• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.3
• /
• pp.95-105
• /
• 2019

In recent years, image processing techniques for detecting road surface damaged spot have been actively researched. Especially, it is mainly used to acquire images through a smart phone or a black box that can be mounted in a vehicle and recognize the road surface damaged region in the image using several algorithms. In addition, in conjunction with the GPS module, the exact damaged location can be obtained. The most important technology is image processing algorithm. Recently, algorithms based on artificial intelligence have been attracting attention as research topics. In this paper, we will also discuss artificial intelligence image processing algorithms. Among them, an object detection method based on an region-based convolution neural networks method is used. To improve the recognition performance of road surface damage objects, 600 road surface damaged images and 1500 general road driving images are added to the learning database. Also, supervised learning using background object recognition method is performed to reduce false alarm and missing rate in road surface damage detection. As a result, we introduce a new method that improves the recognition performance of the algorithm to 8.66% based on average value of mAP through the same test database.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.1970-1974
• /
• 2006

Tank 모형은 유역을 임의의 저류탱크로 가정하여, 유출공의 높이를 초과하는 저류고를 방출함으로써 유출량을 모의한다. 유출분석의 목적에 따라 직렬 3단 혹은 4단의 탱크로 구성하여 적용하는 것이 일반적인데, 국내의 일 단위 장기유출분석 연구에서는 직렬 4단 Tank 모형이 널리 활용되고 있다. 이러한 Tank 모형은 유역의 강우-유출관계를 모의하는 과정에 black box적인 특성을 지니고 있다. 그러나 각 저류탱크와 관련된 매개변수를 최적화하기 위해서는 매개변수들의 물리적인 의미를 이해하여야 한다. 이런 점을 고려하여 일본의 Sugawara는 경험적으로 매개변수들이 결정되는 범위를 제시한 바 있다. 그러나 기저유출을 모의하는 Tank 모형의 최하단 탱크에서 이러한 매개변수 범위에서는 적합한 값을 갖으나 장기적인 모의시에 저류고 및 유출고가 계속 증가하여 물리적인 유출특성을 반영하지 못하는 문제점이 나타났다. 따라서 본 연구에서는 이러한 문제점의 원인을 분석하기 위해 장기간의 자료가 구축된 소양강댐을 적용 유역으로 선정하여 최하단탱크의 유출공계수의 변화에 따른 유출량과 저류고의 변화를 살펴보았다. 분석결과 매개변수가 $0.0001{\sim}0.001$의 범위에서 장기간의 지속적인 저류고와 유출량의 증가가 나타났다. 그리고 유출공계수가 증가함에 따라 최대저류고는 감소하고, 저류고가 증가하는 지속기간이 짧아지는 것으로 나타났다. 그러나 통계치 변화분석에서는 상관계수, 평균제곱근오차, 모형효율성계수에서 거의 변화가 없는 것으로 나타났으며, 유출용적오차에서도 최대 약 6% 정도 유출용적이 변화하는 것으로 나타났다.mber)과 동일한 위치의 수온자료를 기초로 회귀분석을 실시함으로써 수온추출 알고리즘을 도출하여, 분석데이터의 신뢰도를 검증하였으며, 수온, 클로로필, 투명도 등을 위성원격탐사 자료와 GIS를 이용하여 공간분석을 실시하고, 공간분포도를 작성함으로써 대상해역의 해양환경을 파악하였다. 본 연구결과, 분석된 위성자료가 현장조사에 의한 검증이 이루어지지 않을 경우, 영상자료분석을 통한 표층수온 추출은 대기 중의 수증기와 에어로졸에 의한 계산치의 오차가 반영되기 때문에 실측치 보다 낮게 평가 될 수 있으므로, 반드시 이에 대한 검증이 필요함을 알 수 있었다. 현지관측에 비해 막대한 비용과 시간을 절약할 수 있는 위성영상해석방법을 이용한 방법은 해양수질파악이 가능할 것으로 판단되며, GIS를 이용하여 다양하고 복잡한 자료를 데이터베이스화함으로써 가시화하고, 이를 기초로 공간분석을 실시함으로써 환경요소별 공간분포에 대한 파악을 통해 수치모형실험을 이용한 각종 환경영향의 평가 및 예측을 위한 기초자료로 이용이 가능할 것으로 사료된다.염총량관리 기본계획 시 구축된 모형 매개변수를 바탕으로 분석을 수행하였다. 일차오차분석을 이용하여 수리매개변수와 수질매개변수의 수질항목별 상대적 기여도를 파악해 본 결과, 수리매개변수는 DO, BOD, 유기질소, 유기인 모든 항목에 일정 정도의 상대적 기여도를 가지고 있는 것을 알 수 있었다. 이로부터 수질 모형의 적용 시 수리 매개변수 또한 수질 매개변수의 추정 시와 같이 보다 세심한 주의를 기울여 추정할 필요가 있을 것으로 판단된다.변화와 기흉 발생과의 인과관계를 확인하고 좀 더 구체화하기 위한 연구가 필요할 것이다.게 이

