• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.89-89
• /
• 2021

부영양화된 호수나 유속이 느린 하천에서 발생하는 녹조의 과도한 발생은 하천 생태계 훼손, 동식물의 건강, 담수의 오염 등 환경 사회 경제적으로 큰 피해를 준다. 현재 수질 측정망은 정해진 지점에서 Chlorophyll-a(Chl-a), Phycocyanin(PC)을 대표농도로 산정하고 조류경보에 활용하고 있으나, 일주일에 한번씩 샘플링을 통해 Chl-a 및 PC를 측정하여 시공간적인 신뢰성의 문제가 제기될 수 있다. 본 연구에서는 기존 점단위 조류 모니터링의 한계점을 개선하기 위해 초분광영상 자료를 머신러닝 기법에 적용하여 Chl-a 및 PC 산정 알고리즘을 개발하였다. 이를 위해 Chl-a와 PC의 최대 흡수, 반사 파장대, 주요 물 흡수 파장대 자료를 조합하여 9개의 파장비를 구축하였으며, 기존 연구에서 활용한 머신러닝 기법인 Partial Least Square, Random Forest, Gradient Boosting, Support Vector Machine, K-Nearest Neighbor, Artificial Neural Network를 검토하여 최적 모델을 선정하였다. 학습된 머신러닝의 성능을 R², NSE, RMSE 목적함수를 이용해 평가하였으며, 그 결과 ANN이 각각 PC 0.801, 0.755, 11.774 mg/m³, Chl-a 0.733, 0.622, 8.736 mg/m³로 가장 우수한 성능을 보였다. 최적화 된 ANN 모델을 백제보 상류 2016-2017년 항공 초분광영상에 적용하여 시공간에 따른 조류 분포변화를 평가하고자 한다.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.719-722
• /
• 2024

뇌 연령은 신경퇴행성 질환과 인지 저하를 예측하는 중요한 바이오마커로 주목받고 있으며, 이를 통해 개인의 뇌 건강 상태를 보다 정밀하게 확인할 수 있다. 특히, 회백질과 백질은 뇌 구조와 기능을 평가하는 데 핵심적인 역할을 하며, 뇌 구조적 변화를 분석함으로써 뇌 연령 예측의 정확도를 높일 수 있다. 또한, 특정 데이터셋만 활용될 경우 일반화된 성능을 기대하기 어려워 뇌 연령 예측에 다양한 데이터셋을 활용한 연구가 필요하다. 따라서, 본 연구에서는 다중 모달 MRI 데이터를 결합한 3D CNN 기반 뇌 연령 예측 모델을 제안한다. 제안된 모델은 회백질과 백질의 특징을 전처리된 T1 이미지에 결합하여 더욱 풍부한 뇌 구조 정보를 학습할 수 있도록 설계하여, 뇌 연령 예측의 정확성을 향상시켰다. 실험 결과 회백질과 백질 정보를 추가로 활용한 모델이 T1 이미지만을 사용한 기존 CNN 및 ResNet 모델보다 MAE(Mean Absolute Error) 평가지표에서 더 우수한 성능을 보였으며, 이를 통해 회백질과 백질 정보가 뇌 연령 예측에 중요한 기여를 한다는 사실을 확인하였다.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.47 no.6
• /
• pp.45-52
• /
• 2019

As international criteria and standards are required in the fields of design and construction, landscape performance must also be considered not only for the value of the landscape but also for providing quality assurance and sustainability. Given the lack of research on landscape performance, the present research was purposed to analyze the importance of potential assessment categories and items using an analytical network process. A list of assessment items, which is composed of 20 items and 6 categories, was derived through a literature review and a preliminary survey of 11 landscape professionals. An ANP model was established and a survey was conducted among 30 landscape practitioners to determine the weight of priorities considering the criteria. The results of ANP showed that the categories of site selection, preservation and health, and convenience had high priorities while materials had the lowest importance score. For the assessment items, a monitoring plan was the highest importance, followed by cultural/ historic preservation, management cost reduction, and natural ground areas. Despite the difficulties in quantifying landscape achievements, most respondents agreed that there needs to be an evaluation system for landscape performance in order to assure the quality and sustainability of landscape development. More research and discussion are needed to develop an assessment system for landscape performance that is applicable to Korean context.

• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.1
• /
• pp.47-62
• /
• 2012

To propagate clinical disease management service, there should be built a ecosystem where service developers, service providers, device suppliers closely cooperate for u-Health platform. However, most u-Health platform is difficult to build an effective ecosystem due to the lack of secure and effective PHR(Personal Health Record) management, the lack of personalized and intelligent service, difficulties of N-screen service. To solve these problems we suggest the CDMP(Chronic Disease Management Platform) architecture. The CDMP is a software platform that provides the core functions to develop the chronic disease management services and performs a hub function for the link and integration rbetween various services and systems. CDMP is SOA based platform that enables a provision of reusability, expansibility and it provides open API where everybody can share information, contents and services easily. CDMP supports the multi platform system foN-screen service and the self management functions via SNS. In this paper, we design and implement the CDMP including PHR service based on hybrid data model for privacy preservation. Experiment results prove the effectiveness of hybrid model-based PHR service.

• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.5
• /
• pp.15-24
• /
• 2014

In 2005, there was a partial revision of the Act on the Promotion of Education for the Handicapped people, so that students with health impairments would be able to receive special education support. Since the amendment of the bill, to support them classified into weak people in education, education support systems have been proposed and settled so that they may receive the support for free. According to the pre-study, after the amendment of this bill, there has been many studies on the form of educational service to support them, and recently, there have been a lot of researches to investigate their satisfaction with the current services and draw its problems. And yet these studies have been carried out by the preceeding researchers at the drawing of problems, but have a limitation to present fundamental countermeasures to the problems. Therefore, this study attempted to understand the meaning of health impairment through the pre-study and investigate the forms of the services currently supporting them and analyze the problem of each service. In addition, to solve the identified problems, a new support system was proposed. In order to confirm the performance of the system, we design the user satisfaction survey composed of a Likert 5-point scale per each question, and to make the task, comparing stories and clapping for increasing quality of their subjective evaluation about the image and voice transmission when the user uses it. As a result, in the overall evaluation of the robot system, the average score of each question was recorded to 4.31 points, and through the two tasks, it was found that there were effective data transmission of image and voice.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.2
• /
• pp.210-220
• /
• 2013

Recently, as we move toward a society with an increasingly aged population, wearable U-health devices in various shapes with smart wear have been developed in order to conveniently measure health variables without using hands in daily life or at home. However, the problem is that only supply of the wearable U-health devices is focused and its applicable devices are studied and developed, which has resulted in lack of awareness of importance of performance evaluation. In this study, two electrodes were fabricated using conductive fabric which can be used as electrode if attached to wearable U-health devices or smart wear in order to measure ECG signal. Two electrodes those were fabricated using conductive fabric were compared the correlation, impedance and CMRR with patch typed Ag-AgCl electrode-normally used for measurement of ECG signal, so that the study would find out if the fabricated electrode can be used with the wearable U-health devices by testing and evaluating performances.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.3
• /
• pp.705-711
• /
• 2013

QRS detection of ECG is the most popular and easy way to detect cardiac-disease. But it is difficult to analyze the ECG signal because of various noise types. Also in the healthcare system that must continuously monitor people's situation, it is necessary to process ECG signal in realtime. In other words, the design of algorithm that exactly detects QRS wave using minimal computation and classifies PVC by analyzing the persons's physical condition and/or environment is needed. Thus, efficient QRS detection and PVC classification based on profiling method is presented in this paper. For this purpose, we detected QRS through the preprocessing method using morphological filter, adaptive threshold, and window. Also, we applied profiling method to classify each patient's normal cardiac behavior through hash function. The performance of R wave detection, normal beat and PVC classification is evaluated by using MIT-BIH arrhythmia database. The achieved scores indicate the average of 99.77% in R wave detection and the rate of 0.65% in normal beat classification error and 93.29% in PVC classification.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.3
• /
• pp.33-40
• /
• 2012

Recently, to enhance the quality and efficiency of health care services, smart health care services utilizing a smart phone and tablet pc are mainly being introduced into the large hospital and the interest for personal health record service which manages personal medical information utilizing PC and smart phone and technology development is actively underway. In order to build a next generation hospital information systems, the interoperability of data between the different hospital information systems and UX consistency between the various user devices must be assured. Therefore, in this paper, medical information standard and the study for improving the performance is conducted to design the gateway system which converting and integrating the medical information under multi-platform environment. In this paper, CCR standard is adapted to the mapping method between the data which is the key feature of the gateway system and design the system utilizing multi-table-based mapping method. To evaluate the performance of the system, experiment was done and the result has been improved approximately 20 percent for response rate.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.6 no.2
• /
• pp.16-27
• /
• 2003

For the last decade, some 400 small and large oil spill accidents have occurred every year. Such accident blow a serious damage to the marine resource and ecosystem, which can't be estimated in terms of economic and environmental losses. The physical/chemical methods used currently may be effective at the initial stage of accidents, but they can't serve to remove the spilled oil completely. Moreover, the dispersant may lead to a secondary contamination detrimental to the lives inhabiting wet lands, beaches and tidal zone. Thus, a new decomposing technology Is required for the environmentally sensitive areas. Bioremediation is the active use of biological techniques to mitigate the consequences of a spill using biological processes and refers both of stimulation of pollutant biodegradation and/or to enhance ecosystem recovery Bioremediation is an economically attractive method for the clean-up of oil-contaminated area. Bioremediation has been demonstrated to be an effective oil spill countermeasure for use in cobble, sand beach, salt marsh, and mud flat environment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.4
• /
• pp.1954-1961
• /
• 2011

The concept of sustainability in residential area is closely related to the notion of amenity, livability, health and interchangeably used with the terms of urban growth management, urban regeneration, new urbanism, urban village movement, compact development, smart growth and the quality of residential environment. Residential Performance Management System (RPMS This system is developed granted by 'Korea Institute of Construction & Transportation Technology Evaluation and Planning' (KICTEP) during 2005-2008.)is a kind of decision making supportive program developed for the evaluation and assessment of various urban performances. It can be utilized for various purposes by many of urban related expert like planner, manager, etc. This study investigate the schematic model for the applying the indicators of quality of residential environment developed by SDI (Seoul Development Institute) using RPMS.