• Journal of Convergence for Information Technology
• /
• v.11 no.7
• /
• pp.14-20
• /
• 2021

Most accidents caused by road icing in winter lead to major accidents. Because it is difficult for the driver to detect the road icing in advance. In this work, we study how to accurately detect road traffic emerging risk using AutoML and CNN's ensemble model that use both structured and unstructured data. We train CNN-based road traffic emerging risk classification model using images that are unstructured data and AutoML-based road traffic emerging risk classification model using weather data that is structured data, respectively. After that the ensemble model is designed to complement the CNN-based classification model by inputting probability values derived from of each models. Through this, improves road traffic emerging risk classification performance and alerts drivers more accurately and quickly to enable safe driving.

• Journal of the Society of Disaster Information
• /
• v.18 no.2
• /
• pp.364-373
• /
• 2022

Purpose: The purpose of this paper is to develope smoke detection using AI model for detecting the initial fire in underground utility tunnels using CCTV Method: To improve detection performance of smoke which is high irregular, a deep learning model for fire detection was trained to optimize smoke detection. Also, several approaches such as dataset cleansing and gradient exploding release were applied to enhance model, and compared with results of those. Result: Results show the proposed approaches can improve the model performance, and the final model has good prediction capability according to several indexes such as mAP. However, the final model has low false negative but high false positive capacities. Conclusion: The present model can apply to smoke detection in underground utility tunnel, fixing the defect by linking between the model and the utility tunnel control system.

• Broadcasting and Media Magazine
• /
• v.28 no.1
• /
• pp.13-28
• /
• 2023

이미지 압축은 이미지 및 영상처리에서 주요한 역할을 하며, 자율주행, 클라우드, 영상 송출 등의 분야에서 빅데이터를 처리해야 하는 수요가 늘어남에 따라 지속적인 연구가 진행 중이다. 그 중심에는 딥러닝(deep learning)의 발전이 자리잡고 있으며, 심층 신경망(deep neural network)을 효과적으로 학습하는 알고리즘들을 적용한 논문들은 기존 압축 포맷인 JPEG, JPEG 2000, MPEG 등의 압축 성능을 뛰어넘는 결과를 보여 주고 있다. 이에 따라 JPEG AI는 딥러닝 기반 학습 이미지 압축의 표준을 제정하는 일을 진행 중이다. 본 기고에서는 JPEG AI가 표준화하고자 하는 기술과 JPEG AI에 제안한 압축 프레임워크들을 분석하고, 활용 사례들을 소개하여 JPEG AI 기반 학습 이미지 압축 모델의 동향에 대해 알아보고자 한다.

• Journal of the Korea Convergence Society
• /
• v.12 no.12
• /
• pp.83-91
• /
• 2021

For agricultural products, supply is irregular due to changes in meteorological conditions, and it has high price elasticity. For example, if the supply decreases by 10%, the price increases by 50%. Due to these fluctuations in the prices of agricultural products, the Korean government guarantees the safety of prices to producers through small merchants' auctions. However, when prices plummet due to overproduction, protection measures for producers are insufficient. Therefore, in this paper, we designed a business model that can be used in the electronic transaction system by predicting the price of agricultural products with an artificial intelligence algorithm. To this end, the trained model with the training pattern pairs and a predictive model was designed by applying ARIMA, SARIMA, RNN, and CNN. Finally, the agricultural product forecast price data was classified into short-term forecast and medium-term forecast and verified. As a result of verification, based on 2018 data, the actual price and predicted price showed an accuracy of 91.08%.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.1
• /
• pp.27-38
• /
• 2023

In this paper, we propose a conversational AI agent based on continual learning that can continuously learn and grow with new data over time. A continual learning-based conversational AI agent consists of three main components: Task manager, User attribute extraction, and Auto-growing knowledge graph. When a task manager finds new data during a conversation with a user, it creates a new task with previously learned knowledge. The user attribute extraction model extracts the user's characteristics from the new task, and the auto-growing knowledge graph continuously learns the new external knowledge. Unlike the existing conversational AI agents that learned based on a limited dataset, our proposed method enables conversations based on continuous user attribute learning and knowledge learning. A conversational AI agent with continual learning technology can respond personally as conversations with users accumulate. And it can respond to new knowledge continuously. This paper validate the possibility of our proposed method through experiments on performance changes in dialogue generation models over time.

