• Journal of Information Technology Services
• /
• v.21 no.2
• /
• pp.97-107
• /
• 2022

Recently, interest in smart factories is increasing. Investments to improve intelligence/automation are also being made continuously in manufacturing plants. Facility automation based on sensor data collection is now essential. In addition, we are operating our factories based on data generated in all areas of production, including production management, facility operation, and quality management, and an integrated standard information system. When producing LCD polarizer products, it is most important to link trace information between data generated by individual production processes. All systems involved in production must ensure that there is no data loss and data integrity is ensured. The large-capacity data collected from individual systems is composed of key values linked to each other. A real-time quality analysis processing system based on connected integrated system data is required. In this study, large-capacity data collection, storage, integration and loss prevention methods were presented for optimization of LCD polarizer production. The identification Risk model of inspection products can be added, and the applicable product model is designed to be continuously expanded. A quality inspection and analysis system that maximizes the yield rate was designed by using the final inspection image of the product using big data technology. In the case of products that are predefined as analysable products, it is designed to be verified with the big data knn analysis model, and individual analysis results are continuously applied to the actual production site to operate in a virtuous cycle structure. Production Optimization was performed by applying it to the currently produced LCD polarizer production line.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.27-33
• /
• 2021

With increasing interest in key areas of the 4th industrial revolution such as artificial intelligence, deep learning and big data, scientific approaches have developed in order to overcome the limitations of traditional decision-making methodologies. These scientific techniques are mainly used to predict the direction of financial products. In this study, the factors of apartment prices, which are of high social interest, were analyzed through SOM. For this analysis, we extracted the real prices of the apartments and selected a total of 16 input variables that would affect these prices. The data period was set from 1986 to 2021. As a result of examining the characteristics of the variables during the rising and faltering periods of the apartment prices, it was found that the statistical tendencies of the input variables of the rising and the faltering periods were clearly distinguishable. I hope this study will help us analyze the status of the real estate market and study future predictions through image learning.

• Journal of Broadcast Engineering
• /
• v.27 no.5
• /
• pp.718-727
• /
• 2022

The Korean domestic cartoon industry has grown explosively by 65% compared to the previous year. Then the market size is expected to exceed KRW 1 trillion. However, excessive work results in health deterioration. Moreover, this working environment makes the production of human resources insufficient, repeating a vicious cycle. Although some tasks require creation activity during cartoon production, there are still a lot of simple repetitive tasks. Therefore, this study aimed to develop a method for creating a character pose through human pose estimation (HPE). The HPE is to detect key points for each joint of a user. The primary role of the proposed method was to make each joint of the character match that of the human. The proposed method enabled us to create the pose of the two-dimensional cartoon character through the results. Furthermore, it was possible to save the static image for one character pose and the video for continuous character pose.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.92-98
• /
• 2022

Identifying defects in textiles is a key procedure for quality control. This study attempted to create a model that detects defects by analyzing the images of the fabrics. The models used in the study were deep learning-based VGGNet and ResNet, and the defect detection performance of the two models was compared and evaluated. The accuracy of the VGGNet and the ResNet model was 0.859 and 0.893, respectively, which showed the higher accuracy of the ResNet. In addition, the region of attention of the model was derived by using the Grad-CAM algorithm, an eXplainable Artificial Intelligence (XAI) technique, to find out the location of the region that the deep learning model recognized as a defect in the fabric image. As a result, it was confirmed that the region recognized by the deep learning model as a defect in the fabric was actually defective even with the naked eyes. The results of this study are expected to reduce the time and cost incurred in the fabric production process by utilizing deep learning-based artificial intelligence in the defect detection of the textile industry.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.474-476
• /
• 2021

Glioblastoma is the most common brain malignancies arising from glial cells. Early diagnosis and treatment plan establishment are important, and cancer is diagnosed mainly through T1CE imaging through injection of a contrast agent. However, the risk of injection of gadolinium-based contrast agents is increasing recently. Region segmentation that marks cancer regions in medical images plays a key role in CAD systems, and deep neural network models for synthesizing new images are also being studied. In this study, we propose a model that simultaneously learns the generation of T1CE images and segmentation of cancer regions. The performance of the proposed model is evaluated using similarity measurements including mean square error and peak signal-to-noise ratio, and shows average result values of 21 and 39 dB.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.51.1-51.1
• /
• 2019

At the commissioning phase of KVN from 2009 to 2013, single-dish survey and monitoring observations were performed toward about 1000 evolved stars and about 60 relatively strong SiO and H2O maser sources respectively. Based on these single-dish results and VLBI feasibility test observations at K/Q/W/D bands in 2014, KVN Key Science Project (KSP) has started from 2015 and will be completed in 2019 as KSP phase I. Here we present the overview of observational studies on evolved stars using KVN. In KSP phase I, we have focused on nine KSP sources which show a successful astrometrically registered maps of SiO and H2O masers using the source frequency phase referencing method. We aim at investigating the spatial structure and dynamical effect from 43/42/86/129 GHz SiO to 22 GHz H2O maser regions associated with a stellar pulsation and development of asymmetry in circumstellar envelopes. Using the combined network KaVA (KVN+Japanese VLBI network VERA), KaVA Large Program titled on "Expanded Study on Stellar Masers: ESTEMA Phase I" was performed from 2015 to 2016. Based on ESTEMA Phase I, EAVN Large Program titled on "EAVN Synthesis of Stellar Maser Animations: ESTEMA Phase II" was also performed from 2018. The ESTEMA II project aims to publish composite animations of circumstellar H2O and SiO masers, which taken from up to 6 long-period variable stars with a variety of the pulsation periods (333-1000 days). The animations will exhibit the three-dimensional kinematics of the maser gas clumps with complexity caused by stellar pulsation-driven shock waves and anisotropy of clump ejections from the stellar surface. Adding three EAVN telescopes (Tianma 65m, Nanshan 26m and NRO 45m telescopes) with KaVA always secures the high quality of the maser image frames through the monitoring program.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.61-62
• /
• 2023

