• Title/Summary/Keyword: 활성화모델

Search Result 1,505, Processing Time 0.025 seconds

Estimation of Compressive Strength of the Fly Ash Substitution Cement Mortar by Equivalent age (등가재령 방법에 의한 플라이애시를 치환한 시멘트 모르타르의 강도 증진 해석)

  • Han, Min-Cheol
    • Journal of the Korean Recycled Construction Resources Institute
    • /
    • v.7 no.4
    • /
    • pp.121-127
    • /
    • 2012
  • This paper is to present the strength estimation of the cement mortar incorporating 20% of fly ash by equivalent age method. ASTM C 1074 was applied to achieve apparent activation energy($E_a$). Cement mortar was cured at the temperature of $5^{\circ}C$, $20^{\circ}C$ and $35^{\circ}C$ respectively to measure the setting time and compressive strength at designed age. According to test results, it is found that an increase in curing temperature resulted in an acceleration of setting time. $E_a$ was achieved to 34.75 KJ/mol. It was also found that by estimating strength development with Plowman and Gompertz model, good agreement between calculated value and measured one was achieved.

  • PDF

Performance Evaluation of YOLOv5 Model according to Various Hyper-parameters in Nuclear Medicine Phantom Images (핵의학 팬텀 영상에서 초매개변수 변화에 따른 YOLOv5 모델의 성능평가)

  • Min-Gwan Lee;Chanrok Park
    • Journal of the Korean Society of Radiology
    • /
    • v.18 no.1
    • /
    • pp.21-26
    • /
    • 2024
  • The one of the famous deep learning models for object detection task is you only look once version 5 (YOLOv5) framework based on the one stage architecture. In addition, YOLOv5 model indicated high performance for accurate lesion detection using the bottleneck CSP layer and skip connection function. The purpose of this study was to evaluate the performance of YOLOv5 framework according to various hyperparameters in position emission tomogrpahy (PET) phantom images. The dataset was obtained from QIN PET segmentation challenge in 500 slices. We set the bounding box to generate ground truth dataset using labelImg software. The hyperparameters for network train were applied by changing optimization function (SDG, Adam, and AdamW), activation function (SiLU, LeakyRelu, Mish, and Hardwish), and YOLOv5 model size (nano, small, large, and xlarge). The intersection over union (IOU) method was used for performance evaluation. As a results, the condition of outstanding performance is to apply AdamW, Hardwish, and nano size for optimization function, activation function and model version, respectively. In conclusion, we confirmed the usefulness of YOLOv5 network for object detection performance in nuclear medicine images.

A Study on Promotion of Value Added Logistics(VAL) Activities of Lumber Hinterland in Incheon Northport (인천북항 목재배후단지 부가가치물류 활성화방안)

  • Chung, Tae-Won;Han, Jong-Khill
    • Journal of Navigation and Port Research
    • /
    • v.35 no.10
    • /
    • pp.847-853
    • /
    • 2011
  • The purpose of this study is to develope a series of business models for promotion of value added logistics activities of lumber hinterland in Incheon Northport. A set of policies to create value added developed are an import market diversification model using Incheon Northport, a model by building integration processing center by pallet facility, a model by joint logistics center, an export-oriented Pre-cut material development model and an export processing value added model.

Bottom-up Approach of Rule Rewriting in Neural Network Rule Extraction (신경망 규칙 추출에서 규칙 결합의 bottom-up 접근 방법)

  • Lee, Eun Hun;Kim, Hyeoncheol
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2018.10a
    • /
    • pp.916-919
    • /
    • 2018
  • 심층신경망 모델은 우수한 성능을 갖고 있음에도 불구하고 모델이 어떤 판단 과정을 통해 결론을 내렸는지 파악하기 어렵다. 그에 따라 판단에 대한 근거가 중요한 분야에서는 심층신경망 모델을 적용한 실제 사례를 찾기 어렵다. 인공신경망 모델을 해석하기 어렵다는 문제를 해결하기 위해 내부 구조를 이용하여 규칙을 추출하는 decompositional 접근법이 제안되었으나 기존의 연구는 대부분 은닉층이 1개인 다층 퍼셉트론 모델에서 규칙을 생성하는 것을 가정하고 있다. 오늘날 사용하는 심층신경망 모델은 일반적으로 여러 은닉층을 가지고 있기 때문에 기존의 접근법을 그대로 적용할 경우 규칙 불확실성에 따라 잘못된 규칙을 추출하는 문제가 발생한다. 본 논문은 decompositional 접근법에 존재하는 규칙 불확실성 문제를 완화하고 깊이가 깊은 심층신경망 모델에 규칙을 추출하는 방법을 제안한다. 제안한 접근법은 실제 활성화 값을 통해 지식을 추출하며, 이를 통해 규칙 불확실성 문제를 완화할 수 있었다.

Development of Business Model for Partnership Support Management (PSM) in Private Infrastructure Projects (민간투자사업 파트너쉽지원관리(PSM) 업무모델 개발)

  • Jung, Pyung Ki;Baek, Hyeon Gi;Seo, Jong Won
    • KSCE Journal of Civil and Environmental Engineering Research
    • /
    • v.31 no.1D
    • /
    • pp.99-109
    • /
    • 2011
  • In order to maintain construction industry and to improve efficiency, the Korean government legislated private investment system. Private sector takes advantages by adopting CM (Construction Management) project system, however, public sector has relatively less capability and experience than private sector. In this paper, PSM (Partnership Support Management) is addressed in order to actively strengthen the project capabilities for public sector. PSM maintains existing CM project system because it is a concept of expended business model of CM. PSM needed to categorize the private participated project into 5 phases; project development phase, project announcement phase, project evaluation phase, project agreement phase and agreement changing phase. PSM business structure and business process considering stakeholder's interrelationship and participation are suggested in this research. Suggested PSM business model may not only increase a comprehensive understanding of government and project capability, but also may reduce work load and responsibilities for project momentum of private participation in infrastructure.

