• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.2
• /
• pp.1-8
• /
• 2024

The expansion joint is installed to offset the expansion of the superstructure and must ensure sufficient gap during its service life. In detailed guideline of safety inspection and precise safety diagnosis for bridge, damage due to lack or excessive gap is specified, but there are insufficient standards for determining the abnormal behavior of superstructures. In this study, a data-based maintenance was proposed by continuously monitoring the expansion-gap data of the same expansion joint. A total of 2,756 data were collected from 689 expansion joint, taking into account the effects of season. We have developed a method to evaluate changes in the expansion joint-gap that can analyze the thermal movement through four or more data at the same location, and classified the factors that affect the superstructure behavior and analyze the influence of each factor through deep learning and explainable artificial intelligence(AI). Abnormal behavior of the superstructure was classified into narrowing and functional failure through the expansion joint-gap evaluation graph. The influence factor analysis using deep learning and explainable AI is considered to be reliable because the results can be explained by the existing expansion gap calculation formula and bridge design.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.313-315
• /
• 2019

Although Korea's fashion market has negative growth, it has been growing again since 2018. This phenomenon means that people are becoming more interested in fashion. As interest in fashion grows, people visit various community sites for reference to find a suitable coordination for themselves. Most community sites, however, are manually categorizing each garment. Not only do these tasks take a lot of time, but they also make it difficult to search for multiple clothing at the same time. In other words, I can't choose what I want at the same time, and if I choose what I want, I have to look at what the model is wearing and refer to it. The problem with this may not help because the coordination in which the model provided is worn is more likely to be the one that the user does not want. In this paper, when the image is uploaded to improve the problem, the clothing is analyzed with AI analysis model and automatically classified and stored. Therefore, not only can you search for one clothes in the existing way, but you can also search for multiple clothes at the same time. The service is expected to allow more people to easily find and refer to the code for themselves.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.280-282
• /
• 2020

최근 사물인터넷(IoT), 자율주행과 같이 기계 간의 통신이 요구되는 서비스가 늘어감에 따라, 기계 임무 수행에 최적화된 데이터의 생성 및 압축에 대한 필요성이 증가하고 있다. 또한, 사물인터넷과 인공지능(AI)이 접목된 기술이 주목을 받으면서 딥러닝 모델에서 추출되는 특징(feature)을 디바이스에서 클라우드로 전송하는 방안에 관한 연구가 진행되고 있으며, 국제 표준화 기구인 MPEG에서는 '기계를 위한 부호화(Video Coding for Machine: VCM)'에 대한 표준 기술 개발을 진행 중이다. 딥러닝으로 특징을 추출하는 가장 대표적인 방법으로는 합성곱 신경망(Convolutional Neural Network: CNN)이 있으며, 오토인코더는 입력층과 출력층의 구조를 동일하게 하여 출력을 가능한 한 입력에 근사시키고 은닉층을 입력층보다 작게 구성하여 차원을 축소함으로써 데이터를 압축하는 딥러닝 기반 이미지 압축 방식이다. 이에 본 논문에서는 이러한 오토인코더의 성질을 이용하여 CNN 기반의 이미지 분류 네트워크의 합성곱 신경망으로부터 추출된 feature에 오토인코더를 적용하여 압축하는 방안을 제안한다.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.7
• /
• pp.299-306
• /
• 2022

In recent, sensors embedded in robots, equipment, and circuits have become common, and research for diagnosing device failures by learning measured sensor data is being actively conducted. This failure diagnosis study is divided into a classification model for predicting failure situations or types and a regression model for numerically predicting failure conditions. In the case of a classification model, it simply checks the presence or absence of a failure or defect (Class), whereas a regression model has a higher learning difficulty because it has to predict one value among countless numbers. So, the reason that regression modeling is more difficult is that there are many irregular situations in which it is difficult to determine one output from a similar input when predicting by matching input and output. Therefore, in this paper, we focus on input and output data with periodicity, analyze the input/output relationship, and secure regularity between input and output data by performing sliding window-based input data patterning. In order to apply the proposed method, in this study, current and temperature data with periodicity were collected from MMC(Modular Multilevel Converter) circuit system and learning was carried out using ANN. As a result of the experiment, it was confirmed that when a window of 2% or more of one cycle was applied, performance of 97% or more of fit could be secured.

