• Journal of Broadcast Engineering
• /
• v.25 no.3
• /
• pp.386-398
• /
• 2020

We developed an integrative annotation program that can perform data labeling process for deep learning models in object recognition. The program utilizes the basic GUI library of Python and configures crawler functions that allow data collection in real time. Retinanet was used to implement an automatic annotation function. In addition, different data labeling formats for Pascal-VOC, YOLO and Retinanet were generated. Through the experiment of the proposed method, a domestic vehicle image dataset was built, and it is applied to Retinanet and YOLO as the training and test set. The proposed system classified the vehicle model with the accuracy of about 94%.

• Journal of the Korean Society of Radiology
• /
• v.81 no.6
• /
• pp.1274-1289
• /
• 2020

Deep learning has recently become one of the most actively researched technologies in the field of medical imaging. The availability of sufficient data and the latest advances in algorithms are important factors that influence the development of deep learning models. However, several other factors should be considered in developing an optimal generalized deep learning model. All the steps, including data collection, labeling, and pre-processing and model training, validation, and complexity can affect the performance of deep learning models. Therefore, appropriate optimization methods should be considered for each step during the development of a deep learning model. In this review, we discuss the important factors to be considered for the optimal development of deep learning models.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.83-88
• /
• 2024

In this research, we propose a Semi-Supervised learning based railroad surface defect detection method. The Resnet50 model, pretrained on ImageNet, was employed for the training. Data without labels are randomly selected, and then labeled to train the ResNet50 model. The trained model is used to predict the results of the remaining unlabeled training data. The predicted values exceeding a certain threshold are selected, sorted in descending order, and added to the training data. Pseudo-labeling is performed based on the class with the highest probability during this process. An experiment was conducted to assess the overall class classification performance based on the initial number of labeled data. The results showed an accuracy of 98% at best with less than 10% labeled training data compared to the overall training data.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.29 no.4
• /
• pp.359-365
• /
• 2011

This study aims at filtering ALS points into ground and non-ground effectively through labeling and window based algorithm by utilizing 2D adjacency based on scan line. Firstly, points adjacency is constructed through minimal search based on scan line. Connected component labeling algorithm is applied to classify raw ALS points into ground and non-ground by utilizing the adjacency structure. Then, some small objects are removed by morphology filtering, and isolated ground points are restored by IDW estimation. The experimental results shows that the method provides good filtering performance( about 97% accuracy) for diverse sites, and the overall processing takes less time than converting raw data into TIN or raster grid.

• Journal of the East Asian Society of Dietary Life
• /
• v.25 no.6
• /
• pp.942-951
• /
• 2015

This study compared nutrition knowledge, dietary behaviors, and checking behaviors of food and nutrition labels between Korean and Chinese university students to obtain useful data for development of an education program for healthy dietary life. The data were collected by a self-administered questionnaire in Korea and China. Frequencies, t tests, ${\chi}^2$ tests, Cronbach's ${\alpha}$, and Pearson's correlation coefficient analysis were conducted by SPSS Win. V.21.0. The levels of nutrition knowledge and dietary behaviors were not high. Korean students showed higher percentage of correct answers about nutrition knowledge and levels of dietary behaviors than Chinese students. The means of degree of checking contents of food labels were 3.46 points for Korean students and 3.11 for Chinese students. Both groups of students showed the highest degree of checking milk and dairy products. The degree of understanding nutritive component labeling of Chinese students was higher than that of Korean students. Both groups of students showed higher than normal levels of confidence about nutritive component labeling and necessity of education on food and nutrition labels. The most preferred method of education on food and nutrition labels was broadcast media for both groups of students. In addition, there were significant correlations among nutrition knowledge, dietary behaviors, checking degree of food labels, checking degree of nutritive component labeling, and experience of nutrition education.

• Archives of Plastic Surgery
• /
• v.32 no.4
• /
• pp.533-538
• /
• 2005

Due to its safety and softness, autologous fat transplantation has been commonly performed for soft tissue correction. However, the injected fat is absorbed resulting in the reduction of volume of the graft by 40-60% within a few months. Thus, there was an attempt to use adipocytes differentiated from preadipocytes in vitro for transplantation. Differentiated adipocytes were biocompatible and matured with gradual volume increase at transplantation site in clinical study(unpublished data). In addition, they did not induce immune rejection in response to nonself lymphocytes in a mixed lymphocyte reaction(MLR)(unpublished data). The purpose of this study is to differentiate mouse preadipocytes following labeling into adipocytes to establish an animal model for allogenic transplantation. Preadipocytes isolated from inguinal and retroperitoneal fat pad of C57BL/6 mice were proliferated with growth medium by passage 3 and differentiated into adipocytes with different culture conditions after labeled with BrdU. At most suitable conditions, above 90% of preadipocytes were differentiated and BrdU labeling did not affect differentiation rate and function of differentiated adipocytes. These results demonstrate that BrdU-labeled adipocytes resulting from this in vitro differentiation protocol are useful for allogenic transplantation study.