Design and Implementation of Gamified Tourtainment Service Platform (게임화 투어테인먼트 서비스 플랫폼 설계 및 구현)

  • Lee, Deuk-Woo;Cho, Byung-Sang;Hwang, Ho-Sung;Kim, Se-Hyeon
    • The Journal of the Institute of Internet, Broadcasting and Communication
    • /
    • v.17 no.3
    • /
    • pp.153-158
    • /
    • 2017
  • With the fast increment of fusion ICT(Information and Communication Technology), new creative tourism model is required for the revival of local economy. We come up with new frame that is able to increase tourists and revive local economy through ICT fusion such as spatial informaion, contents, game, smart phone, wearable device and wireless internet. In this paper, we propose 'Gamified Tourtainment Service Platform' that can be provided tourists with user custom tour guidance and gamified tour services for motivation, fun, contest, reward in the environment of outdoor. This platform is a kind of framework for creating and providing gamified creative tour services and a gateway for increasing tourist and reviving commercial local economy. We also propose the real tourtainment service application that is implemented with this platform and smart phone.

Adsorption Characteristics of Acetone, Benzene, and Metylmercaptan by Activated Carbon Prepared from Waste Citrus Peel (폐감귤박으로 제조한 활성탄에 의한 아세톤, 벤젠 및 메틸메르캅탄의 흡착특성)

  • Kam, Sang-Kyu;Kang, Kyung-Ho;Lee, Min-Gyu
    • Applied Chemistry for Engineering
    • /
    • v.28 no.6
    • /
    • pp.663-669
    • /
    • 2017
  • Activated carbons were prepared from waste citrus peels using KOH, NaOH, and $ZnCl_2$ as activating chemicals. They were prepared at optimal conditions including the chemical ratio of 300%, activation time of 1.5h, and activation temperature of $900^{\circ}C$ for KOH, $700^{\circ}C$ for NaOH, and $600^{\circ}C$ for $ZnCl_2$, which were named as ACK, ACN, and ACZ, respectively. Using the activated carbons, their adsorption characteristics for three target gases such as acetone, benzene, and methylmercaptan (MM) were carried out in a batch reactor. The adsorption behavior of activated carbons for three target gases followed the Freundlich model better than the Langmuir. And the experimental kinetic data followed a pseudo-second-order kinetic model more than pseudo-first-order one. Following the intraparticle diffusion model suggested that the external mass transfer and particle diffusion were occurred simultaneously during the adsorption process.

Performance Analysis of Adaptive Radio Activation in Dual-Radio Aggregation System (이중 무선 인터페이스 결합 시스템을 위한 적응적 인터페이스 활성화 기법의 성능 분석)

  • Mulya Saputra, Yuris;Yun, Ji-Hoon
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.19 no.8
    • /
    • pp.1901-1907
    • /
    • 2015
  • Today's smartphones and user devices are equipped with multiple radio interfaces increasingly. Aggregating theses multiple radio interfaces and using them concurrently will increase a user's communication speed immediately, but at the expense of increased power consumption. In this paper, we develop a mathematical performance model of an adaptive radio activation scheme by which a radio interface is activated only when needed for performance increase and deactivated otherwise. The developed model shows that the adaptive scheme reduces delay significantly and almost halves power consumption below a certain level of traffic input.

A Comparison Study on the Parcel Transport Service Model-using High-Speed Passenger/Freight Mixed Train (여객/화물 고속복합열차를 이용한 소화물 운송 서비스 모델 비교연구)

  • Yum, ByongSoo;Ha, Ohkeun;Lee, Jinsun
    • Journal of the Korean Society for Railway
    • /
    • v.18 no.5
    • /
    • pp.471-480
    • /
    • 2015
  • Reliance on road transport in domestic logistics can lead to intensified congestion and greenhouse gas emissions due to the rise in oil prices, any increase in logistics costs can have a high social cost. The government policy on Low Carbon Green Growth is seeking to take advantage of the railway system. However, existing railway transport logistics systems, for reasons such as low speed, low track capacity constraints, and the impossibility of implementing a Door to Door system, make it difficult to activate a railway logistics program. As a result of this study, a national R&D project to develop a High-Speed Passenger/Baggage Mixed Train(Hy-SoBex) utilizing the rail capacity constraints to overcome the difficulties of linking an air cargo and freight railway logistics system, we propose a variety of service models and select the optimal service model.

Government as a Platform Revitalization Strategy Derived from Webtoon Platform Success Factors (웹툰플랫폼 성공요인에서 도출한 플랫폼 정부 활성화 방안)

  • Seo, Hyungjun
    • Journal of Digital Convergence
    • /
    • v.19 no.10
    • /
    • pp.1-13
    • /
    • 2021
  • This study postulates that Government as a Platform(GaaP) could be a next generation government model, drawing activation strategy from success factors of webtoon platform by NAVER. It suggests success factors of Naver webtoon case based on three main components of platform strategy. First, in the aspect of platform infrastructure, Naver webtoon was established based on powerful portal site by parent company Naver and improved platform accessibility through using webtoon without login. Second, in the aspect of stakeholder engagement, Naver webtoon has offed webtoon at no cost and it has taken an intermediary role for supporting the stakeholders with PPS. Third, in the aspect of outputs production, Naver webtoon offered real time feedback from webtoon users that could affect webtoon production and established incubating system that allow users to create their webtoon. And due to PPS, it makes possible to various create second outputs based on webtoon, which can contribute to activating webtoon ecosystem.