• 한국통신학회논문지
• /
• 제38A권1호
• /
• pp.1-9
• /
• 2013

본 논문에서는 CS-LBP (Center-Symmetric Local Binary Pattern) 특징과 공간 피라미드를 이용한 BoF (Bag of Features)를 생성하고 이를 랜덤 포레스트(Random Forest) 분류기에 적용하여 인간의 행동을 인식하는 알고리즘을 제안한다. BoF를 생성하기 위해 영상을 균일한 패치로 나누고, 각 패치 마다 CS-LBP 특징을 추출한다. 행동 분류 성능을 향상시키기 위해 패치들마다 추출한 특징벡터들에 대해 K-mean 클러스터링을 적용하여 코드 북을 생성한다. 본 논문에서는 영상의 지역적인 특성을 고려하기 위해 공간 피라미드 방법을 적용하고 각 공간 레벨에서 추출된 BoF에 대해 가중치를 적용하여 최종적으로 하나의 특징 벡터로 결합한다. 행동 분류를 위해 결정트리의 앙상블로 이루어진 랜덤 포레스트는 학습 단계에서 각 행동 클래스를 위한 분류 모델을 만든다. 가중 BoF가 적용된 랜덤 포레스트는 다양한 인간 행동 영상을 포함하고 있는 Standford Actions 40 데이터를 성공적으로 분류하였다. 또한 기존 방법에 비해 분류 성능이 유사하거나 우수하며, 한 장의 영상에 대해 빠른 인식속도를 보였다.

• 동아시아식생활학회지
• /
• 제20권2호
• /
• pp.313-320
• /
• 2010

In this study, kiwi fruit, which has various medical actions and excellent functionality, was used together with gruels for comparison and analysis of to compare and analyze the changes in physico-chemical quality of a kiwi fruit-gruel according to the amount of kiwi concentrate added to the gruel. Analysis on the sensual quality of kiwi fruit-gruel spread was done to indicate to it was executed to find out the merchandising potential of the preparation and to provide basic data for gruel commercialization. Increasing content of kiwi concentrate decreased the solid content and pH of kiwi fruit, and increased sweetness and viscosity. Significant changes in chromaticity to L(lightness) and a(redness), texture qualities of hardness and cohesiveness, gumminess, springiness, color, flavor, sourness, and sweetness were evident with increasing kiwi concentrate. Concerning overall acceptability and sensory evaluation, 20% softness samples were most preferred. The findings may lead to enhanced medical/nutritional activity and functionality of kiwi fruit-supplemented gruel. The more the kiwi concentrate was added to the gruel, solid content and pH of the kiwi fruit-grue decreased. The sweetness and viscosity of kiwi fruit-gruel increased. For chromaticity to L(lightness) and a(redness) of the gruel decreased and showed significant difference between sample groups. For texture, there were significant changes shown for hardness and cohesiveness the amount of kiwi concentrate added was different, increased There was a significant change between samples for gumminess. As for springiness, there was a significant change between samples. There were significant difference between each sample for color, flavor, sourness, sweetness. For softness, overall-acceptability and sensory evalution 20% group was most preferred based on this study, from the physico-chemical and sensual perspective, 20% group is highly recommended for adding kiwi concentrate into gruel. In conclusion, this study is meaningful as it enhanced the functionality and quality of a gruel by combining kiwi fruit with various medical action and excellent functionality, with Kiwi Fruit and presented a basic material for creating a kiwi concentrate combined gruels.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권7호
• /
• pp.3191-3194
• /
• 2014

Adenocarcinoma of lung has high incidence and a poor prognosis, woith chemotherapy as the main therapeutic tool, most commonly with cisplatin. However, chemotherapy resistance develops in the majority of patients during clinic treatment. Mechanisms of resistance are complex and still unclear. Although annexin play important roles in various tumor resistance mechanisms, their actions in cisplatin-resistant lung adenocarcinoma remain unclear. Preliminary studies by our group found that in cisplatin-resistant lung cancer A549 cells and lung adenocarcinoma tissues, both mRNA and protein expression of annexins A1, A2 and A3 is increased. Using a library of annexin A1, A2 and A3 targeting combined molecules already established by ourselves we found that specific targeting decreased cisplatin-resistance. Taken together, the underlined effects of annexins A1, A2 and A3 on drug resistance and suggest molecular mechanisms in cisplatin-resistant A549 cells both in vivo and in vitro. Furthermore, the study points to improved research on occurrence and development of lung adenocarcinoma, with provision of effective targets and programmes for lung adenocarcinoma therapy in the clinic.

