• Journal of the Korean Data and Information Science Society
• /
• 제25권6호
• /
• pp.1207-1219
• /
• 2014

본 연구의 목적은 1) 보험전문가인 보험회사에 근무하는 임직원들이 가지고 있는 채널의 유형, 채널평가 항목, 보험 상품과 보험판매 시 요구되는 채널특성들간의 적합성에 대한 의견을 가지고 생명보험업계와 손해보험업계 각각 2-mode 소셜네트워크 데이터를 구성하고 2) 생명보험업계와 손해보험업계 2-mode 소셜 네트워크 데이터를 1-mode 소셜 네트워크 데이터로 변환하여 생명보험업계와 손해보험업계 두 소셜 네트워크의 구조와 네트워크 특성 변수들을 찾아 비교 분석하고, 소셜 네트워크 기반 측면에서의 생명보험사와 손해보험사의 판매채널 전략의 방향을 제시하고자 한다. 보험 판매채널의 평가에 의한 소셜 네트워크를 비교한 결과, 생명보험업계 소셜 네트워크가 손해보험업계 소셜 네트워크보다 더욱 강한 연결을 보였다. 즉 생명보험업계의 소셜 네트워크의 중심성 변수들이 손해보험업계의 것 들 보다 모두 높게 나타나 생명보험회사들은 채널전략 운영 측면에서 한 방향으로 움직이기가 수월함을 보이는 반면, 손해보험사들은 회사의 규모나 처한 환경에 따라 판매채널 전략이 다를 수 있어 각사의 개별적인 보험 판매 채널전략을 운영해야 함을 의미한다.

• Steel and Composite Structures
• /
• 제19권2호
• /
• pp.485-501
• /
• 2015

Channels are implemented in composite beams as shear connectors in two arrangements, face to face and back to back. No relevant explanation is found in the design codes to clarify the preference of the mentioned arrangements. Besides, the designers do not have a common opinion on this subject; i.e., some recommend the face to face and others, back to back status. In this research, channel shear connectors in composite beams are studied analytically for both arrangements using ABAQUS software. For this purpose, they have been modeled in simply supported beams in the arrangements of face to face and back to back; their effects on the crack initiation load of concrete slabs were monitored. The stiffness values of composite beams were also compared in the two arrangements using force-displacement curve; the results are relatively the same in both cases. Furthermore, the effects of compressive strength of concrete, channel size, length and spacing of channels as well as steel type of channels on the performance of composite beams have been investigated. According to the results obtained in this research, the face to face status shows better performance in comparison with that of back to back, considering the load of concrete fracturing.

• Steel and Composite Structures
• /
• 제29권1호
• /
• pp.23-37
• /
• 2018

In this paper, seismic provisions related to built-up special concentrically braced frames (BSCBFs) are investigated under cyclic loading using non-linear finite element analysis of a single-bay single-story frame. These braces, which contain double angle and double channel brace sections, are considered in two types of single diagonal and X-braced frames. The results of this study show that current seismic provisions such as observing the 0.4 ratio for slenderness ratio of individual elements between stitch connectors are conservative in BSCBFs, and can be increased according to the type of braces. Furthermore, such increments will lead to decreasing or remaining the current middle protected zone requirements of each BSCBFs. Failure results of BSCBFs, which are related to the plastic equivalent strain growth of members and ductility capacity of the models, show that the behaviors of double channel back-to-back diagonal braces are more desirable than those of similar face-to-face ones. Also, for double angle diagonal braces, results show that the failure of back-to-back BSCBFs occurs faster in comparison with face-to-face similar braces. In X-braced frames, cyclic and failure behaviors of built-up face-to-face models are more desirable than similar back-to-back braces in general.