• The Journal of Bigdata
• /
• v.7 no.1
• /
• pp.37-47
• /
• 2022

Autonomous delivery operation data is the key to driving a paradigm shift for last-mile delivery in the Corona era. To bridge the technological gap between domestic autonomous delivery robots and overseas technology-leading countries, large-scale data collection and verification that can be used for artificial intelligence training is required as the top priority. Therefore, overseas technology-leading countries are contributing to verification and technological development by opening AI training data in public data that anyone can use. In this paper, 326 objects were collected to trainn autonomous delivery robots, and artificial intelligence models such as Mask r-CNN and Yolo v3 were trained and verified. In addition, the two models were compared based on comparison and the elements required for future autonomous delivery robot research were considered.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.10
• /
• pp.137-148
• /
• 2024

The European honeybee (Apis mellifera) is an important pollinator threatened by colony collapse disorder (CCD), primarily due to infestation by the Varroa mite (Varroa destructor). Traditional detection methods are invasive and time-consuming, often causing additional stress to colonies. We propose a two-stage framework using the You Only Look Once version 8 (YOLOv8) model for non-destructive and rapid detection of Varroa mite infestation. The framework uses comb light images from inside the hives. In the first stage, a YOLOv8-n model detects bees and extracts individual bee images. In the second stage, a YOLOv8-cls model classifies the infestation status of each bee. Our object detection model achieved a mAP@0.5 of 0.701, and the classification model achieved an average accuracy of 91%. These results demonstrate the effectiveness of the framework as a non-destructive method for Varroa mite detection. Based on this research, we expect to provide beekeepers with an efficient tool for early detection and management of Varroa mite infestations, potentially reducing the incidence of CCD and supporting the sustainability of apiculture.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.2
• /
• pp.91-108
• /
• 2022

In most cases, optical images have been used as training data of DL (Deep Learning) models for object detection, recognition, identification, classification, semantic segmentation, and instance segmentation. However, properties of 3D objects in the real-world could not be fully explored with 2D images. One of the major sources of the 3D geospatial information is DSM (Digital Surface Model). In this matter, characteristic information derived from DSM would be effective to analyze 3D terrain features. Especially, man-made objects such as buildings having geometrically unique shape could be described by geometric elements that are obtained from 3D geospatial data. The background and motivation of this paper were drawn from concept of the intrinsic image that is involved in high-level visual information processing. This paper aims to extract buildings after classifying terrain features by training DL model with DSM-derived information including slope, aspect, and SRI (Shaded Relief Image). The experiments were carried out using DSM and label dataset provided by ISPRS (International Society for Photogrammetry and Remote Sensing) for CNN-based SegNet model. In particular, experiments focus on combining multi-source information to improve training performance and synergistic effect of the DL model. The results demonstrate that buildings were effectively classified and extracted by the proposed approach.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.434-436
• /
• 2021