• 한국콘텐츠학회논문지
• /
• 제8권11호
• /
• pp.286-295
• /
• 2008

본 연구는 점포충성도에 영향을 미치는 브랜드만족, 브랜드신뢰, 점포만족, 점포신뢰 간의 구조적 관계를 검증하였다. 분석결과 점포충성도는 단순하게 브랜드나 점포 각각에 대한 만족과 신뢰 형성보다는 브랜드와 점포가 결합된 만족과 신뢰를 제공함으로써 강력한 점포충성도를 형성할 수 있었다. 또한 점포충성도 형성 과정으로는 관계지향적 고객의 경우 브랜드만족이 브랜드신뢰에 영향을 주고, 브랜드신뢰는 점포신뢰에 영향을 주어 점포충성도를 형성하였고, 거래지향적 고객의 경우 브랜드만족이 점포만족에 영향을 주어 직접적으로 점포충성도를 형성하는 두 가지 경로를 확인하였다. 따라서 관계지향적 고객에게는 소비자가 점포브랜드에 신뢰할 수 있도록 차별화할 수 있는 파워브랜드를 제공하여야 하고, 거래지향적 고객에게는 점포에 만족할 수 있도록 점포환경을 개선하고 서비스를 강화하여야 하는 마케팅시사점을 제공한다.

• 한국시스템다이내믹스연구
• /
• 제16권3호
• /
• pp.5-29
• /
• 2015

The importance of the concept of safety culture has increased in the security of high-risk facility after Chernobyl accident in 1986. This paper elaborated the concept of safety culture and its main factors by Causal Loop Diagram. Due to the decline of safety culture, the occurrence of incidents and accidents require more and more corrective actions to the members of high-risk facilities and thereby increasing their workloads. Employees who must complete the task within the given time have to have time pressures and don't comply with the rules and procedures. Also, a schedule pressure is a big stress for employees, causing mistakes in precision work. In order to improve these problems, CLD of the safety culture in this paper suggests hiring more workers, re-allocation of given workloads and strengthen the learning, communication capabilities and safety leadership. In addition, the two real accident cases were analyzed to test the feasibility of the System Dynamic simulation model through the process of structuring the fault trees on the stationary black out accident in Kori unit 1 in South Korea and Kleen Energy power station explosion in US. The simulation results show that the various safety factors cause the serious accident combined with mechanical failure and safety culture will reduce the possibility of the accidents in these high-risk organizations. This simulation model can contribute to analyzing the impact of the organizational and human factors of safety culture and can provide the alternatives in high-risk facilities.

• Smart Structures and Systems
• /
• 제25권1호
• /
• pp.111-122
• /
• 2020

Researchers up to date have introduced several Structural Health Monitoring (SHM) techniques with varying advantages and drawbacks for each. Satellite positioning systems (GPS, GLONASS and GALILEO) based techniques proved to be promising, especially for high natural period structures. Particularly, the GPS has proved sufficient performance and reasonable accuracy in tracking real time dynamic displacements of flexible structures independent of atmospheric conditions, temperature variations and visibility of the monitored object. Tall structures are particularly sensitive to oscillations produced by different sources of dynamic actions; such as typhoons. Wind forces induce in the structure both longitudinal and perpendicular displacements with respect to the wind direction, resulting in torsional effects, which are usually more complex to be detected. To efficiently track the horizontal rotations of the in-plane sections of such flexible structures, two main issues have to be considered: a suitable sensor topology (i.e., location, installation, and combination of sensors), and the methodology used to process the data recorded by sensors. This paper reports the contributions of the measurements recorded from dual frequency GPS receivers and uni-axial accelerometers in a full-scale experimental campaign. The Canton tower in Guangzhou-China is the case study of this research, which is instrumented with a long-term structural health monitoring system deploying both accelerometers and GPS receivers. The elaboration of combining the obtained rather long records provided by these two types of sensors in detecting the torsional behavior of the tower under ambient vibration condition and during strong wind events is discussed in this paper. Results confirmed the reliability of GPS receivers in obtaining the dynamic characteristics of the system, and its ability to capture the torsional response of the tower when used alone or when they are combined with accelerometers integrated data.

