• 한국방사선학회논문지
• /
• 제18권1호
• /
• pp.21-26
• /
• 2024

You only look once v5 (YOLOv5)는 객체 검출 과정에 우수한 성능을 보이고 있는 딥러닝 모델 중 하나다. 그러므로 본 연구의 목적은 양전차방출단층촬영 팬텀 영상에서 다양한 하이퍼 파라미터에 따른 YOLOv5 모델의 성능을 평가했다. 데이터 세트는 500장의 QIN PET segmentation challenge로부터 제공되는 오픈 소스를 사용하였으며, LabelImg 소프트웨어를 사용하여 경계박스를 설정했다. 학습의 적용된 하이퍼파라미터는 최적화 함수 SDG, Adam, AdamW, 활성화 함수 SiLu, LeakyRelu, Mish, Hardwish와 YOLOv5 모델 크기에 따라 nano, small, large, xlarge다. 학습성능을 평가하기 위한 정량적 분석방법으로 Intersection of union (IOU)를 사용하였다. 결과적으로, AdmaW의 최적화 함수, Hardwish의 활성화 함수, nano 크기에서 우수한 객체 검출성능을 보였다. 결론적으로 핵의학 영상에서의 객체 검출 성능에 대한 YOLOV5 모델의 유용성을 확인하였다.

• Journal of Electrical Engineering and Technology
• /
• 제10권1호
• /
• pp.18-29
• /
• 2015

State Estimation (SE) is the basis of a variety of advanced applications used in most modern power systems. An SE problem formed with enough phasor measurement units (PMUs) data is simply a linear weighted least squares problem requiring no iterations. Thus, designing a minimum-cost placement of PMUs that guarantees observability of a power system becomes a worthy challenge. This paper proposes an equivalent integer linear programming method for substation-oriented optimal PMU placement (SOOPP). The proposed method uses an exhaustive search to determine a globally optimal solution representing the best PMU placement for that particular power system. To obtain a more comprehensive model, contingencies and the limitation of the number of PMU measurement channels are considered and embodied in the model as changes to the original constraints and as additional constraints. The proposed method is examined for applicability using the IEEE 14-bus, 118-bus and 300-bus test systems. The comparison between SOOPP results and results obtained by other methods reveals the excellence of SOOPP. Furthermore, practical large-scale power systems are also successfully analyzed using SOOPP.

• 보건의료산업학회지
• /
• 제6권3호
• /
• pp.1-11
• /
• 2012

The purpose of this study is to identify the effects of emergency medical technicians' personal traits and job characteristics on job/social stress, job satisfaction, role conflicts, organizational commitment, and self-efficacy. The study population was emergency medical technicians who is working at the hospitals in Yeungnam province. The 200 questionnaires were administered by using ground mail, e-mail, and personal visits and 156 questionnaires were returned(82.5% response rate). Social stress, job satisfaction and role conflicts were significantly determined by personal traits and job characteristic variables. However, job stress, organizational commitment and self-efficacy was only significantly determined by job characteristic variables. In social stress and role conflicts, the subjects with challenge, sociability, acceptance and prudence traits were tend to be significantly higher scores compared to the stability traits. Additionally, among social stress regression models, adding job characteristics to the personal traits model, $R^2$ was increased up to 19% and adding personal traits to the job characteristics model, $R^2$ was increased up to 14%. In conclusion, the study found that personal traits and job characteristics are important variables in explaining social stress, job satisfaction and role conflicts, that would have important managerial implications for recruiting, hiring and managing either new or current emergency medical technicians efficiently.

• 디지털융복합연구
• /
• 제10권9호
• /
• pp.111-122
• /
• 2012

본 연구는 인터넷 웹사이트에서 소비자운동을 설명하는데 있어서 인지적인 요인과 감정적 요인의 효과를 살펴본 연구이다. 특히, 인지적 요인을 설명하는 그루닉의 상황적 이론과 감정적 요인을 설명하는 분노활성화 이론을 결합하여 통합모델을 제시하였다. 이 모델을 검증하기 위해서 소비자에 웹사이트에서 이슈화된 제품과 관련 소비자들의 커뮤니케이션 활동을 살펴보았다. 웹사이트에서 멤버들의 커뮤니케이션 활동을 분석하기 위해 가입자를 대상으로 온라인 설문을 실시하였다. 분석결과, 상황적 이론 및 분노는 온라인에서 소비자들의 능동적인 커뮤니케이션 활동을 유의미하게 설명하였다. 더불어 상황적 이론의 요인과 분노를 결합한 모델은 더욱 설명력이 높았고 그 중에서도 분노는 강력한 요인으로 판명되었다.

• 한국IT서비스학회지
• /
• 제15권3호
• /
• pp.195-209
• /
• 2016

IoT (Internet of Things) is a collective term referring to application services that provide information through sensors/devices connected to the internet. The real world application of IoT is expanding fast along with growing number of sensors/devices. However, since IoT application relies on vertical combination of sensors/devices networks, information sharing within IoT services remains unresolved challenge. Consequently, IoT sensors/devices demand high construction and maintenance costs, rendering the creation of new IoT services potentially expensive. One solution is to launch an IoT open market for information sharing similar to that of App Store for smart-phones. Doing so will efficiently allow novel IoT services to emerge across various industries, because developers can purchase licenses to access IoT resources directly via an open market. Sharing IoT resource information through an open market will create an echo-system conducive for easy utilization of resources and communication between IoT service providers, resource owners, and developers. This paper proposes the new business model of IoT open market for information sharing, and the requirements for ensuring security and standardization of open markets.

