• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11a
• /
• pp.65-68
• /
• 2003

본 논문에서는 가장 최근의 동영상 표준인 H.264에서 가변 블록 움직임 예측 시 인접한 블록과의 상관성을 분석하여 병합 절차를 추가함으로써 매크로블록의 최종 모드를 결정하는 시간을 줄이기 위한 알고리즘을 제안한다. H.264에서는 매크로블록의 모드를 결정하기 위하여 총 7가지 모드를 사용하여 움직임 예측은 실시함으로써 부호화 효율을 극대화시킨 반면 이러한 움직임 예측이 부호화기의 복잡도를 높이는 주요 요인으로 현재 커다란 단점으로 지적되고 있다. 본 논문에서는 $8{\times}8$ 움직임 예측이 끝난 후 인접한 두 블록 사이의 거리론 임계값(Threshold)과 비교하여 다음 모드의 움직임 예측의 실시 여부를 먼저 절정함으로써 필요한 움직임 예측에 소비되는 시간을 단축시켰다. 여기서 실험 조건으로 명시하고 있는 것은 대표적인 단일모드 중에서 수행 성능이 가장 좋은 $8{\times}8$ 모드를 기본모드로 사용하고 병합 시 $16{\times}16$ 모드 쪽으로 상향식(bottom-up) 방법의 병합을 수행해 나아간다 모의실험을 통해 수행 성능과 전체 부호화 시간 측면을 본 논문에서 제안한 방법과 4가지 모드인 $16{\times}16,\;16{\times}8,\;8{\times}16,\;8{\times}8$ 모드를 모두 사용한 경우, $8{\times}8$ 단일모드를 사용한 경우를 비교하였다. 실험 결과 $8{\times}8$ 단일모드보다 수행 성능이 향상되었으며, 시간 단축 면에서 제안한 방법이 4가지 모드인 $16{\times}16,\;16{\times}8,\;8{\times}16,\;8{\times}8$ 모드를 모두 사용한 경우와 $8{\times}8$ 단일모드를 사용한 경우보다 계산 시간이 감소하였음을 확인하였다.행중인 MoIM-Messge서버의 네트워크 모듈로 다중 쓰레드 소켓폴링 모델을 적용하였다.n rate compared with conventional face recognition algorithms. 아니라 실내에서도 발생하고 있었다. 정량한 8개 화합물 각각과 총 휘발성 유기화합물의 스피어만 상관계수는 벤젠을 제외하고는 모두 유의하였다. 이중 톨루엔과 크실렌은 총 휘발성 유기화합물과 좋은 상관성 (톨루엔 0.76, 크실렌, 0.87)을 나타내었다. 이 연구는 톨루엔과 크실렌이 총 휘발성 유기화합물의 좋은 지표를 사용될 있고, 톨루엔, 에틸벤젠, 크실렌 등 많은 휘발성 유기화합물의 발생원은 실외뿐 아니라 실내에도 있음을 나타내고 있다.>10)의 $[^{18}F]F_2$를 얻었다. 결론: $^{18}O(p,n)^{18}F$ 핵반응을 이용하여 친전자성 방사성동위원소 $[^{18}F]F_2$를 생산하였다. 표적 챔버는 알루미늄으로 제작하였으며 본 연구에서 연구된 $[^{18}F]F_2$가스는 친핵성 치환반응으로 방사성동위원소를 도입하기 어려운 다양한 방사성의 약품개발에 유용하게 이용될 수 있을 것이다.었으나 움직임 보정 후 영상을 이용하여 비교한 경우, 결합능 변화가 선조체 영역에서 국한되어 나타나며 그 유의성이 움직임 보정 전에 비하여 낮음을 알 수 있었다. 결론: 뇌활성화 과제 수행시에 동반되는 피험자의 머리 움직임에 의하여 도파민 유리가 과대평가되었으며 이는 이 연구에서 제안한 영상정합을 이용한 움직임 보정기법에 의해서 개선되었다. 답이 없는 문제, 문제 만들기, 일반화가 가능한 문제 등으로 보고, 수학적 창의성 중 특히 확산적 사고에 초점을 맞추어 개방형 문제가 확산적 사고의 요소인 유창성, 독창성, 유연성 등에 각각 어떤 영향을 미치는지 20주의 프로그램을 개발, 진행하여 그 효과를 검증하고자 한다. 개방형 문

