• Journal of the Society of Disaster Information
• /
• v.18 no.3
• /
• pp.433-443
• /
• 2022

Purpose: In this paper, we propose a study of deep learning algorithms that estimate and predict sarcopenia by exploiting the high penetration rate of smart devices. Method: To utilize deep learning techniques, experimental data were collected by using the inertial sensor embedded in the smart device. We implemented a smart device application for data collection. The data are collected by labeling normal and abnormal gait and five states of running, falling and squat posture. Result: The accuracy was analyzed by comparative analysis of LSTM, CNN, and RNN models, and binary classification accuracy of 99.87% and multiple classification accuracy of 92.30% were obtained using the CNN-LSTM fusion algorithm. Conclusion: A study was conducted using a smart sensoring device, focusing on the fact that gait abnormalities occur for people with sarcopenia. It is expected that this study can contribute to strengthening the safety issues caused by sarcopenia.

• Journal of the Microelectronics and Packaging Society
• /
• v.30 no.4
• /
• pp.69-78
• /
• 2023

With the rapid growth of artificial intelligence, the demand for semiconductors is enormously increasing everywhere. To ensure the manufacturing quality and quantity simultaneously, the importance of automatic defect detection during the packaging process has been re-visited by adapting various deep learning-based methodologies into automatic packaging defect inspection. Deep learning (DL) models require a large amount of data for training, but due to the nature of the semiconductor industry where security is important, sharing and labeling of relevant data is challenging, making it difficult for model training. In this study, we propose a new framework for securing sufficient data for DL models with fewer computing resources through a divide-and-conquer approach. The proposed method divides high-resolution images into pre-defined sub-regions and assigns conditional labels to each region, then trains individual sub-regions and boundaries with boundary loss inducing the globally coherent and seamless images. Afterwards, full-size image is reconstructed by combining divided sub-regions. The experimental results show that the images obtained through this research have high efficiency, consistency, quality, and generality.

• Journal of Internet Computing and Services
• /
• v.25 no.2
• /
• pp.11-19
• /
• 2024

This paper was studied abouta technology for detecting damage to temporary works equipment used in construction sites with explainable artificial intelligence (XAI). Temporary works equipment is mostly composed of steel or aluminum, and it is reused several times due to the characters of the materials in temporary works equipment. However, it sometimes causes accidents at construction sites by using low or decreased quality of temporary works equipment because the regulation and restriction of reuse in them is not strict. Currently, safety rules such as related government laws, standards, and regulations for quality control of temporary works equipment have not been established. Additionally, the inspection results were often different according to the inspector's level of training. To overcome these limitations, a method based with AI and image processing technology was developed. In addition, it was devised by applying explainableartificial intelligence (XAI) technology so that the inspector makes more exact decision with resultsin damage detect with image analysis by the XAI which is a developed AI model for analysis of temporary works equipment. In the experiments, temporary works equipment was photographed with a 4k-quality camera, and the learned artificial intelligence model was trained with 610 labelingdata, and the accuracy was tested by analyzing the image recording data of temporary works equipment. As a result, the accuracy of damage detect by the XAI was 95.0% for the training dataset, 92.0% for the validation dataset, and 90.0% for the test dataset. This was shown aboutthe reliability of the performance of the developed artificial intelligence. It was verified for usability of explainable artificial intelligence to detect damage in temporary works equipment by the experiments. However, to improve the level of commercial software, the XAI need to be trained more by real data set and the ability to detect damage has to be kept or increased when the real data set is applied.