• Smart Media Journal
• /
• v.12 no.8
• /
• pp.60-68
• /
• 2023

After COVID-19, hotel industry is rapidly experiencing changes in the business environment, and under the influence of the Fourth Industrial Revolution, hotel industry is striving to secure competitive advantages through differentiation, including the use of big data and the IoT in service provision, as well as the introduction of artificial intelligence(AI) robot services. This study analyzed the perceived value of AI robot services and their impact on usage attitudes and behavioral intentions of customers who have used hotels that have introduced AI robot services. The results of the study showed that the value of robot services perceived by customers who have used robot services in hotels is categorized into three dimensions: social, experiential, and functional, and all of them have a positive effect on usage attitudes, with social, functional, and experiential values having a positive effect on usage attitudes in that order. Attitude toward use was also analyzed to have a positive effect on behavioral intention, which is consistent with the value-attitude-behavior model. Therefore, it is necessary for hotels to improve the satisfaction of hotel guests through diversified services of AI robot service.

• Journal of The Korean Association For Science Education
• /
• v.44 no.2
• /
• pp.141-153
• /
• 2024

The 2022 revised science curriculum aims to develop the ability to solve scientific problems arising in daily life and society based on convergent thinking stimulated through participation in research activities using artificial intelligence (AI). Therefore, we developed a science-AI convergence education program that combines the science curriculum with artificial intelligence and employed it in convergence classes for high school students. The aim of the science-AI convergence class was for students to qualitatively understand the movement of a damped pendulum and build an AI model to predict the position of the pendulum using the block coding platform KNIME. Individual in-depth interviews were conducted to understand and interpret the learners' experiences. Based on Giorgi's phenomenological research methodology, we described the learners' learning processes and changes, challenges and limitations of the class. The students collected data and built the AI model. They expected to be able to predict the surrounding phenomena based on their experimental results and perceived the convergence class positively. On the other hand, they still perceived an with the unfamiliarity of platform, difficulty in understanding the principle of AI, and limitations of the teaching method that they had to follow, as well as limitations of the course content. Based on this, we discussed the strengths and limitations of the science-AI convergence class and made suggestions for science-AI convergence education. This study is expected to provide implications for developing science-AI convergence curricula and implementing them in the field.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.136-138
• /
• 2022

스포츠 방송/미디어 데이터에서 특정 이벤트 시점을 효율적으로 검출하는 방법은 정보 검색이나 하이라이트, 요약 등을 위해 중요한 기술이다. 이 논문에서는, 야구 중계 방송 데이터에서 투구에 대한 타격 및 포구 이벤트 시점을 강인하게 검출하는 방법으로, 음향 및 영상 정보를 융합하는 방법에 대해 제안한다. 음향 정보에 기반한 이벤트 검출 방법은 계산이 용이하고 정확도가 높은 반면, 영상 정보의 도움 없이는 모호성을 해결하기 힘든 경우가 많이 발생한다. 특히 야구 중계 데이터의 경우, 투수의 투구 시점에 대한 영상 정보를 활용하여 타격 및 포구 이벤트 검출의 정확도를 보다 향상시킬 수 있다. 이 논문에서는 음향 기반의 딥러닝 이벤트 시점 검출 모델과 영상 기반의 보정 방법을 제안하고, 실제 KBO 야구 중계 방송 데이터에 적용한 사례와 실험 결과에 대해 기술한다.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.571-572
• /
• 2024

본 연구에서는 딥러닝 기술을 활용하여 LA 갈비 이미지에서 살코기 부분을 검출하는 방안을 제안한다. 검출에는 이미지 픽셀을 미리 정의된 클래스로 분류하는 딥러닝 기술인 세그먼테이션을 활용한다. 세그먼테이션은 이미지 내 객체를 정교하게 탐지할 수 있는 특성을 가진다. 각 이미지에는 한 개의 손질 전 LA 갈비 이미지를 포함한다. 학습된 모델은 높은 정확도와 정밀도로 살코기 부분을 검출하였으며, 이는 육류 가공 과정에 추후 딥러닝 기술 활용이 유용할 수 있음을 시사한다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.452-454
• /
• 2021

Currently, the security control market is operating based on AI technology. The reason for using AI is to detect large amounts of logs and big data between security equipment, and to alleviate time and human problems. However, problems are still occurring in the application of AI. The security control market is responding to many problems other than the problems introduced in this paper, and this paper attempts to deal with five problems. We would like to consider problems that arise in applying AI technology to security control environments such as 'AI model selection', 'AI standardization problem', 'Big data accuracy', 'Security Control Big Data Accuracy and AI Reliability', 'responsibility material problem', and 'lack of AI validity.'