• 방송공학회논문지
• /
• 제16권1호
• /
• pp.178-181
• /
• 2011

최근, 모바일 기기의 발전, 팬/틸트 카메라를 사용한 지능형 감시 시스템 등이 발전하면서 비정적 환경에서의 추적알고리즘에 대한 연구가 활발하게 이루어지고 있다. 비정적 환경에서는 배경에 대한 모델링이 어렵기 때문에 주로 컬러, 텍스쳐 등의 피쳐를 이용한 객체 추적이 이루어진다. 이 경우 배경에 나타나는 객체의 컬러와 유사한 컬러들로 인해 추적이 불안정해진다. 본 논문에서는 컬러에 기반한 객체추적 방법들 중에 상대적으로 안정적이고 속도가 빠른 중심 이동 (Centroid Shifting) 기반의 추적 알고리즘을 더욱 안정화하기 위해 배경에 대한 영향을 줄이고, 시간 필터링을 접목하는 방법에 대하여 제안하고자 한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권10호
• /
• pp.4976-4994
• /
• 2018

The diffusion-based salient object detection methods have shown excellent detection results and more efficient computation in recent years. However, the current diffusion-based salient object detection methods still have disadvantage of detecting the object appearing at the image boundaries and different scales. To address the above mentioned issues, this paper proposes a multi-scale diffusion-based salient object detection algorithm with background and objectness seeds. In specific, the image is firstly over-segmented at several scales. Secondly, the background and objectness saliency of each superpixel is then calculated and fused in each scale. Thirdly, manifold ranking method is chosen to propagate the Bayessian fusion of background and objectness saliency to the whole image. Finally, the pixel-level saliency map is constructed by weighted summation of saliency values under different scales. We evaluate our salient object detection algorithm with other 24 state-of-the-art methods on four public benchmark datasets, i.e., ASD, SED1, SED2 and SOD. The results show that the proposed method performs favorably against 24 state-of-the-art salient object detection approaches in term of popular measures of PR curve and F-measure. And the visual comparison results also show that our method highlights the salient objects more effectively.

• 대한가정학회지
• /
• 제38권2호
• /
• pp.109-122
• /
• 2000

The purpose of this study is to find a mother's influence on her children regarding environmental preservation. The questions of this study are how a mother's background variables influence on their own behaviors, how the children's background variables influences on their own behaviors, how a mother's background variables influence on their children behaviors. This is used self-writing questionaires the 538 subjects are 4th, 5th and 6th grade children and their mothers living in Kwangiu, in June, 1998. The data are analyzed by Frequency, Percentage, Mean, Pearson's correlation, Regression analysis, Path analysis, Coronbach'$\alpha$ coefficent, and Factor analysis. These are the major resets of this study. 1) In case of mothers, the background variables for them to be influenced the environmental presentation behaviors show the frequency of contact for environmental information, average income, family cohesions, the level of communication, egoistic orientation, and age. 2) In Children, the background variables which are influenced on their environmental presentation behaviors represent the frequency of contact for environmental information, family cohesion, biospheric orientation, and the grade level of children. This is, the children who have the more chance of environmental information contact, the higher family cohesion, the higher biospheric orientation, and the lower grade level show the more environmental preservation behaviors. 3) The children who grow seeing their mothers'environmental behaviors such as sorted disposal, recycle behavior, and purchasing behaviors also show the more same behaviors.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.410-412
• /
• 2021

본 논문은 객체와 배경 히스토그램을 활용한 개선된 보행자 검출 방식을 제안하고 있다. HOG & SVM 알고리즘을 통해 검출한 객체는 사각형 형태로 검출된다. 사각형 영역 안에는 배경과 객체의 영역이 혼합되어있다. 배경을 제외한 객체의 영역만을 검출한다면 객체 관련 다양한 정보를 쉽게 얻을 수 있다. 검출된 사각형의 크기를 객체의 크기에 맞게 x-y축 투영 알고리즘을 사용하여 재조정한다. 그리고 나서 재조정 된 사각형 내의 객체에 대한 히스토그램을 바탕으로 배경과 객체를 구분하여 개선된 객체를 검출한다. 검출한 객체와 원본의 객체를 비교하는 신뢰성 평가인 정밀도와 재현율의 평균값이 각각 97.9%와 90%를 보이고 있다.