• The Journal of Asian Finance, Economics and Business
• /
• 제7권5호
• /
• pp.205-216
• /
• 2020

This study seeks to empirically analyze the effects of core self-evaluation and adaptive selling behavior on sales performance for female salespersons engaged in door-to-door sales through the face-to-face channel in the wellness industry. This study seeks to examine the importance of adaptive selling, through, salespeople derive appropriate strategies in response to market changes. For female salespeople who use face-to-face channels, this study empirically investigated the relationship between core self-evaluation and adaptive selling, and effects on sales performance. A 31-item survey was constructed, based on prior research. We selected six door-to-door sales companies in South Korea and conducted one-to-one interviews with female salespeople in the Seoul metropolitan area and analyzed 208 pieces of significant data. Results demonstrated that among the core self-evaluation factors for female salespeople, self-esteem, self-efficacy, and neuroticism had an effect on adaptive selling factors, while locus of control did not. These factors were found to affect sales performance through the mediating role of adaptive selling. Improvements in the adaptive selling capabilities of female salespeople in charge of face-to-face channels positively affected sales performance. Management efforts are required to enhance self-esteem, self-efficacy, or neuroticism. These results suggest that companies should support enhancing individual adaptive selling capabilities of their salespeople.

• 대한한의학원전학회지
• /
• 제34권3호
• /
• pp.73-83
• /
• 2021

Objectives : To organize the acupoints used to treat disorders of the head and face in the Zhenjiuzishengjing, and examine their characteristics in application. Methods : 1. The head and face area was divided into 8 parts according to the textbook of meridians and acupoints. Channels belonging to each part were marked. 2. Disorders as mentioned in the Zhenjiuzishengjing were categorized into 8 groups, accordingly. 3. Acupoints used to treat each disorder were organized according to the channels each belonged to. 4. The points were divided according to their proximity, and their application frequency was organized. 5. Based on the organized contents, the characteristics of using proximal and distal points, together with the interrelationship between the channel belonging to the afflicted area and the points locations were examined. Results : In treating disorders in the head and face area, various distal points along with proximal points were suggested in the Zhenjiuzishengjing. In some cases, points belonging to a channel that was irrelevant to the afflicted area were used widely; for proximal points, the Governor/Conception/Triple Energizer/Gallbladder channels were used. For distal points, channels that were related to the Five Zhang were used. Conclusions : Based on the contents of the Zhenjiuzishengjing, the following could be concluded: 1. When treating disorders of the head and face caused by heat, distal points were mostly used. 2. In cases where points which are not part of channels that pass the head or face were used, Zhang disfunction was likely behind such points selection.

• 한국멀티미디어학회논문지
• /
• 제8권11호
• /
• pp.1421-1431
• /
• 2005

입체음향을 구현하는 방법은 두 가지가 있으며, 5.1 채널과 같은 서라운드 시스템(surround system)을 이용하는 방법과 2 채널의 바이노럴 시스템(binaural system)을 이용하는 방법이 있다. 바이노럴 시스템은 사람이 두 귀를 이용하여 음상을 정위하는 원리를 이용하는 방법이다. 일반적으로 라우더 스피커 시스템에서 각 채널사이에 발생하는 크로스토크(crosstalk)는 본래의 입체음향을 재생하기 위해서는 제거되어야만 한다. 이 문제를 해결하기 위해서는 머리의 움직임을 추정할 필요가 있다. 본 논문에서는 청취자의 머리 움직임을 제대로 추정하는 새로운 알고리즘을 제안한다. 제안하는 알고리즘은 얼굴과 눈의 영역 검출을 기본으로 한다. 얼굴 검출은 이미지의 밝기 값을 이용하고 눈 검출은 수학적 형태학(mathematical morphology)을 이용한다. 청취자의 머리가 움직일 때 얼굴 영역과 눈 사이의 경계선의 길이가 변한다. 이 정보를 이용하여 머리 움직임을 추정한다. 컴퓨터 시뮬레이션을 통해 제안하는 알고리즘이 +10오차 범위 내에서 머리의 움직임을 효율적으로 추정하는 것을 확인하였다.

