• Journal of Climate Change Research
• /
• v.2 no.1
• /
• pp.15-32
• /
• 2011

Due to greenhouse gas increased by human activities, abnormal climate changes are continuously occurring everywhere in the world and internationally people make efforts to reduce the emission of greenhouse gas. Our country also is making endeavors to realize low carbon society on the foundation of the green growth and for this low carbon consumption pattern settlement through green life is necessary. Therefore for the nationals the offering of the information on greenhouse gas emission that is reduced through the change to low carbon life is required. In this study the objects are beef and wine whose weight of import is high among the beverages and foods consumed in the country and we calculated the food mileage and emission of carbon dioxide of the domestic and foreign product beef and wine and estimated the potential amount that can be reduced when replacing the imported products with domestic products. As the year 2007 being standard if we replace 10% of imported beef with domestic products it is possible to reduce 14,000 tons of carbon dioxide per year and on one day out of a year if we replace imported beef with domestic beef the reduction of 384 tons of carbon dioxide is appeared to be possible. In the same standard year if we replace 10% of imported wine with domestic product we can reduce 1,396 tons and on one day out of a year if we replace imported wine with domestic wine reduction of 38 tons of carbon per year appeared to be possible. Through active promotion and expansion of variety of domestic foods and beverages in the real life of the nationals the consumption pattern of natural low carbon life should be achieved and offering of more systematized greenhouse gas emission DB is thought to be necessary.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.50 no.4
• /
• pp.20-36
• /
• 2022

This study aims to provide basic data for high-quality street tree management by setting reasonable management items and appropriate unit prices by reviewing the adequacy of current street tree management. Currently, street tree management items, except for street tree pruning, use general landscape tree quantity per unit for the street tree management quantity per unit. KEPCO (Korea Electric Power Corporation) applied pruning items from standard electric production infrastructure and carried out the activities at an average unit price of 51% lower for heavy pruning and 39% lower for light pruning than the standard estimate. This was judged to be a level that could not maintain or increase the quality of street tree management. It was determined that an appropriate standard unit price for street tree management was necessary. To improve the quantity per unit for the proper management of street trees, it was necessary to review costs in the field. However, due to the absence of data on actual construction costs in the domestic landscape field, detailed items of the US RSMeans Building Construction Cost Data (RSMeans) were reviewed, and the actual construction costs were calculated by applying personal domestic expenses. As a result, the standard of the estimated unit showed a good ratio of 107% for heavy pruning of street tree pruning compared to the actual construction cost, but light pruning was underestimated with a 59% ratio. Shrub pruning was 82%, weeding was 92%, tree fertilization was 87%, and windbreak wall installation was 91% under-engineered. In addition, it was also confirmed that the watering by sprinkler trucks and chemical spraying were over-designed compared to the actual construction cost at the rates of 118% and 124%, respectively. Due to the specificity of the street trees, the increase in personal expenses and the input cost of equipment, such as road safety controls, were judged to be the main cause of the underestimation of items. Therefore, it is necessary to add items related to street trees and general landscape trees to the landscape maintenance items of the standard of the estimated unit.

• Korean Journal of Environment and Ecology
• /
• v.36 no.6
• /
• pp.676-683
• /
• 2022

Due to the effects of rapid changes in the living environment since 2000 and the recent unforeseen pandemic, people are refraining from domestic and international traveling and movement, and outdoor activities for health and the public value of forest trails, called Dullegil Trail in Korea, have become more important. This study estimated the environmental value of the tentatively named "East-West Trail," which connects the forest trails crossing Chungcheong and Gyeongsang Provinces using CVM (Contingent Valuation Method). It surveyed visitors to the East-West Trail, and 725 questionnaires were used for analysis. The average characteristics of respondents were those who exercised 2-3 times per week, visited a forest trail not far from their residence with friends or family, and showed a tendency to spend 50 thousand Korean won or more per visit. Visitors to the Dullegil Trail felt that there was a shortage of information boards on the forest trail, and they preferred a shelter in appropriate locations. We used a double-bounded dichotomous choice (BDDC) logit model proposed by Hanemann to measure the conservation value of the East-West Trail. It was estimated that the environmental value that a visitor to the East-West Trail could obtain was 30,087 won per trip. The estimated environmental value of the East-West Trail can be converted to about 94 billion won total visitors annually based on the population belonging to the direct-use zone near the East-West Trail. As there has been no study on the environmental value of forest trails using CVM, the results of this study will be able to suggest the feasibility of the government policies on forest trails.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.43 no.4
• /
• pp.257-267
• /
• 2023