• 한국멀티미디어학회논문지
• /
• 제13권1호
• /
• pp.47-57
• /
• 2010

객체 검출은 영상에서 객체의 식별, 위치정보, 상황인식 등을 위해서 필수적이다. 본 논문에서는 강인한 객체 검출 시스템을 제안한다. Principal Component Analysis (PCA)를 이용하여 배경 영상에서 수집한 학습데이터를 주성분으로 선형변환 한다. 객체와 배경에 대하여 판별 능력이 우수한 주성분을 선별하여 Eigen-background를 구성한다. Fuzzy-C-Means (FCM)은 Eigen-background의 정보를 입력 차원으로 하여 영상을 Clustering하고 객체와 배경으로 분류한다. 고정된 카메라에서 배경변화에 적용 가능한 시스템을 구현하기 위해 동일한 시점에서 움직이는 객체가 포함된 영상을 학습데이터로 사용하였다. 제안하는 시스템은 인위적인 한 프레임을 배경으로 정의하여 사용하는 과정이 필요 없이 입력 영상에서 잡음이 제거된 객체와 배경으로 분류하였고, 또한 객체의 부분적인 움직임도 효과적으로 검출하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권4호
• /
• pp.1712-1731
• /
• 2016

In this paper, we propose a novel saliency detection framework via multiple random walks (MRW) which simulate multiple agents on a graph simultaneously. In the MRW system, two agents, which represent the seeds of background and foreground, traverse the graph according to a transition matrix, and interact with each other to achieve a state of equilibrium. The proposed algorithm is divided into three steps. First, an initial segmentation is performed to partition an input image into homogeneous regions (i.e., superpixels) for saliency computation. Based on the regions of image, we construct a graph that the nodes correspond to the superpixels in the image, and the edges between neighboring nodes represent the similarities of the corresponding superpixels. Second, to generate the seeds of background, we first filter out one of the four boundaries that most unlikely belong to the background. The superpixels on each of the three remaining sides of the image will be labeled as the seeds of background. To generate the seeds of foreground, we utilize the center prior that foreground objects tend to appear near the image center. In last step, the seeds of foreground and background are treated as two different agents in multiple random walkers to complete the process of salient object detection. Experimental results on three benchmark databases demonstrate the proposed method performs well when it against the state-of-the-art methods in terms of accuracy and robustness.

• 대한인간공학회지
• /
• 제36권3호
• /
• pp.213-219
• /
• 2017

Objective: The present study identified the minimum legible text size by an experiment for eight combinations of background and text colors, which will be used in designing visual information on group-view display (GVD). Background: Information on minimum legible text size is needed to design the visual information presented on GVD in a radioactive waste control room. Method: The experiment was conducted for 22 male participants (age: mean = 37, SD = 6.7; visual acuity: over 0.8) who were recruited by considering demographic characteristics of current control room operators. Eight combinations of background and text colors were considered and the minimum legible text size was determined for each combination by applying the method of limits, one of psychophysical methods. Results: The minimum legible text size was significantly different in accordance with the combination of background and text colors. Statistical analysis results showed that luminance contrast and color contrast between background and text influenced the minimum legible text sizes. Conclusion: This study concluded that the minimum legible text size is 8 minute of arc for various combinations of background and text colors. Application: The minimum legible text size identified in the present study can be utilized in designing visual information on GVD at the main control room in a radioactive waste facility.

• 한국정보통신학회논문지
• /
• 제25권3호
• /
• pp.396-402
• /
• 2021

최근 스마트 도시를 구축하기 위해 무인 차량 관제 시스템의 보급이 활성화 되고 있다. 본 논문은 적응적 배경영상 모델링 방법을 이용한 불법주정차 무인단속시스템에 관한 것으로서, 적응적 가우시안 혼합 모델로 배경 영상을 모델링할 때, 이동 물체의 상황 변화에 따라 전역적으로 배경 영상을 업데이트하거나 국소적으로 배경 영상을 업데이트하는 방법에 대해 기술한다. 특히, 이동 물체가 배경 영상에 미치는 영향을 최소화하는 방법과 배경 영상을 정확하게 업데이트하기 위한 방법을 제안한다. 본 논문에서는 시스템의 구현을 통해 제안하는 시스템이 이동하고 있는 물체 또는 정지상태의 물체를 신속하고 정확하게 구분할 수 있음을 증명하였다.

