• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.39-39
• /
• 2021

기상재해를 극소화하기 위해서는 그 원인이 되는 기상현상의 규모와 거동을 명확히 감시하고 분석하여 신뢰성 있는 예측정보가 제공되어야 한다. 최근 위험기상 발생빈도가 증가하여 초단기 및 위험기상 예보의 정확도 향상을 위한 고품질 레이더 정보 활용 연구가 활발하게 진행되고 있다. 레이더는 전자파를 이용하여 강우의 양과 분포, 이동특성을 관측하는 장비로써 우리나라는 초단기적 위험기상 대응능력 향상을 추진하기 위한 목적으로 첨단 성능의 이중편파레이더 관측망을 구축하고 있다. 국내 기상관측용 레이더는 기상예보(기상청), 홍수예보(환경부), 군 작전 기상지원(국방부) 등으로 각 기관이 개별적으로 설치운영 하고 있다. 본 연구에서는 관계부처에서 운영하고 있는 레이더의 합성장을 이용하여 강수장의 상관성을 기반으로 이류(advection) 특성을 도출하였다. 정확도 있는 이류특성을 도출하기 위하여 시간해상도는 10분을 적용하였으며 가우시안 필터링 기법을 적용하여 강수장 상관분석을 수행하였다. 호우와 태풍을 대상으로 강수장의 이류패턴을 추출하여 강수장의 이동방향 및 속도를 고려한 강수량 예측기법의 적용성을 평가하였다. 본 연구 결과는 격자형 강수예측정보를 제공하여 AI 홍수예보 및 수치예보 모델의 초기조건 입력 등에 활용되어 기후변동성에 따른 대국민 안전 실현을 확보하는데 기후변화 대응전략의 핵심기술로 활용될 수 있을 것으로 판단된다. 덧붙어, 4차 산업혁명에 따른 수문기상 빅 데이터(big data) 통합 플랫폼을 구축하여 고해상도 홍수대응 기술 및 GIS 및 모바일 시스템을 연계한 실시간 기후재해 예·경보가 가능할 것으로 사료된다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.474-477
• /
• 2021

Recently, in order to prevent the proliferation of COVID-19, which began in earnest in 2020, an increasing number of places have been measuring the temperature and required to wear a mask. However, as wearing a mask and checking the temperature are typically measured directly by a person or by a single individual positioned in front of the machine, standards may vary based on the person's manual measurement method, wasting workforce. While standing in front of a device often measures the maximum temperature of the face, the standard of fever is also unclear. Both approaches can create bottleneck situations when checking large numbers of people. Furthermore, it is unable to conduct periodic measurements and tracking because the measuring machines are generally put only at the entrance. Thus, this study suggests a method for preventing the spread of infectious diseases by automatically identifying and displaying unmasked people and those with fever in real-time using a general camera, a thermal imaging camera, and an artificial intelligence algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.209-216
• /
• 2023

In this paper, artificial intelligence agents ChatGPT and Stable Diffusion were used to explore the possibility of educational use by implementing a program to recommend picture materials for teaching and learning according to the class topic entered by teachers. The average time spent recommending all picture materials is about 6 minutes. In general, pictures related to keywords were recommended, and the letters in the recommended pictures could only know the intention to represent the letters, and the letters could not be recognized and the meaning could not be known. However, further research seems to be needed on the fact that the type or content of the recommended picture depends entirely on the response of ChatGPT and that it is not possible to accurately recommend the picture for all keywords. In addition, it was concluded that it is true that the recommended picture is related to the keyword, but the evaluation of whether it has educational value is the subject of discussion that should be left to the judgment of human teachers.

• Journal of Software Assessment and Valuation
• /
• v.14 no.2
• /
• pp.1-9
• /
• 2018

The proportion of users and companies using open source continues to grow. The size of open source software market is growing rapidly not only in foreign countries but also in Korea. However, compared to the continuous development of open source software, there is little research on open source software subject classification, and the classification system of software is not specified either. At present, the user uses a method of directly inputting or tagging the subject, and there is a misclassification and hassle as a result. Research on open source software classification can also be used as a basis for open source software evaluation, recommendation, and filtering. Therefore, in this study, we propose a method to classify open source software by using machine learning model and propose performance comparison by machine learning model.