• 콘크리트학회논문집
• /
• 제17권4호
• /
• pp.513-520
• /
• 2005

현재 유럽 및 일본의 경우 기둥 및 보 부재에 대한 전단 설계에 있어, 전단기구로서 전단보강근과 주철근의 부착작용을 필요로 하지 않는 아치기구와 전단보강근과 주근의 부착 작용을 필요로 하는 트러스기구를 고려하여, 양자의 합으로 전단내력을 평가하는 방식을 취하고 있다. 이러한 설계 방법은 매우 명료하고, 또한 힘의 평형 조건에 기초하여, 단적으로 부착의 좋고 나쁨 및 전단보강근 양의 대소로 정해지는 상한 값으로 전단 내력을 평가하는 것이 특징이다. 본 연구에서는, 역대칭 휨 모멘트를 받는 철근콘크리트 보 부재에서 단부의 텐션쉬프트 영역 사이를 대상으로 한 주철근의 축 방향 응력(압축, 인장) 및 주근의 직경, 부착길이의 상이함에 기인하는 부착거동의 차이와 트러스 기구와의 관계를 명확히 하는 것을 목적으로, 트러스기구만이 재현 가능한 모델 실험법을 제안하고, 전단보강근의 양, 부재의 축압축 응력의 크기 및 부착 영역 양단의 주철근에 가하는 힘의 크기를 실험 변수로 한 부재 실험을 실시하였다. 실험에서 얻은 결과에 기초하여, 양단부 주철근에 가한 힘의 차이로 인해 발생하는 부착응력 분포의 차이를 고려한 일정 트러스 기구와 부채형 트러스기구를 중첩한 복합 트러스 모델을 제안하였다.

• The Korean Journal of Physiology and Pharmacology
• /
• 제2권3호
• /
• pp.287-295
• /
• 1998

Diazepam is known to have cardiovascular depressive effects through a combined action on benzodiazepinergic receptor and the GABA receptor-chloride ion channel complex. Moreover, it is known that barbiturates also have some cardiovascular regulatory effects mediated by the central GABAergic system. Therefore, this study was undertaken to delineate the regulatory actions and interactions of these systems by measuring the responses of the cardiovascular system and renal nerve activity to muscimol, diazepam and pentobarbital, administered intracerebroventricularly in rabbits. When muscimol $(0.03{\sim}0.3\;{\mu}\;g/kg)$, diazepam $(10{\sim}100\;{\mu}\;g/kg)$ and pentobarbital $(1{\sim}10\;{\mu}\;g/kg)$ were injected into the lateral ventricle of the rabbit brain, there were similar dose-dependent decreases in blood pressure (BP) and renal nerve activity (RNA). The relative potency of the three drugs in decreasing BP and RNA was muscimol ＞ pentobarbital ＞ diazepam. Muscimol and pentobarbital also decreased the heart rate in a dose-dependent manner; however, diazepam produced a trivial, dose-independent decrease in heart rate. Diazepam $(30\;{\mu}g/kg)$ augmented the effect of muscimol $(0.1\;{\mu}g/kg)$ in decreasing blood pressure and renal nerve activity, but pentobarbital $(3\;{\mu}g/kg)$ did not. Bicuculline $(0.5\;{\mu}g/kg)$, a GABAergic receptor blocker, significantly attenuated the effect of muscimol in decreasing BP and RNA, either alone or with diazepam, and that of pentobarbital in decreasing BP and RNA, either alone or with muscimol. We inferred that the central benzodiazepinergic and barbiturate systems help regulate peripheral cardiovascular function by modulating the GABAergic system, which adjusts the output of the vasomotor center and hence controls peripheral sympathetic tone. Benzodiazepines more readily modulate the GABAergic system than barbiturates.

• 한국산학기술학회논문지
• /
• 제7권6호
• /
• pp.1356-1363
• /
• 2006

본 연구의 목적은 모바일 정보통신 시장에 있어서, 이에 관련한 공급자가 어떻게 성공적으로 기업을 관리함으로써 고객 충성도를 증대시킬 것인가에 대한 것이다. 이러한 목적을 달성하기 위하여 본 연구는 우즈베키스탄의 모바일 정보통신 시장을 사례분석 함으로써 외국의 경우를 고찰하고자 하였다. 본 소고에서는 우선적으로 CRM의 영향력과 제반 요인들이 CRM 시스템을 통하여 고객 이해력에 의한 충성도를 제고시킬 것인가를 분석 검토 하고자 하였다. 고객의 욕구와 고객 만족, 고객 충성도의 관련성들이 어떻게 기업의 수익성과 연계되어 발생하는가는 여러 가지 복잡한 방법들과 관련되어 있는 것으로 나타났다. 본 연구의 특징은 외국의 사례를 중심으로 고객만족과 고객충성도의 관련 속에서 분리한 경우와 결합한 경우의 특성들로 각각 분석 고찰 한 점이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.