• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권7호
• /
• pp.1346-1367
• /
• 2011

As the size of camera modules is decreasing and as the computing performance of portable devices is improving, taking photos has become a part of daily life. However, existing photo management programs and products that manage such photos still require extensive user effort to facilitate the sharing and browsing of images. It is especially difficult for novice users to manage and share photos. In this paper, we develop a round-robin concept-driven user preference feedback mechanism for achieving direct photo sharing, instant display, and easy management using optimized user controls and user preference-driven classification. Compared with commercial photo management systems, our proposed solution provides new features: optimized user controls, direct sharing and instant display, and user preference feedback driven classification. These new features boost the round-robin concept-driven user preference feedback. This paper proposes a photo finder that automatically searches for photos in storage spaces or cameras. The proposed photo finder relies on user preference feedback to share photos by leveraging user preferences, and the round-robin connection transmits photos to the family's digital photo frame or web album by arbiter. The proposed method saves time and spares users the effort required for photo management. Moreover, this method does not merely direct photo sharing and simple photo management, but it also increases the satisfaction level of users viewing the photos.

• 한국정밀공학회지
• /
• 제14권7호
• /
• pp.59-66
• /
• 1997

A general procedure for a motion capture and mimic system has been delineated. Utilizing sensors operated in the magnetic fields, complicated and optimized movements are easily digitized to analyze and repreduce. The system consists of a motion capture module, a motion visualization module, a motion plan module, a motion mimic module, and a GUI module. Design concepts of the system are modular, open, and user friendly to ensure the overall system performance. Custom-built and/or off-the-shelf modules are ease- ly integrated into the system. With modifications, this procedure can be applied for complicated motion controls. This procedure is implemented on tracking a head and balancing a pole. A neural controller based on this control scheme dtilizing human motions can easily evolve from a small amount of learning data.

• Journal of information and communication convergence engineering
• /
• 제12권2호
• /
• pp.97-103
• /
• 2014

In spectrum sensing, there is a tradeoff between the probability of missed detection and the probability of a false alarm according to the value of the sensing threshold. Therefore, it is important to determine the sensing threshold suitable to the environment of cognitive radio networks. In this study, we consider a cognitive radio-based ad hoc network where secondary users directly communicate by using the same frequency band as the primary system and control their transmit power on the basis of the distance between them. First, we investigate a condition in which the primary and the secondary users can share the same frequency band without harmful interference from each other, and then, propose an algorithm that controls the sensing threshold dynamically on the basis of the transmit power of the secondary user. The analysis and simulation results show that the proposed sensing threshold control algorithm has low probabilities of both missed detection and a false alarm and thus, enables optimized spectrum sharing between the primary and the secondary systems.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 추계학술대회 논문집
• /
• pp.243-247
• /
• 1996

A general procedure for motion capture and mimic system has been delineated. Utilizing sensors operated in the magnetic fields, complicated and optimized movements are easily digitized to analyze and reproduce. The system consists of a motion capture module, a motion visualization module, a motion plan module, a motion mimic module, and a GUI module. Design concepts of the system are modular, open, and user friendly to ensure the overall system performance. This procedure is being implemented on a virtual cyber cube. and an inverted pendulum. With modifications, this procedure can applied for complicated motion controls.

• Journal of Communications and Networks
• /
• 제4권3호
• /
• pp.221-229
• /
• 2002

Adaptability is the most promising feature to be applied in future robust multimedia applications. In this paper, we propose the Direct Algorithm to improve the degree of satisfaction at heterogeneous receivers in multi-layered multicast video environments. The algorithm relies on a mechanism that dynamically controls the rates of the video layers and is based on feedback control packets sent by the receivers. The algorithm also addresses scalability issues by implementing a merging procedure at intermediate nodes in order to avoid packet implosion at the source in the case of large multicast groups. The proposed scheme is optimized to achieve high global video quality and reduced bandwidth requirements. We also propose the Direct Algorithm with a virtual number of layers. The virtual layering scheme induces intermediate nodes to keep extra states of the multicast session, which reduces the video degradation for all the receivers. The results show that the proposed scheme leads to improved global video quality at heterogeneous receivers with no cost of extra bandwidth.

• 한국지능시스템학회논문지
• /
• 제19권4호
• /
• pp.574-579
• /
• 2009

복잡한 비선형 회귀문제들에서 최적의 신경망을 구축하기 위해서는 구조의 선정 및 노이즈에 의한 과잉학습(overtraining)등에 따른 많은 문제들이 있다. 본 논문에서는 flexible incremental 알고리즘을 이용하여 단계적으로 최적의 신경망을 구축하는 방법을 제안한다. Flexible incremental 알고리즘은 예측 잔류오차를 최소화하기 위해 단계적으로 추가되어지는 은닉노드 개수를 검증데이터를 이용하여 신축성 있게 조절하고, 빠른 학습을 위하여 ELM (Extreme Learning Machine)을 이용한다. 제안된 방법은 신경망의 구축과정에서 사용자의 어떠한 관여 없이도 빠른 학습과 적은 수의 은닉노드들에 의한 범용 근사화 (universal approximation)가 가능한 신경망의 구축이 가능한 장점을 가지고 있다. 다양한 종류의 벤치마크 데이터들을 이용한 실험 결과를 통하여 제안된 방법이 실제 회귀문제들에서 우수한 성능을 가짐을 확인하였다.

• 터널과지하공간
• /
• 제34권3호
• /
• pp.218-230
• /
• 2024

최근 건설현장의 안전사고 문제를 해결하기 위해 컴퓨터 비전 기술을 활용한 안전관리에 관한 연구를 많이 수행하고 있다. 최근 딥러닝 기반 객체 인식 및 영역 분할 연구에서 앵커 박스 파라미터를 사용하고 있다. 일관적인 정확도를 확보하기 위하여 학습 과정에서 앵커 박스 파라미터의 최적화가 중요하다. 앵커 박스 관련 파라미터는 일반적으로 학습자의 휴리스틱 방법으로 모양과 크기를 고정하여 학습을 수행하고 있고, 파라미터는 단일로 구성된다. 하지만 파라미터는 객체 종류와 객체 크기에 따라 민감하고 수가 증가하면 단일 파라미터로 데이터의 모든 특성을 반영하는데 한계가 발생한다. 따라서 본 논문은 분할 학습을 통해 최적화된 다중 파라미터를 적용하는 방법을 제안하여 단일 파라미터로 모든 객체의 특성을 반영하기 어려운 문제를 해결하고자 한다. 통합 데이터를 객체 크기, 객체 수, 객체의 형상에 따라 효율적으로 분할하는 기준을 정립하였으며, 최종으로 통합 학습과 분할 학습 방법의 성능 비교를 통해 제안한 학습 방법의 효과를 검증하였다.