• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.7
• /
• pp.1047-1055
• /
• 2015

We proposes a method for improving the identification and tracking of the moving objects in intelligent video surveillance system. The proposed method consists of 3 parts: object detection, object recognition, and object tracking. First of all, we use a GMM(Gaussian Mixture Model) to eliminate the background, and extract the moving object. Next, we propose a labeling technique forrecognition of the moving object. and the method for identifying the recognized object by using the optical flow and EMD algorithm. Lastly, we proposes method to track the location of the identified moving object regions by using location information of moving objects and Kalman filter. Finally, we demonstrate the feasibility and applicability of the proposed algorithms through some experiments.

• Nutrition Research and Practice
• /
• v.9 no.6
• /
• pp.667-672
• /
• 2015

BACKGROUND/OBJECTIVES: To encourage healthier food choices for children in fast-food restaurants, many initiatives have been proposed. This study aimed to examine the effect of disclosing nutritional information on parents' meal choices for their children at fast-food restaurants in South Korea. SUBJECTS/METHODS: An online experimental survey using a menu board was conducted with 242 parents of children aged 2-12 years who dined with them at fast-food restaurants at least once a month. Participants were classified into two groups: the low-calorie group (n = 41) who chose at least one of the lowest calorie meals in each menu category, and the high-calorie group (n = 201) who did not. The attributes including perceived empowerment, use of provided nutritional information, and perceived difficulties were compared between the two groups. RESULTS: The low-calorie group perceived significantly higher empowerment with the nutritional information provided than did the high-calorie group (P = 0.020). Additionally, the low-calorie group was more interested in nutrition labeling (P < 0.001) and considered the nutritional value of menus when selecting restaurants for their children more than did the high-calorie group (P = 0.017). The low-calorie group used the nutritional information provided when choosing meals for their children significantly more than did the high-calorie group (P < 0.001), but the high-calorie group had greater difficulty using the nutritional information provided (P = 0.012). CONCLUSIONS: The results suggest that improving the empowerment of parents using nutritional information could be a strategy for promoting healthier parental food choices for their children at fast-food restaurants.

• Journal of the Institute of Convergence Signal Processing
• /
• v.20 no.3
• /
• pp.132-137
• /
• 2019

The amount of data generated from medical images is increasingly exceeding the limits of professional visual analysis, and the need for automated medical image analysis is increasing. For this reason, this study evaluated the classification and accuracy according to the presence or absence of tumor using Inception V3 deep learning model, using MRI medical images showing normal and tumor findings. As a result, the accuracy of the deep learning model was 90% for the training data set and 86% for the validation data set. The loss rate was 0.56 for the training data set and 1.28 for the validation data set. In future studies, it is necessary to secure the data of publicly available medical images to improve the performance of the deep learning model and to ensure the reliability of the evaluation, and to implement modeling by improving the accuracy of labeling through labeling classification.

• Convergence Security Journal
• /
• v.23 no.2
• /
• pp.55-66
• /
• 2023

Recently, there is an increasing movement to increase the value of AI learning data and to secure high-quality data based on previous research on AI learning data in all areas of society. Therefore, quality management is very important in construction projects to secure high-quality data. In this paper, quality management to secure high-quality data when building AI learning data and improvement plans for each construction process are presented. In particular, more than 80% of the data quality of unstructured data built for AI learning is determined during the construction process. In this paper, we performed quality inspection of image/video data. In addition, we identified inspection procedures and problem elements that occurred in the construction phases of acquisition, data cleaning, labeling, and models, and suggested ways to secure high-quality data by solving them. Through this, it is expected that it will be an alternative to overcome the quality deviation of data for research groups and operators participating in the construction of AI learning data.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.189-196
• /
• 2024

Frequent false positives alarm from the Intelligent Selective Control System have raised significant concerns. These persistent issues have led to declines in operational efficiency and market credibility among agents. Developing a new model or replacing the existing one to mitigate false positives alarm entails substantial opportunity costs; hence, improving the quality of the training dataset is pragmatic. However, smaller organizations face challenges with inadequate capabilities in dataset collection and refinement. This paper proposes an automatic human pose data collection system centered around a human pose estimation model, utilizing camera-based sensor fusion techniques and edge devices. The system facilitates the direct collection and real-time processing of field data at the network periphery, distributing the computational load that typically centralizes. Additionally, by directly labeling field data, it aids in constructing new training datasets.

• Journal of Food Hygiene and Safety
• /
• v.38 no.6
• /
• pp.496-507
• /
• 2023

This study investigated species identification and labeling compliance of 48 shrimp products sold in the Korean online markets. Species identification was conducted using the standard DNA barcoding method, using the cytochrome c oxidase subunit I gene. The obtained sequences were compared with those deposited in the NCBI GenBank and BOLD Systems databases. Additionally, phylogenetic analysis was performed to further verify the identified shrimp species. Consequently, 16 shrimp species were identified, including Penaeus vannamei, Pandalus borealis, Palaemon gravieri, Leptochela gracilis, Penaeus monodon, Pleoticus muelleri, Metapenaeopsis dalei, Euphausia pacifica, Lebbeus groenlandicus, Trachypenaeus curvirostris, Argis lar, Metanephrops thomsoni, Metapenaeopsis barbata, Alpheus japonicus, Penaeus chinensis, and Mierspenaeopsis hardwickii. The most prevalent species was Penaeus vannamei, found in 45.8% of the analyzed products. A significant mislabeling rate of 72.9% was found; however, upon excluding generic names such as shrimp, the mislabeling rate reduced to 10.4%. The mislabeling rate was higher in highly-processed products (89.3%) compared with that in minimally-processed products (50%). No correlation was found between the country of origin and mislabeling rate. The results of this study provide crucial data for future monitoring of shrimp products and improving the labeling of shrimp species in Korea.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.6
• /
• pp.527-536
• /
• 2016