• Journal of IKEEE
• /
• v.21 no.2
• /
• pp.136-141
• /
• 2017

In this paper, we propose an algorithm for memory optimization to apply deep learning to PC. The proposed algorithm minimizes the memory and computation processing time by reducing the amount of computation processing and data required in the conventional deep learning structure in a general PC. The algorithm proposed in this paper consists of three steps: a convolution layer configuration process using a random filter with discriminating power, a data reduction process using PCA, and a CNN structure creation using SVM. The learning process is not necessary in the convolution layer construction process using the discriminating random filter, thereby shortening the learning time of the overall deep learning. PCA reduces the amount of memory and computation throughput. The creation of the CNN structure using SVM maximizes the effect of reducing the amount of memory and computational throughput required. In order to evaluate the performance of the proposed algorithm, we experimented with Yale University's Extended Yale B face database. The results show that the algorithm proposed in this paper has a similar performance recognition rate compared with the existing CNN algorithm. And it was confirmed to be excellent. Based on the algorithm proposed in this paper, it is expected that a deep learning algorithm with many data and computation processes can be implemented in a general PC.

• Journal of IKEEE
• /
• v.25 no.2
• /
• pp.273-279
• /
• 2021

In this paper, we propose an algorithm that shortens the learning time by performing individual learning using partitioning the deep learning structure. The proposed algorithm consists of four processes: network classification origin setting process, feature vector extraction process, feature noise removal process, and class classification process. First, in the process of setting the network classification starting point, the division starting point of the network structure for effective feature vector extraction is set. Second, in the feature vector extraction process, feature vectors are extracted without additional learning using the weights previously learned. Third, in the feature noise removal process, the extracted feature vector is received and the output value of each class is learned to remove noise from the data. Fourth, in the class classification process, the noise-removed feature vector is input to the multi-layer perceptron structure, and the result is output and learned. To evaluate the performance of the proposed algorithm, we experimented with the Extended Yale B face database. As a result of the experiment, in the case of the time required for one-time learning, the proposed algorithm reduced 40.7% based on the existing algorithm. In addition, the number of learning up to the target recognition rate was shortened compared with the existing algorithm. Through the experimental results, it was confirmed that the one-time learning time and the total learning time were reduced and improved over the existing algorithm.

• Industry Promotion Research
• /
• v.7 no.3
• /
• pp.77-83
• /
• 2022

This study was conducted with the purpose of helping middle-aged women who experienced insecure attachment during personal growth experience positive changes by re-illuminating their own growth process. During this study period, 14 in-depth interviews were conducted from August 2020 to September 2021, and the interview contents were based on the narrative research methodology to examine the meaning of participants' experiences regarding unstable attachment and post-traumatic growth in existential contexts. The text was described in terms of, relational context, life context, etc. As a result of the study, three participants who experienced unstable attachment and post-traumatic growth were selected and the following conclusions were drawn. First, the meaning in the existential context is the desire for recognition, perfectionism, unstable family environment, how to cope with stress, the courage to face the wounds, self-acceptance and affirmation, gratitude to the people around you, and the hope of life is the meaning in the participant experience. was interpreted as Second, the meaning in the relational context was interpreted as experiences with parents, husbands, children, interpersonal relationships, and religion. Third, the meaning in the life context is the lack of care, the reproduction of control, the responsibility as the eldest daughter, the precious family, and the meaning and value of life is the present experience in which the various experiences with the parents in the past affect the lives of the current participants. interpreted in Through the above research results, this study aims to describe the experiences of insecure attachment during childhood and the post-traumatic growth process of participants using a narrative technique, and to suggest positive alternatives to their lives.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.11
• /
• pp.53-60
• /
• 2009

This paper is focusing on the optimal size of internal space in which various functions are expressed by means of analogizing appropriate size of passage fitted to characters and reverse-calculating it in comparison with preexisting games, and thus increase the objectivity of data. To this end, the work of making the data objective was conducted in advance, and the experiment proceeded in the way that specially-designed Dungeon can make its way through the passage and in the situation where small-scale combats take place. In addition, the efforts were made to standardize the outcome of experiments by restricting the types of game graphics to MMO RPG. Further, the enhance the objectivity of size of the characters, the game types were limited to RPG games, and it is also adjusted to the character's face size based on the graphic designer's preference. The size of shoulders and arms of characters, inter alia, was selected as distinguishing points to be adjusted to the passage. By analyzing these data, the size of passage was re-organized according to the main characters in the MMO RPG games. Four experimental data was utilized by the 200 game experts in order to select the optimal size of passage in the game. As a consequence, the proportional correlation between character and passage movement was evaluated as successful in terms of emotional recognition.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.10
• /
• pp.63-70
• /
• 2018