• Journal of Communications and Networks
• /
• 제18권5호
• /
• pp.725-734
• /
• 2016

For multi-mode radar signals in complex electromagnetic environment, different modes of one emitter tend to be deinterleaved into several emitters, called as "extension", when processing received signals by use of existing sorting methods. The "extension" problem inevitably deteriorates the sorting performance of multi-mode radar signals. In this paper, a novel method based on spatial data mining is presented to address above challenge. Based on theories of data field, we describe the distribution information of feature parameters using potential field, and makes partition clustering of parameter samples according to revealed distribution features. Additionally, an evaluation criterion based on cloud model membership is established to measure the relevance between different cluster-classes, which provides important spatial knowledge for the solution of the "extension" problem. It is shown through numerical simulations that the proposed method is effective on solving the "extension" problem in multi-mode radar signal sorting, and can achieve higher correct sorting rate.

• Journal of Electrical Engineering and Technology
• /
• 제10권6호
• /
• pp.2240-2248
• /
• 2015

The integration of the large-scale wind power brings great challenge to the stability of the power grid. This paper investigates and studies the fault on May 14, 2012 of the large-scale cascading trip-off of wind turbines in North China. According to the characteristics of the voltage variation, the fault process is divided into three stages: the pre-event stage, the critical stage before cascading, and the cascading stage. The scenes in the fault are reproduced, using the full-size actual power system model. Simulation models of double-fed induction generators (DFIGs) and SVCs including protection settings and controller strategies are carefully chosen to find out the reason of voltage instability in each stage. Some voltage dynamic that have never been observed before in the faults of the same kind are analyzed in detail, and an equivalent voltage sensitive dynamic model of DFIG is proposed for the fast computation. The conclusions about the voltage dynamics are validated by the actual PMU observation evidence.

• 한국경영과학회지
• /
• 제30권4호
• /
• pp.131-150
• /
• 2005

One of the reasons for popularity of online game is owing to the interactions with other game users that would lead game the users easily involving in it. Such a holistic state is called flow. Theoretically and empirically flow is one of the antecedents influencing the intension of reusing online game. However, since there are so many kinds in online games, all of them are not the same in the way of feeling flow and level of the relationship with reuse. The objective of this study is to analyze the relationship between flow, fators influencing flow and reuse intension in online games genre by genre. For this purpose, we classily online game into role playing, board, arcade and simulation, and suggest a research model that would explain relationship between online game and flow comparing its difference among four game genres. Sampling 730 from online game users, we analyze the suggested model empirically by PLS and prove it to be a valid one. The result of this study would not only provide differentiate Insights to the online game providers in each genre practically but also contribute to explaining intrinsic motivation for the use of information technology theoretically.

• Advances in nano research
• /
• 제3권3호
• /
• pp.143-168
• /
• 2015

Initiation of crack and its growth simulation requires accurate model of traction - separation law. Accurate modeling of traction-separation law remains always a great challenge. Atomistic simulations based prediction has great potential in arriving at accurate traction-separation law. The present paper is aimed at establishing a method to address the above problem. A method for traction-separation law prediction via utilizing atomistic simulations data has been proposed. In this direction, firstly, a simpler approach of common neighbor analysis (CNA) for the prediction of crack growth has been proposed and results have been compared with previously used approach of threshold potential energy. Next, a scheme for prediction of crack speed has been demonstrated based on the stable crack growth criteria. Also, an algorithm has been proposed that utilizes a variable relaxation time period for the computation of crack growth, accurate stress behavior, and traction-separation atomistic law. An understanding has been established for the generation of smoother traction-separation law (including the effect of free surface) from a huge amount of raw atomistic data. A new curve fit has also been proposed for predicting traction-separation data generated from the molecular dynamics simulations. The proposed traction-separation law has also been compared with the polynomial and exponential model used earlier for the prediction of traction-separation law for the bulk materials.

• Advances in concrete construction
• /
• 제5권3호
• /
• pp.221-240
• /
• 2017

The modeling of the mechanical behavior of quasi-brittle materials is still a challenge task, mainly in failure processes when fracture and plasticity phenomena become important actors in dissipative processes which occur in materials like concrete, as instance. Many homogenization-based approaches have been proposed to deal with heterogeneous materials in the last years. In this context, a computational homogenization modeling for concrete is presented in this work using the concept of Representative Volume Element (RVE). The material is considered as a three-phase material consisting of interface zone (ITZ), matrix and inclusions-each constituent modeled by an independent constitutive model. The Representative Volume Element (RVE) consists of inclusions idealized as circular shapes symmetrically and nonsymmetrically placed into the specimen. The interface zone is modeled by means of cohesive contact finite elements. The inclusion is modeled as linear elastic and matrix region is considered as elastoplastic material. A set of examples is presented in order to show the potentialities and limitations of the proposed modeling. The consideration of the fracture processes in the ITZ is fundamental to capture complex macroscopic characteristics of the material using simple constitutive models at mesoscopic level.