This study aims to generate baseline data for vitalizing the sales of functional cosmetics through an Importance-Performance Analysis (IPA) of the selection attributes of functional cosmetics. From the analysis of consumers' selection criteria, the study will assist functional cosmetics companies in reflecting consumer demands and therefore securing competitiveness. For this, general consumers aged over 20 years were surveyed for 5 weeks from Feb 23 through Mar 30, 2015, and 447 empirical data (response rate 88.9%) were processed through SPSS WIN 21.0 program for analysis. To conduct gender difference analysis on the IPA of the selection attributes of functional cosmetics, 17 selection attributes were categorized into 4 factors: functionality, labeling, popularity, and product. Cronbach's alpha for all factors was 0.5, proving the internal consistency and reliability of the survey. The survey results showed that while the entire average came out significantly higher for females (5.89/7points) than for males (5.66/7points) (p<0.001), the selection attributes 'anti-wrinkling', 'whitening function', 'functionality', 'expiration date', 'full ingredient labeling system' and 'various promotional events' showed significant gender differences. IPA results pertaining to gender showed 'price', 'functionality', 'spreadability' and 'full ingredient labeling system' as 2nd quadrant attributes, whereas female consumers selected 'price', 'whitening function', 'anti-wrinkling', 'functionality' and 'full ingredient labeling system' as attributes. Results show that businesses in the field of cosmetics and related areas need to prioritize improving the following factors that received low satisfaction from all consumers: 'price', 'functionality', and 'total labeling.' In particular, the 'price' aspects are considered to require reasonable and affordable pricing.

• Journal of Korean Home Economics Education Association
• /
• v.23 no.1
• /
• pp.113-141
• /
• 2011

The purpose of this study was to examine the effects of home economics instruction in food labeling using a motivational(ARCS-Attention, Relevance, Confidence, and Satisfaction) strategy to increase middle school students' learning motivation, recognition and use of food labels. To achieve this purpose, teaching-learning plans of food label instruction using a motivation(ARCS) strategy were developed over four class periods using a pretest-posttest experimental design. The experiment was conducted across two groups as follows: 4 experimental groups that received the motivation(ARCS) strategy instruction, and 3 comparative groups that received lecture type instruction. The pretest-posttest scores of the experimental and comparative groups were compared. The 203 data of questionnaires for the experiment were analyzed and evaluated by Analysis of Covariance(ANCOVA) using SPSS Win 12,0. The results of this study were as follows: First, teaching-learning plans, learning materials, and teacher reference materials for the home economics food label instruction that applied the motivation(ARCS) strategy were developed in five subject areas: nutrition labels, food additives, genetically modified food, irradiated food, and food quality verification labels. Second, students' learning motivation of the two groups showed statistically meaningful differences. Home economics instruction using a motivation(ARCS) strategy was more effective in increasing students' learning motivation than lecture type instruction. Third, as a result of ANCOVA which regulated the recognition of food labels in the pre-experimental design, the recognition of food labels in the post-experimental design showed the meaningful differences depending on the instruction style(motivation strategy and lecture type instruction). In addition, comprehensibility, practical use and educational necessity of food label details showed statistically meaningful differences. Home economics instruction using motivation(ARCS) strategy was more effective than lecture type instruction in improving students' recognition of food labeling. Fourth, as a result of ANCOVA which regulated the use of food labels in the pre-experimental stage, the use of food labels in the post-experimental stage showed meaningful differences between experimental and comparative groups depending on the instruction style. Therefore, home economics instruction in food labeling using motivation(ARCS) strategy was more effective than lecture type instruction in increasing students' use of food labels.

• Journal of Biosystems Engineering
• /
• v.39 no.2
• /
• pp.111-114
• /
• 2014

Purpose: This review aims to introduce aptamers and the methods of its development to improve the sensitivity and selectivity to target bacteria. In this review, we have highlighted current developments and directions in the pathogen detection based on aptamers. Background: Aptamers, the specific nucleic acid sequences, can bind to targets with high affinity and specificity. Some of researches on the use of aptamers for the detection of pathogen have been reported in recent years. Aptamers have more applicability than antibodies for the development of pathogen detection using biosensor; such as easy to synthesis and labeling, lack of immunogenicity, and a low cost of production. However, only few reports on the development and use of aptamers for the detection of pathogen have been published. Review: Aptamers specific to pathogen are obtained by whole-cell systematic evolution of ligands by exponential enrichment (SELEX) process. SELEX process is composed of screening random oligonucleotide bound with target cells, multiple separation and amplification of nucleic acids, final identification of the best sequences. For improving those affinity and selectivity to target bacteria, optimization of multiple separating process to remove unbounded oligonucleotides from aptamer candidates and sorting process by flow cytometry are required.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.154-157
• /
• 2022

In the existing Tensorflow CNN environment, the object detection method is a method of performing object labeling and detection by Tensorflow itself. However, with the advent of YOLO, the efficiency of image object detection has increased. As a result, more deep layers can be built than existing neural networks, and the image object recognition rate can be increased. Therefore, in this paper, the detection ability and speed were compared and analyzed by designing an object detection system based on Darknet and YOLO and performing multi-layer construction and learning based on the existing convolutional neural network. For this reason, in this paper, a neural network methodology that efficiently uses Darknet's learning is presented.