Biomedical signal measurement technology using images has been developed, and researches on respiration signal measurement technology for maintaining life have been continuously carried out. The existing technology measured respiratory signals through a thermal imaging camera that measures heat emitted from a person's body. In addition, research was conducted to measure respiration rate by analyzing human chest movement in real time. However, the image processing using the infrared thermal image may be difficult to detect the respiratory organ due to the external environmental factors (temperature change, noise, etc.), and thus the accuracy of the measurement of the respiration rate is low.In this study, the images were acquired using visible light and infrared thermal camera to enhance the area of the respiratory tract. Then, based on the two images, features of the respiratory tract region are extracted through processes such as face recognition and image matching. The pattern of the respiratory signal is classified through the k-nearest neighbor classifier, which is one of the statistical classification methods. The respiration rate was calculated according to the characteristics of the classified patterns and the possibility of breathing rate measurement was verified by analyzing the measured respiration rate with the actual respiration rate.

• Journal of the Korean Society for Library and Information Science
• /
• v.57 no.4
• /
• pp.333-351
• /
• 2023

This study investigates user behavior in library spaces through the lens of AI camera analytics. By leveraging the face recognition and tracking capabilities of AI cameras, we accurately identified the gender and age of visitors and meticulously collected video data to track their movements. Our findings revealed that female users slightly outnumbered male users and the dominant age group was individuals in their 30s. User visits peaked between Tuesday to Friday, with the highest footfall recorded between 14:00 and 15:00 pm, while visits decreased over the weekend. Most visitors utilized one or two specific spaces, frequently consulting the information desk for inquiries, checking out/returning items, or using the rest area for relaxation. The library stacks were used approximately twice as much as they were avoided. The most frequented subject areas were Philosophy(100), Religion(200), Social Sciences(300), Science(400), Technology(500), and Literature(800), with Literature(800) and Religion(200) displaying the most intersections with other areas. By categorizing users into five clusters based on space utilization patterns, we discerned varying objectives and subject interests, providing insights for future library service enhancements. Moreover, the study underscores the need to address the associated costs and privacy concerns when considering the broader application of AI camera analytics in library settings.

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.81-89
• /
• 2023

In modern society, blind people face difficulties in navigating common environments such as sidewalks, elevators, and crosswalks. Research has been conducted to alleviate these inconveniences for the visually impaired through the use of visual and audio aids. However, such research often encounters limitations when it comes to practical implementation due to the high cost of wearable devices, high-performance CCTV systems, and voice sensors. In this paper, we propose an artificial intelligence fusion algorithm that utilizes low-cost video sensors integrated into smartphones to help blind people safely navigate their surroundings during walking. The proposed algorithm combines motion capture and object detection algorithms to detect moving people and various obstacles encountered during walking. We employed the MediaPipe library for motion capture to model and detect surrounding pedestrians during motion. Additionally, we used object detection algorithms to model and detect various obstacles that can occur during walking on sidewalks. Through experimentation, we validated the performance of the artificial intelligence fusion algorithm, achieving accuracy of 0.92, precision of 0.91, recall of 0.99, and an F1 score of 0.95. This research can assist blind people in navigating through obstacles such as bollards, shared scooters, and vehicles encountered during walking, thereby enhancing their mobility and safety.