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.4
• /
• pp.580-586
• /
• 2021

The CAT methodology is a numerical analysis technique using CAE. Recently, a methodology of applying artificial intelligence techniques to a simulation has been studied. A previous study compared the deformation results according to the injection molding process using a machine learning technique. Although MLP has excellent prediction performance, it lacks an explanation of the decision process and is like a black box. In this study, data was generated using Autodesk Moldflow 2018, an injection molding analysis software. Several Machine Learning Algorithms models were developed using RapidMiner version 9.5, a machine learning platform software, and the root mean square error was compared. The decision-tree showed better prediction performance than other machine learning techniques with the RMSE values. The classification criterion can be increased according to the Maximal Depth that determines the size of the Decision-tree, but the complexity also increases. The simulation showed that by selecting an intermediate value that satisfies the constraint based on the changed position, there was 7.7% improvement compared to the previous simulation.

• Korean Chemical Engineering Research
• /
• v.61 no.4
• /
• pp.542-549
• /
• 2023

Since the growing interest in surrogate modeling, there has been continuous research aimed at simulating nonlinear chemical processes using data-driven machine learning. However, the opaque nature of machine learning models, which limits their interpretability, poses a challenge for their practical application in industry. Therefore, this study aims to analyze chemical processes using Explainable Artificial Intelligence (XAI), a concept that improves interpretability while ensuring model accuracy. While conventional sensitivity analysis of chemical processes has been limited to calculating and ranking the sensitivity indices of variables, we propose a methodology that utilizes XAI to not only perform global and local sensitivity analysis, but also examine the interactions among variables to gain physical insights from the data. For the ammonia synthesis process, which is the target process of the case study, we set the temperature of the preheater leading to the first reactor and the split ratio of the cold shot to the three reactors as process variables. By integrating Matlab and Aspen Plus, we obtained data on ammonia production and the maximum temperatures of the three reactors while systematically varying the process variables. We then trained tree-based models and performed sensitivity analysis using the SHAP technique, one of the XAI methods, on the most accurate model. The global sensitivity analysis showed that the preheater temperature had the greatest effect, and the local sensitivity analysis provided insights for defining the ranges of process variables to improve productivity and prevent overheating. By constructing alternative models for chemical processes and using XAI for sensitivity analysis, this work contributes to providing both quantitative and qualitative feedback for process optimization.

• Journal of KIISE:Software and Applications
• /
• v.31 no.8
• /
• pp.1027-1042
• /
• 2004

A Software Component is a module that is reused among a lot of projects, systems, and companies rather than a single application. Components can be reused in various systems if they provide not only the common functionalities required in many applications but also the diverse aspects to be customized for being suitable for customers' demands. From the development phase, components should be designed and developed considering the variable aspects they have for convenient customization. Easily customized components can be frequently reused in lots of applications. In the literature, there are some modeling and customizing techniques. But they suggested only conceptual or basic methods based on Object-Oriented. And the practical instructions for reusing component were not provided sufficiently. Moreover, there are few techniques that consider the proper variability types components have. Thus, those techniques are not appropriate for applying to black box component completely developed and released. In this paper, we classify variabilities that components have in functional aspect into two categories. The one is logic variability, and the other is workflow variability. For each classified variability, we propose the three kind of modeling techniques, which are selection, plug in and externalization. Also detailed instructions for practical design and application are provided.

• The Journal of the Korea Contents Association
• /
• v.20 no.8
• /
• pp.653-665
• /
• 2020

As the use of recommended agents has become more active, the "Explain Facilities" is drawing attention as a way to solve the black-box problem that could not explain internal logic to users. This study wants to look at how the description Method and Detail affects to user experience. The Explanation method was divided into 'why the agent did a particular action' and 'why not do a particular action' and the detail condition were divided into 'high or low'. Studies have found that 'why method' have a positive effect on users' transparency, trust, satisfaction, and behavioral intention to use, and 'high detail condition' higher the user' Psychological reactance. In addition, it was found that the explanation methods and detail influenced the 'Explanation' perception through interaction and tended to affect satisfaction and intention to adopt recommendation. This study suggested that careful attention is needed to determine the method and detail of the Explanation facilities in the context of the recommended agent, based on the research findings that it affects the user experience through the interaction of the method and detail.