• Journal of the Korean Home Economics Association
• /
• v.38 no.12
• /
• pp.29-45
• /
• 2000

The purposes of this study were to determine what factors influence risk perceptions of females and males for milk produced using food-related biotechnology, to test whether risk perceptions or other factors influence self-protection actions and to estimate milk demand response in light of self-protection actions and other economic and demographic factors. The expected utility model was applied to explain the way consumers would take self-protection actions regarding risk perceptions and to drive milk demand. Telephone interviews were conducted and the data were collected from households(females=1,029, males=437) nationwide in the U.S. And the data were analyzed by Heckman two-step method using the software package LIMDEP. Risk perceptions were found to be influenced not by demographic factors but by outrage factors as well as attitudinal factors in both females and males, although some factors were different. In addition, risk perceptions and labeling availability were found to significantly influence self-protection actions in both groups. Furthermore, as an important concern in this study, self-protection action was found to significantly influence milk demand in only male group, implying a consistent behavior of males. Also milk price and household size were found to significantly influence milk demand in both groups. In fact, the results did demonstrate that labeling availability significantly influenced self-protection actions. That is, in markets where labeled laternatives were present, concerned consumers were more likely to self protect by substituting to these products. A policy implication of this result is that labeling food products produced using biotechnology enhances consumer choice. Hence, consumer could express a more accurate demand response and reduce the perceived food safety risk. Furthermore, education for females might be necessary to have a consistent behavior because self-protection action did not significantly influence female's milk demand, though they have greater risk perceptions than males have.

• The KIPS Transactions:PartA
• /
• v.9A no.4
• /
• pp.525-534
• /
• 2002

Race conditions, races in short, need to be detected for debugging parallel programs, because the races result in unintended non-deterministic executions. To detect the races in an execution of program, previous techniques use a centralized data structure which may incur serious bottleneck in generating concurrency information, or show inefficient time complexity which depends on the degree of nested parallelism in comparing any two of them. We propose a new labeling scheme in this paper, which is scalable in generating the concurrency information without bottleneck by using private data structure, and improves time complexity into constant in checking concurrency. The scalability and time efficiency therfore makes on-the-fly race detection efficient not only for programs with either shared-memory or message-passing, but also for programs with mixed model of the two.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.35 no.12
• /
• pp.43-53
• /
• 2019

The construction industry has made great strides in the past decades by utilizing computer programs including CAD. However, compared to other manufacturing sectors, labor productivity is low due to the high proportion of workers' knowledge-based task in addition to simple repetitive task. Therefore, the knowledge-based task efficiency of workers should be improved by recognizing the visual information of computers. A computer needs a lot of training data, such as the ImageNet project, to recognize visual information. This study, aim at proposing building facade datasets that is efficiently constructed by quickly collecting building facade data through portal site road view and automatically labeling using deep learning as part of construction of image dataset for visual recognition construction by the computer. As a method proposed in this study, we constructed a dataset for a part of Dongseong-ro, Daegu Metropolitan City and analyzed the utility and reliability of the dataset. Through this, it was confirmed that the computer could extract the significant facade information of the portal site road view by recognizing the visual information of the building facade image. Additionally, In contribution to verifying the feasibility of building construction image datasets. this study suggests the possibility of securing quantitative and qualitative facade design knowledge by extracting the facade design knowledge from any facade all over the world.

• Journal of the Korean Dietetic Association
• /
• v.30 no.3
• /
• pp.161-180
• /
• 2024

Social and environmental changes, such as the rise of single-person households and advances in the food industry, have led to the replacement of home-cooked meals with home meal replacements products (HMRs). This study compared the nutrient content of a total of 1,680 HMRs and 158 restaurant foods by collecting data on the nutrient content of comparable food types from the Food composition data for restaurant foods published by the Ministry of Food and Drug Safety (MFDS) and evaluating the calorie and nutrient content of HMRs based on nutrition labeling through market research from May 2022 to May 2024, focusing on rice, porridge, noodles, soup/stew. The nutritional content and price of the HMRs varied widely, even for similar foods, depending on the detailed food type. Therefore, it is necessary to make an appropriate choice based on nutrition labeling according to the purpose of consumption. The HMRs had a lower calorie and nutrient content due to the smaller serving size when compared with restaurant foods. However, when the same weights were compared, the sugar and sodium content in the rice and soup/stew were higher in the HMRs than in the restaurant foods. In addition, due to the wide variety of HMRs available, many HMRs that can replace restaurant foods are being produced. However, even for the same type of food, the serving sizes of the HMRs and the restaurant foods were widely different, suggesting the need for a study to examine the appropriateness of the serving sizes of HMRs and restaurant foods.