• Asia pacific journal of information systems
• /
• 제24권1호
• /
• pp.93-112
• /
• 2014

As personal data breach reared up as a problem domestically and globally, organizations appointing chief privacy officers (CPOs) are increasing. Related Korean laws, 'Personal Data Protection Act' and 'the Act on Promotion of Information and Communication Network Utilization and Information Protection, etc.' require personal data processing organizations to appoint CPOs. Research on the characteristics and role of CPO is called for because of the importance of CPO being emphasized. There are many researches on top management's role and their impact on organizational performance using the Upper Echelon theory. This study investigates what influence the characteristics of CPO gives on the organizational privacy performance. CPO's definition varies depending on industry, organization size, required responsibility and power. This study defines CPO as 'a person who takes responsibility for all the duties on handling the organization's privacy,' This research assumes that CPO characteristics such as role, personality and background knowledge have an influence on the organizational privacy performance. This study applies the part relevant to the upper echelon's characteristics and performance of the executives (CEOs, CIOs etc.) for CPO. First, following Mintzberg and other managerial role classification, information, strategic, and diplomacy roles are defined as the role of CPO. Second, the "Big Five" taxonomy on individual's personality was suggested in 1990. Among these five personalities, extraversion and conscientiousness are drawn as the personality characteristics of CPO. Third, advance study suggests complex knowledge of technology, law and business is necessary for CPO. Technical, legal, and business background knowledge are drawn as the background knowledge of CPO. To test this model empirically, 120 samples of data collected from CPOs of domestic organizations are used. Factor analysis is carried out and convergent validity and discriminant validity were verified using SPSS and Smart PLS, and the causal relationships between the CPO's role, personality, background knowledge and the organizational privacy performance are analyzed as well. The result of the analysis shows that CPO's diplomacy role and strategic role have significant impacts on organizational privacy performance. This reveals that CPO's active communication with other organizations is needed. Differentiated privacy policy or strategy of organizations is also important. Legal background knowledge and technical background knowledge were also found to be significant determinants to organizational privacy performance. In addition, CPOs conscientiousness has a positive impact on organizational privacy performance. The practical implication of this study is as follows: First, the research can be a yardstick for judgment when companies select CPOs and vest authority in them. Second, not only companies but also CPOs can judge what ability they should concentrate on for development of their career relevant to their job through results of this research. Cultural social value, citizen's consensus on the right to privacy, expected CPO's role will change in process of time. In future study, long-term time-series analysis based research can reveal these changes and can also offer practical implications for government and private organization's policy making on information privacy.

• 한국문헌정보학회지
• /
• 제50권3호
• /
• pp.455-490
• /
• 2016

본 연구는 한국의 다문화가정 청소년이 마주하고 있는 사회적, 문화적 요인이 그들의 일상 정보 행태, 특히 미디어 사용을 통한 정보 행태에 미치는 영향의 양상을 살펴보고자 하였다. 13세에서 18세 사이의 다문화가정 청소년 11명을 대상으로 정보요구기록지, 미디어 이용에 관한 설문조사, 두 차례에 걸친 심층 면담이 실시되었으며, 이를 통해 얻은 자료를 바탕으로 다문화가정 청소년의 일상 문제 상황과 미디어를 통한 정보 요구 해결의 양상을 다각적인 시각에서 분석하였다. 다문화가정 청소년이 일상의 문제 상황에서 필요로 하는 정보의 주제는 학업, 여가, 외모, 자기개발, 생활, 진로 등이었으며, 특히 학업과 여가에 대한 정보 요구가 압도적으로 높았다. 본 연구에 참가한 다문화가정 청소년들은 정보 요구의 해결을 위해 능동적으로 미디어를 사용하고 있었다. 이들에게 미디어는 심리적 도피처이자 안식처였고, 자기표현을 위한 도구였으며, 지식과 생활 정보를 얻을 수 있는 해답의 책이자 여가 시간의 동반자였다. 본 연구에서 제시된 미디어 이용 관련 방안들은 추후 도서관에서 다문화가정 자녀를 대상으로 한 서비스를 기획하는데 활용될 수 있을 것이라 기대한다.