• Journal of Science and Technology Studies
• /
• v.13 no.1
• /
• pp.145-180
• /
• 2013

STS studies into science documentaries for television are rare, and they do not even reflect the constructivist fruits of STS. STS have been calling for the need of analyzing 'science-in-the-making' in order to understand science more deeply. Similarly, our starting point is the assumption that science documentary can be better understood when we look into its making process. Under this assumption, we adopted the method of participant observation in analyzing 'documentary practice', trying to open the 'black box' called 'science documentary'. We have here focused on the documentary named "Light" made by a scientific documentary team of EBS, who made "Culture and Mathematics" and "Life". Each of us worked as a main consultant and an assistant staff in making "Light". We will address two main points in this study. First, based on our participant observation and interviews, we will show that the members of documentary making team are thinking about 'science' in distinctive ways. The team tended to emphasize visualization, knowledge linked to people's everyday life, and the distinctive characters of scientists who appear in the documentary. Second, by looking closely into the interaction between the team members and the consultant in the process of completing the script of the documentary, it was possible to understand how the contents of the documentary was constructed more accurately. In the making process, consultant's idea was not simply accepted by the making team, but there were conflicts and compromises. By showing this, we will be able to bring up a reflexive question about the role of consultant in the process of making a science documentary.

• Journal of KIISE:Software and Applications
• /
• v.37 no.12
• /
• pp.892-908
• /
• 2010

Services in SOA are typically perceived as black-box to service consumers, and can be dynamically evolved at runtime, and run on a number of unknown and heterogeneous environments. Because of these characteristics of the services, effective and efficient monitoring of various aspects on services is an essential functionality for autonomous management of service. But the problem with or limitation in conventional or existing approaches is, that they focus on services themselves, ignoring the effects by business processes. Consequently, there is a room for service monitoring which provides more useful information of business level by acquisition of only external monitoring data that depend on specific BPEL engine and middleware. Moreover, there is a strong demand to present effective methods to reduce monitoring overhead which can degrade quality of services. EDA can cope with such limitations in SOA by collecting and analyzing events efficiently. In this paper, we first describe EDA benefits in service monitoring, and classify monitorring target, and present an appropriate monitoring method for each monitoring target. Also to provide the applicability of our approach, an event meta-model is defined, and event processing model and architecture based on the meta-model are proposed. And, with the proposed architecture and method, we implement a prototype of an event-driven dynamic monitoring framework which can collect and process internal and external data at runtime. Finally, we present the result of a case study to demonstrate the effectiveness and applicability of the proposed approach.

• Cartoon and Animation Studies
• /
• s.22
• /
• pp.17-30
• /
• 2011

This paper analyzes Blood: the Last Vampire (2009) in comparison with the original Blood: the Last Vampire (2000). This digital movie, released in 2009, faithfully re-presents the visual of the original animation using more budget and more developed technology. Though these two movies share very similar visual style, they have slightly different modes of expression that can be easily detected in their opening scenes. The original Blood chooses to tell its story in more indirect way. Some shots are intentionally vague and forcefully separated by black credit screen, leaving audience guessing exactly what happened. The remake version of Blood provides faster, more diverse, and more sophisticated images. It also uses invisible and slick editing, leaving no room to guess. These differences comes mainly from the difference of media. However, since the differences between two movies lie mainly in their narratives, we can ascribe the box office slump of the digital movie version of Blood: the Last Vampire to its narrative flaws. The heroin's inner conflicts were not convincing enough, supporting roles were not probable enough, and the whole story was not unique enough. As a result, the new Blood: the Last Vampire became another bad example that shows the importance of narrative in films.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.6
• /
• pp.43-55
• /
• 2021

Recently, as the domestic job problem has become serious, government ministries are investing a lot of budgets to encourage startups by prospective entrepreneurs. What is important to the success of startups is the recognition of various startup opportunities before starting a startup, and the experience through trial. However, in reality, prospective entrepreneurs recognize and seek various startup opportunities through support such as startup education and initial commercialization funds through various government supported projects, but it is difficult to actually start a business. Previous studies have revealed that the recognition of entrepreneurial opportunities affects entrepreneurial intentions by various variables such as gender, but research is insufficient on what kind of black box exists between the recognition of entrepreneurial opportunities and entrepreneurial intentions. The purpose of this study is to emphasize the importance of positive psychological capital as a major method for improving the entrepreneurial intention, and to analyze exploratorily whether positive psychological capital plays a mediating role between the recognition of entrepreneurial opportunities and the entrepreneurial intention. As a result of the study, it was confirmed that the recognition of startup opportunities affects the intention to start a business, and positive psychological capital has a mediating effect between the recognition of the startup opportunity and the intention to start a business. This means that positive psychological capital is important in order to lead to actual entrepreneurial intentions after recognizing a startup opportunity. Therefore, in order to revitalize the startups of prospective entrepreneurs in the current startup ecosystem, it is necessary to prepare a plan to improve the recognition of startup opportunities and positive psychological capital.