There have been cases where the number of port workers has temporarily decreased due to COVID-19. To prevent the spread of COVID-19, a number of workers were quarantined, resulting in bottlenecks and waiting throughout the port operation process, increasing the congestion of terminals. As a result, problems such as a decrease in terminal operation efficiency occurred. However, it is understood that congestion centered on unloading equipment inside the terminal is not clearly calculated. Terminal congestion is thought to be a key factor directly related to the operational efficiency of the terminal. The congestion calculation method generally used in various fields measures congestion based on image-based data. Due to the nature of the loading and unloading equipment that moves according to the quantitative loading plan, it is unreasonable to apply the existing congestion calculation method. Therefore, this study presented a method of calculating terminal congestion using equipment waiting time and turnaround time, and verified the statistical significance of the congestion calculated using data from Terminal A of Busan Port.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.93-101
• /
• 2024

Accurate field crop classification is essential for various agricultural applications, yet existing methods face challenges due to diverse crop types and complex field conditions. This study aimed to address these issues by combining support vector machine (SVM) models with multi-seasonal unmanned aerial vehicle (UAV) images, texture information extracted from Gray Level Co-occurrence Matrix (GLCM), and RGB spectral data. Twelve high-resolution UAV image captures spanned March-October 2021, while field surveys on three dates provided ground truth data. We focused on data from August (-A), September (-S), and October (-O) images and trained four support vector classifier (SVC) models (SVC-A, SVC-S, SVC-O, SVC-AS) using visual bands and eight GLCM features. Farm maps provided by the Ministry of Agriculture, Food and Rural Affairs proved efficient for open-field crop identification and served as a reference for accuracy comparison. Our analysis showcased the significant impact of hyperparameter tuning (C and gamma) on SVM model performance, requiring careful optimization for each scenario. Importantly, we identified models exhibiting distinct high-accuracy zones, with SVC-O trained on October data achieving the highest overall and individual crop classification accuracy. This success likely stems from its ability to capture distinct texture information from mature crops.Incorporating GLCM features proved highly effective for all models,significantly boosting classification accuracy.Among these features, homogeneity, entropy, and correlation consistently demonstrated the most impactful contribution. However, balancing accuracy with computational efficiency and feature selection remains crucial for practical application. Performance analysis revealed that SVC-O achieved exceptional results in overall and individual crop classification, while soybeans and rice were consistently classified well by all models. Challenges were encountered with cabbage due to its early growth stage and low field cover density. The study demonstrates the potential of utilizing farm maps and GLCM features in conjunction with SVM models for accurate field crop classification. Careful parameter tuning and model selection based on specific scenarios are key for optimizing performance in real-world applications.

• Journal of radiological science and technology
• /
• v.46 no.5
• /
• pp.417-425
• /
• 2023

The glomerular filtration rate (GFR) has been the subject of much research as a key indicator for diagnosing, treating, and monitoring kidney function. The gamma camera method (Gates method) is simple and allows simultaneous acquisition of GFR and renal scintigraphy for each kidney, however its accuracy is inferior. This study aimed to investigate changes in GFR depending on how region of interest (ROI) are set up, which is one of many factors influencing accuracy. GFR was calculated by setting the ROI for each phase of the image acquisition time (Gates-1: 0~1 minutes, Gates-2: 1~3 minutes, Gates-3: 3~27 minutes), and statistical significance was verified based on probability value 0.05 through ANOVA analysis. While there was no statistically significant difference among results from Gates-1, 2, 3 (p=0.481>0.05), overall results from the Gates method tended to overestimate compared to those from the multiple blood sampling-dual exponential (MBSDE) method. When comparing averages between phases, results from Gates-2 were most similar to those from the MBSDE method. Moreover, paired t-test p-values between MBSDE method and phases were as follows Gates-1: 0.021 (p<0.05), Gates-2: 0.280 (p>0.05), and Gates-3: 0.164 (p>0.05) indicating that only Gates-1 had statistically significant differences compared with MBSDE method. Thus, setting ROI around 2~3 minutes is calculated can aid in accurately determining GFR when Gates Method.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.292-300
• /
• 2023

This paper proposes a patent discovery method and strategy for Generative AI-related patents by utilizing qualitative evaluation indicators established based on the core components of the technology. Currently, the evaluation of patent quality relies on quantitative indicators, but existing quantitative indicators cannot represent the characteristics of Generative AI technology, making it difficult to accurately evaluate. Therefore, there is a need for additional qualitative indicators that consider technical characteristics based on patent claims, which can reveal the actual strength of the patent. In this paper, we propose a new evaluation index considering the technical characteristics of Generative AI. Core patents were selected using the proposed evaluation index, and the appropriateness of the proposed index was verified through the existing quantitative evaluation method for the selected core patents.