• 전자통신동향분석
• /
• 제37권2호
• /
• pp.42-52
• /
• 2022

Intra- and inter- datacenter data traffic is rapidly increasing due to the spread of smart devices, cloud computing, and non-face-to-face services. Recently, 400-Gbps optical transceivers based on 100-Gbps/channel have been released primarily by major overseas companies. Various solutions for next-generation datacenter interconnect are being proposed by international standardization and multiple source agreement groups. Following this trend, ETRI has developed a 400-Gbps optical transmission/reception engine using 100-Gbps/channel light sources and photodetectors as well as a silica-based AWG. In the future, technologies of optical devices and components for intra-datacenter communication are expected to be developed based on a data rate of 200-Gbps/channel. Thus, 1.6-Tbps class optical transceivers will be released.

• 한국항행학회논문지
• /
• 제14권4호
• /
• pp.553-561
• /
• 2010

본 논문에서는 로봇의 시각시스템에 효과적으로 적용할 수 있는 얼굴 추적 방법을 제안하였다. 제안한 알고리즘은 동영상의 움직임 영역을 검출한 후 얼굴 영역을 추적한다. 동영상의 움직임 검출은 연속되는 2개의 프레임을 사용하여 차영상을 구한 후, 잡음을 제거하기 위한 방법으로 메디안 필터와 침식 및 팽창연산을 사용하여 움직임 영역을 검출한다. 움직임 영역에서 피부색을 추출하기 위하여 표본영상의 칼라 정보를 이용하였다. 칼라정보의 MIN-MAX값을 퍼지화 데이터로 멤버십 함수를 생성한 후, 유사도를 평가하여 피부색 영역과 배경영역을 분리하였다. 얼굴 후보영역에 대하여 CMY 칼라 공간 C 채널에서 눈을 검출하고, YIQ 칼라 공간 Q 채널에서 입을 검출하였다. 지식기반으로 검출된 눈과 입의 특징을 찾아가며 얼굴영역을 추적하였다. 실험영상으로는 10명 각각에 대하여 150프레임의 동영상 총 1,500프레임을 입력받아 실험한 결과, 1,435프레임의 영상에 대하여 움직임 영역이 검출되어 95.7%의 프레임 검출율을 보였으며, 1,401개에 대한 얼굴을 추적 97.6%의 우수한 얼굴 추적결과를 나타내었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권10호
• /
• pp.4080-4097
• /
• 2020

In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.

• 전자공학회논문지CI
• /
• 제44권4호통권316호
• /
• pp.36-44
• /
• 2007

본 논문에서는 무선 오디오와 영상 인터페이스를 이용한 스위치 제어 및 인식 등의 다양한 유비쿼터스 센서 네트워크 응용 서비스를 지원하기 위한 음성과 얼굴인식 시스템을 구현하였다. 제안된 시스템은 하드웨어 부분으로 무선 오디오 및 이미지 센서, 심리응용모델을 이용한 음성인식과 주성분 분석법(PCA: Principal Components Analysis)을 이용한 얼굴이식 알고리즘, 그리고 LDPC(Low Density Parity Check)로 구성되어 있다. 제안된 음성과 얼굴인식 시스템은 센서의 효율적인 에너지 사용을 위하여 HOST PC에 삽입된다. 그리고 인식 시스템의 정확도를 개선하기 위하여 전방향 에러 정정 시스템을 구현하였다. 또한, 무선 채널 잡음의 효과적인 제거와 정정을 위하여 테스트환경과 시뮬레이션 계수를 최적화하였다. 결과적으로 사람 음성과 음성센서의 거리가 1.5m 이하일 경우에 FAR과 FRR을 각각 0.126%, 7.5%를 얻었고, 얼굴인식 알고리즘을 2회로 제한하였을 경우, GAR과 FAR을 각각 98.5%와 0.036%를 획득하였다.