• Journal of Food Hygiene and Safety
• /
• v.16 no.1
• /
• pp.66-75
• /
• 2001

A survey was conducted to investigate consumers'attitudes toward the foods developed by gene recombination techniques from December, 1999 to April, 2000. The questionnaires were mailed to 1,500 people, and the 1,101 people responded. The consumers were asked about knowledge, acceptance, intention of purchasing, and labeling information. Although the portion of the consumers (88.8%) knowing the genetically modified floods (GMF) was lower than that of the flood expert group (98.7%), many consumers had some knowledge on the GMF, which may be influenced by news released from mass media. Seventy-nine percent of the consumers responded that gene recombination technology is necessary in food production, which is similar to the findings on the survey of the expert group. The portion of the consumers responding that these foods are potentially hazard was 88.1%, which is a little higher than the data (80.9%) from the expert group. The consumers having greater knowledge less worried about a potential hazard of the gene recombinant foods (p<0.01). Although 62.9% of the consumers responded to be willing to purchase those foods, only 16.2% of them responded to purchase the foods with no conditions, which is lower to that from the expert group (23.5%). There was no statistically significant relationship between the knowledge and the intention of purchasing. The ninety point three percent of the consumers wanted the information on gene recombination to be labeled on the foods. The data from this survey suggest that knowledge of the consumers on the GMF are not accurate, so proper strategy for consumer education may need to be developed. In addition, it is necessary to improve safety assessment system and analytical techniques for genetically modified foods (GMF) and to build pre- and post-market surveillance system fur efficient implementation of the GMF labeling.

• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.185-197
• /
• 2022

Recently, deep learning networks with high performance for object recognition are emerging. In the case of object recognition using deep learning, it is important to build a training data set to improve performance. To build a data set, we need to collect and label the images. This process requires a lot of time and manpower. For this reason, open data sets are used. However, there are objects that do not have large open data sets. One of them is data required for license plate detection and recognition. Therefore, in this paper, we propose an artificial license plate generator system that can create large data sets by minimizing images. In addition, the detection rate according to the artificial license plate arrangement structure was analyzed. As a result of the analysis, the best layout structure was FVC_III and B, and the most suitable network was D2Det. Although the artificial data set performance was 2-3% lower than that of the actual data set, the time to build the artificial data was about 11 times faster than the time to build the actual data set, proving that it is a time-efficient data set building system.

• Journal of the Korean Society of Costume
• /
• v.55 no.2 s.92
• /
• pp.116-128
• /
• 2005

The aim of this study is to understand problems with both fit and sizing system of gloves by analyzing the glove production industries with an emphasis on the sizing system, production measurement system, and general marking situation. Also, to suggest the basic raw materials for improving sizing system for gloves, actual glove control sizes were compared to the anthropometric data of the previous study Fifteen domestic production companies were participated in this study-Domestic glove production companies established their own sizing system by copying inter-national brand's glove sizing system or by their experience. The Korean Standard of gloves and the 1997 Korean Standard Anthropometrics Measurement for producing glove patterns are not considered because of its discordance with the reality of the required measurements. Domestic glove production companies used different size designation and labeling system. Size measurement unit also showed difference among the glove production companies. Some companies used 'cm', some used 'mm', some used 'inch' for the measurement unit. In general, companies produced 5 to 4 sizes in one design of glove and the production was the highest in M and L size. In 9 out of 15 companies preferred control size as hand length and hand circumference. For reference size, most of the companies preferred finger circumference, finger length, palm length, hand breadth, crotch height, and hand thickness. Actual glove sizes were compared to the anthropometric data of the previous study. The results indicated that most of the measurements of actual glove sizes were significantly larger than the anthropometric data.

• Journal of the Korean Society of Food Culture
• /
• v.12 no.5
• /
• pp.509-517
• /
• 1997

The purpose of this study is to establish the one serving size of Korean Indigeneous Food. Serving size is necessary to make Nutrition Labeling which is required to export Korean food product especially to the United States of America. The basic data of 100 food items were collected through searching traditional and recent cookbooks. 4 industrial foodservices as noncommercial foodservice and 30 traditional ethnic restaurants and 12 gourmet restaurants in hotels as commercial foodservlce were explored to collect the data of actual serving size of each items. Moreover, experimental cooking and sensory evaluation by trained panels were conducted to assess quantity preference of selected food items. All data were rearranged through food type, that is, main dish, side dish, dessert and health food. One serving sizes showed wide variety according to the different menus that include selected food items. Therefore, means and ranges of serving size by four research methods were presented item by item. There were wide differences in intakes of main dishes, for example, noodles were around $50{\sim}100g$, cereals were 20 g, which means the one serving size can be differenciated by the food usage. In intakes of side dishes, average of side dishes were $20{\sim}30g$, but Kimches, the first traditional Korean food, were $30{\sim}50g$, and the other condiments, pepper paste and soy paste were $5{\sim}10g$. About desserts, liquid types were around 200 g, the other sugars were $10{\sim}20g$, the kind of teas were almost $2{\sim}3g$. The health foods-many kinds of that were Ginseng-were averaged 20 g; but dried mushrooms were around 2 g.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.4
• /
• pp.231-241
• /
• 2006