의료산업은 진단 및 치료 위주의 기술개발이 진행되어왔다. 최근 의료 빅데이터를 기반으로 진단, 치료 및 재활뿐만 아니라 예방과 예후관리까지 지원하는 의료서비스에 대한 패러다임이 변화되고 있다. 특히, 여러 의료 중심의 플랫폼 기술 가운데 객관적인 진단지표를 가지고 있는 의료영상을 기반으로 인공지능 학습에 적용하여 진단 및 예측을 중심으로 한 플랫폼 개발이 진행되고 있다. 하지만, 인공지능 연구에는 많은 학습 데이터가 요구될 뿐만 아니라 학습에 적용하기 위해서는 데이터 특성에 따른 전처리 기술과 분류 작업에 많은 시간 소요되어 이와 같은 문제점을 해결할 수 있는 방법들이 요구되고 있다. 따라서, 본 논문은 인공지능 학습까지 적용하기 위한 의료영상 데이터에 대한 확장 모델을 개발하여 공통적인 조건에 따라 의료영상 데이터가 표준화되어 변환하며, 자동화 시스템 구조에 따라 데이터가 분류·저장되어 인공지능 학습까지 지원할 수 있는 플랫폼을 제안하고자 한다. 그리고 근감소증 학습데이터 관리 및 적용 결과를 통해 플랫폼의 수행성을 검증하였다. 향후 제안한 플랫폼을 통해 의료데이터에 대한 전처리, 분류, 관리까지 지원함으로써 CDM 확장 표준 의료데이터 플랫폼으로 활용 가능성을 보였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.507-507
• /
• 2023

지진해일은 많은 인명피해를 입힐 수 있는 위험한 자연재해이며, 예를 들어 각각 약 25만명과 약 2만명의 사상자가 발생하였던 2004년 수마트라 지진해일과 2011년 동일본 지진해일 등이 있다. 우리나라 동해안 또한 향후 지진 발생 가능성이 큰 지진공백역이 존재하여 안전한 지역으로 볼 수 없다. 지진해일 방재대책 수립과 관련된 연구는 지속적으로 이루어지고 있지만 지진해일의 발생빈도는 적고 완벽히 대응하는 것은 현실적으로 불가능하다. 따라서 본 연구에서는 지진해일 방재대책의 가장 기본적인 자료로 이용될 수 있는 지진해일 침수예상도를 효율적인 방법으로 제작하는 것을 연구했다. 현재 우리나라의 지진해일 최대 침수예상도는 과거 및 향후 발생가능한 지진해일의 경우에 대한 모든 범람구역이 고려된 보수적인 방법으로 제작되고 있다. 지진원의 위치와 각 매개변수의 특성에 따라 범람구역이 다양하게 나타날 수 있기 때문에 보수적인 최대 침수예상도는 과도한 범람구역이 고려될 수 있다. 따라서 본 연구에서는 보수적인 최대 침수예상도와 비교하여 AI기술과 로직트리 기법을 통해 더 정확한 최대 침수예상도를 제작하는 것을 목표로 한다. 연구방법은 1) 고려된 모든 지진해일 시나리오에 대한 수치해석 2) 입력자료인 지진해일 초기수면 변위 이미지 증강 3) CNN모델을 활용한 초기수면 변위 이미지 분류 4) 분류된 결과의 범람 구역으로 최대 침수예상도를 제작하였다. 향후 연구결과는 지진해일 재해정보도 제작 및 지진해일 침수예측 모델 개발에 활용될 수 있을 것이다.

• Journal of The Korean Association For Science Education
• /
• v.43 no.6
• /
• pp.573-582
• /
• 2023

This study aimed to explore the potential use of artificial intelligence in science education for gifted students by analyzing the structure of abstracts written by students at a gifted science academy and comparing the performance of various elements extracted using AI. The study involved an analysis of 263 graduation theses from S Science High School over five years (2017-2021), focusing on the frequency and types of background, objectives, methods, results, and discussions included in their abstracts. This was followed by an evaluation of their accuracy using AI classification methods with fine-tuning and prompts. The results revealed that the frequency of elements in the abstracts written by gifted students followed the order of objectives, methods, results, background, and discussions. However, only 57.4% of the abstracts contained all the essential elements, such as objectives, methods, and results. Among these elements, fine-tuned AI classification showed the highest accuracy, with background, objectives, and results demonstrating relatively high performance, while methods and discussions were often inaccurately classified. These findings suggest the need for a more effective use of AI, through providing a better distribution of elements or appropriate datasets for training. Educational implications of these findings were also discussed.