• Convergence Security Journal
• /
• v.23 no.3
• /
• pp.29-36
• /
• 2023

With the recent surge in exposure to fake advertising phishing sites in search engines, the damage caused by poor search quality and personal information leakage is increasing. In particular, the seriousness of the problem is worsening faster as the possibility of automating the creation of advertising phishing sites through tools such as ChatGPT increases. In this paper, the source code of fake advertising phishing sites was statically analyzed to derive structural commonalities, and among them, a detection crawler that filters sites step by step based on foreign domains and redirection was developed to confirm that fake advertising posts were finally detected. In addition, we demonstrate the need for new guide lines by verifying that the redirection page of fake advertising sites is divided into three types and returns different sites according to each situation. Furthermore, we propose new detection guidelines for fake advertising phishing sites that cannot be detected by existing detection methods.

• Information Systems Review
• /
• v.23 no.4
• /
• pp.69-86
• /
• 2021

As artificial intelligence technology advances, personalized recommendation systems using big data have attracted huge attention. In the case of beauty products, product preferences are clearly divided depending on customers' skin types and sensitivity along with individual tastes, so it is necessary to provide customized recommendation services based on accumulated customer data. Therefore, by employing deep learning methods, this study proposes a neural network-based recommendation model utilizing both product search history and context information such as gender, skin types and skin worries of customers. The results show that our model with context information outperforms collaborative filtering-based recommender system models using customer search history.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.2
• /
• pp.133-141
• /
• 2024

Considering fluid-structure-soil interactions, a finite-element model for a liquid-storage tank is presented and the nonlinear earthquake response analysis is formulated. The tank structure is modeled considering shell elements with geometric and material nonlinearities. The fluid is represented by acoustic elements and combined with the structure using interface elements. To consider the soil-structure interactions, the near- and far-field regions of soil are modeled with solid elements and perfectly matched discrete layers, respectively. This approach is applied to the seismic fragility analysis of a 200,000 kL liquid-storage tank. The fragility curve is observed to be influenced by the amplification and filtering of rock outcrop motions at the site when the soil-structure interactions are considered.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.2
• /
• pp.143-149
• /
• 2024

Segmentation of material phases through image analysis is essential for analyzing the microstructure of materials. Micro-CT images exhibit variations in grayscale values depending on the phases constituting the material. Phase segmentation is generally achieved by comparing the grayscale values in the images. In the case of waste concrete used as a recycled aggregate, it is challenging to distinguish between hydrated cement paste and natural aggregates, as these components exhibit similar grayscale values in micro-CT images. In this study, we propose a method for automatically separating the aggregates in concrete, in micro-CT images. Utilizing the Unet-VGG16 deep-learning network, we introduce a technique for segmenting the 2D aggregate images and stacking them to obtain 3D aggregate images. Image filtering is employed to separate aggregate particles from the selected 3D aggregate images. The performance of aggregate segmentation is validated through accuracy, precision, recall, and F1-score assessments.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.101-108
• /
• 2024

AI-OCR (Artificial Intelligence Optical Character Recognition) combines OCR technology with Artificial Intelligence to overcome limitations that required human intervention. To enhance the performance of AI-OCR, training on diverse data sets is essential. However, the recognition rate declines when image colors have similar brightness levels. To solve this issue, this study employs Homomorphic filtering as a preprocessing step to clearly differentiate color levels, thereby increasing text recognition rates. While Homomorphic filtering is ideal for text extraction because of its ability to adjust the high and low frequency components of an image separately using a gamma value, it has the downside of requiring manual adjustments to the gamma value. This research proposes a range for gamma threshold values based on tests involving image contrast, brightness, and entropy. Experimental results using the proposed range of gamma values in Homomorphic filtering suggest a high likelihood for effective AI-OCR performance.

• Information Systems Review
• /
• v.21 no.4
• /
• pp.31-45
• /
• 2019

Despite the expansion of the market of performing arts and culture, small and medium size theaters are still experiencing difficulties due to poor accessibility of information by consumers. This study proposes a machine learning based genre recommendation system as an alternative to enhance the marketing capability of small and medium sized theaters. We developed five recommendation systems that recommend three genres per customer using customer master DB and transaction history DB of domestic venues. We propose an optimal recommendation system by comparing performances of recommendation system. As a result, the recommendation system based on the ensemble model showed better performance than the single predictive model. This study applied the personalized recommendation technique which was scarce in the field of performing arts and culture, and suggests that it is worthy enough to use it in the field of performing arts and culture.