Creating controllable, responsive avatars is an important problem in computer games and virtual environments. Recently, large collections of motion capture data have been exploited for increased realism in avatar animation and control. Large motion sets have the advantage of accommodating a broad variety of natural human motion. However, when a motion set is large, the time required to identify an appropriate sequence of motions is the bottleneck for achieving interactive avatar control. In this paper, we present a novel method for training avatar behaviors from unlabelled motion data in order to animate and control avatars at minimal runtime cost. Based on machine learning technique, called Q-teaming, our training method allows the avatar to learn how to act in any given situation through trial-and-error interactions with a dynamic environment. We demonstrate the effectiveness of our approach through examples that include avatars interacting with each other and with the user.

• Journal of Pharmaceutical Investigation
• /
• v.36 no.1
• /
• pp.1-9
• /
• 2006

Human solid tumors exhibit a multicellular resistance (MCR) resulting from limited drug penetration and decreased sensitivity of tumor cells when interacting with their microenvironments. Multicellular cultures represent solid tumor condition in vivo and provide clinically relevant data. There is little data on antitumor effect of paclitaxel (PTX) in multicellular cultures although its MCR has been demonstrated. In the present study, we evaluated antiproliferative effects of PTX in multicellular layers (MCL) of DLD-1 human colorectal carcinoma cells. BrdU labeling index (LI), thickness of MCL, cell cycle distribution and cellular uptake of calcein were measured before and after exposure to PTX at 0.1 to 50 ${\mu}M$ for 24, 48 and 72 hrs. BrdU LI and thickness of MCL showed a concentration- and time-dependent decrease and the changes in both parameters were similar, i.e., 34.2% and 40.6% decrease in BrdU LI and thickness, respectively, when exposed to $50\;{\mu}M$ for 72 hr. The DLD-1 cells grown in MCL showed increase in $%G_{0}/G_{1}$ and resistance to cell cycle arrest and apoptosis compared to monolayers. Calcein uptake in MCL did not change upon PTX exposure, indicating technical problems in multicellular system. Overall, these data indicate that antitumor activity of PTX may be limited in human solid tumors (a multicellular system) and MCL may be an appropriate model to study further pharmacodynamics of PTX.

• Korean Journal of Community Nutrition
• /
• v.21 no.3
• /
• pp.293-300
• /
• 2016

Objectives: The purpose of the study was to develop a dietary fiber composition table (DFCT) and to assess dietary fiber intakes in Korea National Health and Nutrition Examination Survey (KNHANES). Methods: The DFCT was developed by compiling the food composition tables published by the agencies of Korea, United States, or Japan. When there was no available data from the same species or status (dried, boiled, etc.) of food, the values were imputed by estimating from the same species with different status or substituting familiar species in biosystematic grouping. Using KNHANES VI-2 (2014) microdata and DFCT, intake of dietary fiber of Koreans was estimated. Results: Among the 5,126 food items of DFCT, the proportion of items of which dietary fiber contents were taken from the analytical values of the same foods was 40.9%. The data from the domestic food composition tables was 37.5%, and the data from the foreign tables was 49.6%. The rest was assumed as zero, or estimated with recipe database and nutrition labeling. Mean daily intake of dietary fiber was 23.2 g, and mean intake per 1,000 kcal was 10.7 g in men and 12.6 g in women. The mean percentage of dietary fiber intake compared to adequate intake was higher than 100%. The major food groups contributing to dietary fiber intakes were vegetables and cereals, and the percent contribution were 32.9% and 23.0% of total dietary fiber intakes, respectively. Conclusions: This DFCT could serve as a useful database for assessing dietary fiber intakes and for investigating the association between dietary fiber intakes and noncommunicable diseases.