• Journal of Service Research and Studies
• /
• v.14 no.1
• /
• pp.13-26
• /
• 2024

Multi-modal generation is the process of generating results based on a variety of information, such as text, images, and audio. With the rapid development of AI technology, there is a growing number of multi-modal based systems that synthesize different types of data to produce results. In this paper, we present an AI system that uses speech and text recognition to describe a person and generate a montage image. While the existing montage generation technology is based on the appearance of Westerners, the montage generation system developed in this paper learns a model based on Korean facial features. Therefore, it is possible to create more accurate and effective Korean montage images based on multi-modal voice and text specific to Korean. Since the developed montage generation app can be utilized as a draft montage, it can dramatically reduce the manual labor of existing montage production personnel. For this purpose, we utilized persona-based virtual person montage data provided by the AI-Hub of the National Information Society Agency. AI-Hub is an AI integration platform aimed at providing a one-stop service by building artificial intelligence learning data necessary for the development of AI technology and services. The image generation system was implemented using VQGAN, a deep learning model used to generate high-resolution images, and the KoDALLE model, a Korean-based image generation model. It can be confirmed that the learned AI model creates a montage image of a face that is very similar to what was described using voice and text. To verify the practicality of the developed montage generation app, 10 testers used it and more than 70% responded that they were satisfied. The montage generator can be used in various fields, such as criminal detection, to describe and image facial features.

• Asia pacific journal of information systems
• /
• v.22 no.3
• /
• pp.53-74
• /
• 2012

A challenge in fostering virtual communities is the continuous supply of knowledge, namely members' willingness to contribute knowledge to their communities. Previous research argues that giving away knowledge eventually causes the possessors of that knowledge to lose their unique value to others, benefiting all except the contributor. Furthermore, communication within virtual communities involves a large number of participants with different social backgrounds and perspectives. The establishment of mutual understanding to comprehend conversations and foster knowledge contribution in virtual communities is inevitably more difficult than face-to-face communication in a small group. In spite of these arguments, evidence suggests that individuals in virtual communities do engage in social behaviors such as knowledge contribution. It is important to understand why individuals provide their valuable knowledge to other community members without a guarantee of returns. In virtual communities, knowledge is inherently rooted in individual members' experiences and expertise. This personal nature of knowledge requires social interactions between virtual community members for knowledge transfer. This study employs the social capital theory in order to account for interpersonal relationship factors and identity theory for individual and group factors that may affect knowledge contribution. First, social capital is the relationship capital which is embedded within the relationships among the participants in a network and available for use when it is needed. Social capital is a productive resource, facilitating individuals' actions for attainment. Nahapiet and Ghoshal (1997) identify three dimensions of social capital and explain theoretically how these dimensions affect the exchange of knowledge. Thus, social capital would be relevant to knowledge contribution in virtual communities. Second, existing research has addressed the importance of identity in facilitating knowledge contribution in a virtual context. Identity in virtual communities has been described as playing a vital role in the establishment of personal reputations and in the recognition of others. For instance, reputation systems that rate participants in terms of the quality of their contributions provide a readily available inventory of experts to knowledge seekers. Despite the growing interest in identities, however, there is little empirical research about how identities in the communities influence knowledge contribution. Therefore, the goal of this study is to better understand knowledge contribution by examining the roles of social capital and identity in virtual communities. Based on a theoretical framework of social capital and identity theory, we develop and test a theoretical model and evaluate our hypotheses. Specifically, we propose three variables such as cohesiveness, reciprocity, and commitment, referring to the social capital theory, as antecedents of knowledge contribution in virtual communities. We further posit that members with a strong identity (self-presentation and group identification) contribute more knowledge to virtual communities. We conducted a field study in order to validate our research model. We collected data from 192 members of virtual communities and used the PLS method to analyse the data. The tests of the measurement model confirm that our data set has appropriate discriminant and convergent validity. The results of testing the structural model show that cohesion, reciprocity, and self-presentation significantly influence knowledge contribution, while commitment and group identification do not significantly influence knowledge contribution. Our findings on cohesion and reciprocity are consistent with the previous literature. Contrary to our expectations, commitment did not significantly affect knowledge contribution in virtual communities. This result may be due to the fact that knowledge contribution was voluntary in the virtual communities in our sample. Another plausible explanation for this result may be the self-selection bias for the survey respondents, who are more likely to contribute their knowledge to virtual communities. The relationship between self-presentation and knowledge contribution was found to be significant in virtual communities, supporting the results of prior literature. Group identification did not significantly affect knowledge contribution in this study, inconsistent with the wealth of research that identifies group identification as an important factor for knowledge sharing. This conflicting result calls for future research that examines the role of group identification in knowledge contribution in virtual communities. This study makes a contribution to theory development in the area of knowledge management in general and virtual communities in particular. For practice, the results of this study identify the circumstances under which individual factors would be effective for motivating knowledge contribution to virtual communities.