The grassland section of the greenhouse gas inventory has limitations due to a lack of review and verification of biomass compared to organic carbon in soil while grassland is considered one of the carbon storages in terrestrial ecosystems. Considering the situation at internal and external where the calculation of greenhouse gas inventory is being upgraded to a method with higher scientific accuracy, research on standards and methods for calculating carbon accumulation of grassland biomass is required. The purpose of this study was to identify international trends in the calculation method of the grassland biomass sector that meets the Tier 2 method and to conduct a review of variables applicable to the Republic of Korea. Identify the estimation methods and access levels for grassland biomass through the National Inventory Report in the United Nations Framework Convention on Climate Change and type the main implications derived from overseas cases. And, a field survey was conducted on 28 grasslands in the Republic of Korea to analyse the applicability of major issues. Four major international issues regarding grassland biomass were identified. 1) country-specific coefficients by land use; 2) calculations on woody plants; 3) loss and recovery due to wildfire; 4) amount of change by human activities. As a result of field surveys and analysis of activity data available domestically, it was found that there was a significant difference in the amount of carbon in biomass according to use type classification and climate zone-soil type classification. Therefore, in order to create an inventory of grassland biomass at the Tier 2 level, a policy and institutional system for making activity data should develop country-specific coefficients for climate zones and soil types.

• The Korean Journal of Nuclear Medicine
• /
• v.1 no.1
• /
• pp.55-66
• /
• 1967

In view of its prevalence in the Far East area, a more detailed knowledge on the hookworm infection is one of the very important medical problems. The present study was aimed to; determine the infectivity of the artificially hatched ancylostoma duodenale larvae in man after its oral administration, evaluate the clinical symptomatology of such infection, determine the date of first appearance of the ova in the stool, calculate the blood loss per worm per day, assess the relation-ships between the ova count, infectivity(worm load), blood loss and severity of anemia. An erythrokinetic study was also done to analyse the characteristics of hookworm anemia by means of $^{59}Fe\;and\;^{51}Cr$. Materials and Methods Ten healthy male volunteers(doctors, medical students and laboratory technicians) with the ages ranging from 21 to 40 years were selected as the experimental materials. They had no history of hookworm infection for preceding several years, and care was taken not to be exposed to reinfection. A baseline study including a through physical examinations and laboratory investigations such as complete blood counts, stool examination and estimation of the serum iron levels was done, and a vermifuge, bephenium hydroxynaphoate, was given 10 days prior to the main experiment. The ancylostoma duodenale filariform larvae were obtained in the following manner; The pure ancylostoma duodenale ova were obtained from the hookworm anemia patients and a modified filter paper method was adopted to harvest larger number of infective larvae, which were washed several times with saline. The actively moving mature larvae were put into the gelatine capsules, 150 in each, and were given to the volunteers in the fasting state with 300ml. of water. The volunteers were previously treated with intramuscular injection of 15mg. of chlorpromazine in order to prevent the eventual nausea and vomiting after the larvae intake. The clinical symptoms and signs mainly of the respiratory and gastrointestinal tracts, appearance of the ova and occult blood in the stool etc. were checked every day for the first 20 days and then twice weekly until the end of the experiment, which usually lasted for about 3 months. Roentgenological survey of the lungs was also done. The hematological changes such as the red blood cell, white blood cell and eosinophil cell counts, hemoglobin content and serum iron levels were studied. The appearance of the ova in the stool was examined by the formalin ether method and the ova were counted in triplicate on two successive days using the Stoll's dilution method. The ferrokinetic data were calculated by the modified Huff's method and the apparent half survival time of the red blood cells by the modified Gray's method. The isotopes were simultaneously tagged and injected intravenously, and then the stool and blood samples were collected as was described by Roche et al., namely, three separate 4-day stool samples with the blood sample drawing before each 4-day stool collection. The radio-activities of the stools ashfied and the blood were separately measured by the pulse-height analyser. The daily blood loss was calculated with the following formula; daily blood loss in $ml.=\frac{cpm/g\;stool{\times}weight\;in\;g\;of\;4-day\;stool}{cpm/ml\;blood{\times}4}$ The average of these three 4-day periods was given as the daily blood loss in each patient. The blood loss per day per worm was calculated by simply dividing the daily blood loss by the number of the hookworm recovered after the vermifuge given twice a week at the termination of the experiment. The iron loss in mg. through the gastrointestinal tract was estimated with the daily iron loss in $mg=\frac{g\;Hgb/100ml{\times}ml\;daily\;blood\;loss{\times}3.40}{100}$ 3.40=mg of iron per g Hgb following formula; Results 1. The respiratory symptoms such as cough and sputum were noted in almost all cases within a week after the infection, which lasted about 2 weeks. The roentgenological findings of the chest were essentially normal. A moderate degree of febril reaction appeared within 2 weeks with a duration of 3 or 4 days. 2. The gastrointestinal symptoms such as nausea, epigastric fullness, abdominal pain and loose bowel appeared in all cases immediately after the larvae intake. 3. The reduction of the red blood cell count was not remarkable, however, the hemoglobin content and especially the serum iron level showed the steady decreases until the end of the experiment. 4. The white blood cells and eosinophil cells, on the contrary, showed increases in parallel and reached peaks in 20 to 30 days after the infection. A small secondary rise was noted in 2 months. 5. The ova first appeared in the stool in 40. 1 days after the infection, ranging from 29 to 51 days, during which the occult blood reaction of the stool became also positive in almost cases. 6. The number of ova recovered per day was 164, 320 on the average, ranging from 89,500 to 253,800. The number of the worm evacuated by vermifuge was in rough correlation with the number of ova recovered. 7. The infectivity of ancylostoma duodenale was 14% on the average, ranging from 7.3 to 20.0%, which is relatively lower than those reported by other workers. 8. The mean fecal blood loss was 5.78ml. per day, with a range of from 2.6 to 11.7ml., and the mean blood loss per worm per day was 0.30ml., with a range of from 0.13 to 0.73ml., which is in rough coincidence with those reported by other authors. There appeared to exist, however, no correlation between the blood loss and the number of ova recovered. 9. The mean fecal iron loss was 2.02mg. per day, with a range of from 1.20 to 3.89mg., which is less than those appeared in the literature. 10. The mean plasma iron disappearance rate was 0.80hr., with a range of from 0.62 to 0.95hr., namely, a slight accerelation. 11. The hookworm anemia appeared to be iron deficiency in origin caused by continuous intestinal blood loss.

• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.1-19
• /
• 2018

Large amount of data is now available for research and business sectors to extract knowledge from it. This data can be in the form of unstructured data such as audio, text, and image data and can be analyzed by deep learning methodology. Deep learning is now widely used for various estimation, classification, and prediction problems. Especially, fashion business adopts deep learning techniques for apparel recognition, apparel search and retrieval engine, and automatic product recommendation. The core model of these applications is the image classification using Convolutional Neural Networks (CNN). CNN is made up of neurons which learn parameters such as weights while inputs come through and reach outputs. CNN has layer structure which is best suited for image classification as it is comprised of convolutional layer for generating feature maps, pooling layer for reducing the dimensionality of feature maps, and fully-connected layer for classifying the extracted features. However, most of the classification models have been trained using online product image, which is taken under controlled situation such as apparel image itself or professional model wearing apparel. This image may not be an effective way to train the classification model considering the situation when one might want to classify street fashion image or walking image, which is taken in uncontrolled situation and involves people's movement and unexpected pose. Therefore, we propose to train the model with runway apparel image dataset which captures mobility. This will allow the classification model to be trained with far more variable data and enhance the adaptation with diverse query image. To achieve both convergence and generalization of the model, we apply Transfer Learning on our training network. As Transfer Learning in CNN is composed of pre-training and fine-tuning stages, we divide the training step into two. First, we pre-train our architecture with large-scale dataset, ImageNet dataset, which consists of 1.2 million images with 1000 categories including animals, plants, activities, materials, instrumentations, scenes, and foods. We use GoogLeNet for our main architecture as it has achieved great accuracy with efficiency in ImageNet Large Scale Visual Recognition Challenge (ILSVRC). Second, we fine-tune the network with our own runway image dataset. For the runway image dataset, we could not find any previously and publicly made dataset, so we collect the dataset from Google Image Search attaining 2426 images of 32 major fashion brands including Anna Molinari, Balenciaga, Balmain, Brioni, Burberry, Celine, Chanel, Chloe, Christian Dior, Cividini, Dolce and Gabbana, Emilio Pucci, Ermenegildo, Fendi, Giuliana Teso, Gucci, Issey Miyake, Kenzo, Leonard, Louis Vuitton, Marc Jacobs, Marni, Max Mara, Missoni, Moschino, Ralph Lauren, Roberto Cavalli, Sonia Rykiel, Stella McCartney, Valentino, Versace, and Yve Saint Laurent. We perform 10-folded experiments to consider the random generation of training data, and our proposed model has achieved accuracy of 67.2% on final test. Our research suggests several advantages over previous related studies as to our best knowledge, there haven't been any previous studies which trained the network for apparel image classification based on runway image dataset. We suggest the idea of training model with image capturing all the possible postures, which is denoted as mobility, by using our own runway apparel image dataset. Moreover, by applying Transfer Learning and using checkpoint and parameters provided by Tensorflow Slim, we could save time spent on training the classification model as taking 6 minutes per experiment to train the classifier. This model can be used in many business applications where the query image can be runway image, product image, or street fashion image. To be specific, runway query image can be used for mobile application service during fashion week to facilitate brand search, street style query image can be classified during fashion editorial task to classify and label the brand or style, and website query image can be processed by e-commerce multi-complex service providing item information or recommending similar item.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.