food material.

• Geophysics and Geophysical Exploration
• /
• v.14 no.1
• /
• pp.80-87
• /
• 2011

Autonomous underwater vehicles (AUVs) present the important advantage of being able to approach the seafloor more closely than surface vessel surveys can. To collect bathymetric data, bottom material information, and sub-surface images, multibeam echosounder, sidescan sonar (SSS) and subbottom profiler (SBP) equipment mounted on an AUV are powerful tools. The 3000m class AUV URASHIMA was developed by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC). After finishing the engineering development and examination phase of a fuel-cell system used for the vehicle's power supply system, a renovated lithium-ion battery power system was installed in URASHIMA. The AUV was redeployed from its prior engineering tasks to scientific use. Various scientific instruments were loaded on the vehicle, and experimental dives for science-oriented missions conducted from 2006. During the experimental cruise of 2007, high-resolution acoustic images were obtained by SSS and SBP on the URASHIMA around the northern Kumano Basin off Japan's Kii Peninsula. The map of backscatter intensity data revealed many debris objects, and SBP images revealed the subsurface structure around the north-eastern end of our study area. These features suggest a structure related to the formation of the latest submarine fan. However, a strong reflection layer exists below ~20 ms below the seafloor in the south-western area, which we interpret as a denudation feature, now covered with younger surface sediments. We continue to improve the vehicle's performance, and expect that many fruitful results will be obtained using URASHIMA.

• Korean Journal of Poultry Science
• /
• v.15 no.3
• /
• pp.229-246
• /
• 1988

These studies were conducted to investigate the effects of dietary protein and metabolizable energy(ME) levels on layer performance during summer period. Total 480 ISA Brown egg-type layers ageing 49 weeks housed 2 birds per cage with 4 replictes of 20 birds were employed in this study. Mean environmental temperature over experimental period ranged from 22.3$^{\circ}$to 29.5$^{\circ}C$. The treatments consisted of dietary ME levels of 2500, 2700, 2900kcal/kg of diet containing 15% and 17% protein, respectively, to provide $3{\times}2$factorial design. As metabolizable energy level increased form 2500 to 2900 kcal/kg of diet, daily feed and protein intake, egg production. egg weight, egg mass decreased, but the reverse was true for the daily energy intake, energy requirement and feed costs per kg egg. Feed conversion(kg feed/kg egg) and viability were not affected by the dietary energy levels. However, there were no significant difference in egg production, protein requirement per kg egg, and egg weight between those hens fed 2500 kcal ME/kg diets and those fed 2700 kcal ME/kg diets, and no difference was found in egg weight between those fed 2700 kcal ME/kg and those fed 2900 kcal ME/kg, either. In addition, no specific trend was observed in protein requirement per egg by the different level of metabolizable energy in diets. On the other hand, as dietary protein level increased from 15 to 17%, daily protein in-take, egg production, egg weight, egg mass, and protein requirement and feed costs per kg egg increased, but feed and energy requirement per kg egg decreased, and no significant difference in the daily feed and energy intake and viability were observed among dietary protein levels. It was concluded that metabolizable energy level of 2500 kcal/kg of diet and 17% dietary protein level were considered to be adequate to support the optimum productivity of layers during summer period.

• Resources Recycling
• /
• v.15 no.3 s.71
• /
• pp.38-45
• /
• 2006

The thermal filter press dewatering(TFPD) technology to improve the dewaterability through increasing the inner vapor pressure, lowering the filtration viscosity and forming the porosity easily within cake as applying the heat at the sludge layer was developed in this study. The hot water with temperature of $95^{\circ}C$ and pressure of $1.2kg_f/cm^2$ was supplied to the heating plate equipped between filter plates with plate size of $470{\times}470mm$ and material of polypropylene. Sludge was dewaterd by supplying pressure of $5kg_f/cm^2$ and then by squeezing pressure of $15kg_f/cm^2$. As a results of estimating the characteristics of thermal dewatering to consider the initial water content and organic content to be influenced by a period of water shortage and rainwater, the dewatered cake water content was about 35 wt% and dewatering velocity was $4DSkg/m^2{\cdot}hr$ under the rainwater period, and the dewatered cake water content was about 50 wt% and dewatering velocity was $1.5DSkg/m^2{\cdot}hr$ in the case of sludge of water shortage season. These results was superior to the mechanical dewatering performance with water content of 70wt% and dewatering velocity of $0.9DSkg/m^2{\cdot}hr$. On the base of the results of TFPD, energy consumpted to deal with DS(Dry Solid) of 1kg was estimated by 300 kJ. It was analyzed that the energy consumption of TFPD was decreased about one third with comparison to the dryer system. Dewatering velocity of this technology was faster than the one of mechanical dewatering equipment and it was easier to product low water content cake. Therefore, this technology was recognized that dewaterability was predominant because of the fast of dewatering velocity and production of low water content cake, and also this known as economical efficiency was excellent because of low energy consumption in comparison with dryer.