• 한국항해항만학회지
• /
• 제42권4호
• /
• pp.277-282
• /
• 2018

최근 해상교통량이 증가하고 선박교통 관제구역이 확대됨에 따라 관제사의 업무 부하가 증가하고 있으며, 이로 인해 교통량이 급증하는 경우 관제사가 위험을 인지하지 못하는 상황도 발생하게 된다. 이러한 배경에서 본 논문에서는 관제 업무의 지원을 위해 이상 거동 선박을 자동으로 식별하는 방법을 제안한다. 본 방법은 누적된 AIS 데이터를 이용하여 관제구역 내의 통항 패턴을 학습하고, 학습된 모델과의 비교를 통해 이상치를 계산하여 이상 거동 선박을 식별한다. 특히, 선박의 거동 상태에 대한 분류 정보가 없더라도 비지도 학습법을 기반으로 항적 데이터를 자동으로 분류하여 통항 패턴을 학습할 수 있으며, 항적의 군집화와 분류 과정을 통해 이상 거동 선박을 실시간으로 식별할 수 있는 특징을 가진다. 또한, 본 논문에서는 선박운항 시뮬레이터 및 실제 AIS 항적 데이터를 이용한 식별 실험을 수행하였으며, 이를 통해 선박교통관제 시스템에의 활용 가능성을 고찰하였다.

• 정보과학회 논문지
• /
• 제43권8호
• /
• pp.848-856
• /
• 2016

해상 안전을 위한 선박의 탐색 및 식별은 매우 중요하다. 선박의 탐색은 레이더로 가능하나, 식별은 선박자동식별장치, 통신장비, 시각 등에 의해 이루어지며, 이러한 식별수단이 불능 시 레이더 운용자의 경험과 지식을 바탕으로 선박의 기동특성을 참고하여 식별하는 매우 어려운 경우가 발생한다. 본 논문에서는 지속적인 관찰임무를 수행해야 할 선박 탐색요원의 임무를 보조하기 위하여 레이더 상 선박의 기동특성을 이용, 자동식별 및 사고발생 가능성을 탐지하는 방법을 제안한다. 4가지 유형의 선박 정보, 레이더 상 접촉거리 및 침로, 속력을 이용하여 그 특징을 추출하고, SVM을 활용하여 식별 정확도를 평가하였으며, 이를 이용한 자동식별 알고리즘을 통해 사고발생 가능성이 있는 선박을 선별하는 방법을 제시하였다. 실험 결과 90% 이상의 식별 정확도를 보였으며, 실제 사고선박인 세월호의 정보를 자동식별 알고리즘에 적용하여 선별 가능함을 보였다. 이 방법은 다양한 상황에서 선박 탐색요원의 경험과 지식을 효과적으로 보완하고, 다수의 선박 중 관심필요선박을 사전 식별하여 정보를 제공함으로서 탐색요원의 노력을 경감시키고, 문제점을 보다 빨리 인지하는데 도움이 될 것이다.

• 농업과학연구
• /
• 제31권2호
• /
• pp.161-174
• /
• 2004

This study is aimed at identifying China's land operating system in the great leap periods. the findings are as follows. For one thing, though the landownership in China was basically performed in public, a portion of private management was allowed. The more this kind of private management allowed, the more the life standard of farmers enhanced. For another thing, the conversion from agricultural cooperatives to people's commune happened swiftly, and the problems according to the conversion arose instantly, which made the operation system changed partially. The last, but not the least, even in the point of the rapid communistic movement, private management was locally accepted in the name of three self and one private management. Besides, as we can recognize from the case study of Daichai village, the operation of a large working group and a small working group was done in political context.

• 농촌지도와개발
• /
• 제8권2호
• /
• pp.191-199
• /
• 2001

The objectives of this study was to explore the concept, the origin and history, and classification of amenity. Amenity is a new approach about nature, environment and the world. The concept, origin and classification of amenity was widely diversified, however, the concept of amenity was inclusive and abstract, so amenity was easier to recognize than to define. Even though many people were interested in amenity, however, the movement or practice of amenity in Korea was still in infancy stage. This paper was an exploratory study of amenity and more detailed further studies on amenity should conducted in Korea. It was suggested that we should not start on philosophical question of amenity. We should rather start on practical problems of how to put into practice to insure quality of life in the community, to conserve nature, to study practical dimensions of amenity, and to develop indicators to measure amenity.

• 한국멀티미디어학회논문지
• /
• 제19권5호
• /
• pp.932-938
• /
• 2016

Recently, interactive media which maximizes audience engagement by making the audience appeal on a stage in digital media environment has been distributed more widely. In fact, there has been active movement to develop and promote a new participatory media genre with higher immersion by applying this kind of interactive media concept to advertisement, film, game and e-learning. In the conventional interactive media, digital media had to be enjoyed in particular environment where diverse sensors were installed or through a certain device to recognize a user's motion and voice. This study attempted to design and implement an image creation system which ensures interactions between a user and media in popular distribution-enabled web environment and through PC and smart devices to minimize the image producer-user constraints.

• Journal of Multimedia Information System
• /
• 제7권2호
• /
• pp.115-124
• /
• 2020

This paper is a study of deep learning among artificial intelligence technology which has been developing many technologies recently. Especially, I am talking about emotional computing that has been mentioned a lot recently during deep learning. Emotional computing, in other words, is a passive concept that is dominated by people who scientifically analyze human sensibilities and reflect them in product development or system design, and a more active concept that studies how devices and systems understand humans and communicate with people in different modes. This emotional signal extraction, sensitivity, and psychology recognition technology is defined as a technology to process, analyze, and recognize psycho-sensitivity based on micro-small, hyper-sensor technology, and sensitive signals and information that can be sensed by the active movement of the autonomic nervous system caused by human emotional changes in everyday life. Chapter 1 talks about overview and Chapter 2 shows related research. Chapter 3 shows the problems and models of real emotional computing and Chapter 4 shows this paper as a conclusion.

• 한국운동역학회지
• /
• 제16권3호
• /
• pp.183-188
• /
• 2006

This research is to know differentiation of CP(center of pressure) pattern among four pitching sort(straight ball, raise ball, change up ball, drop ball). Subject are three national or junior athletes. We use the one camera, Novel Win pressure measurement system. Conclusions are as follows : 1. When we throw the straight ball, CP of left foot is effective to end movement at middle of foot in body balance on arm angular motion and enhanced speed. 2. When we throw the raise ball, to change CP from middle to post is more effective in order to raise the ball. 3.In drop ball pitching, in order to fall down the ball in front of hitter, CP of left foot move from post foot to interior part of forefoot 4. In change up ball pitching, if CP of left foot move into forefoot, it is a cause of high ball and hitter can recognize the change up ball because of late arm rotation motion.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2009년도 정보 및 제어 심포지움 논문집
• /
• pp.273-275
• /
• 2009

The currently used methods to get this information include corneal reflection, electro-oculograph, and so on. In this paper, we propose an algorithm that is developed for the reflection of pupil by infrared. I assume that the pupil is a perfect circle in the captured eye images. The method to recognize an existing image has many data throughput very much. And the time to deal with an image data is very long. So, we did not use the camera. However we studied the algorithm to pursue the pupil to a mapping technique.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.312-314
• /
• 2006

Researches on biped robot walking have been mostly focusing on walking on even surfaces. Therefore, robot walking has been only realized on pre-specified spaces with pre-specified movements according to the previous researches. In this paper a walking system for a biped robot using fuzzy system and neural networks to overcome those constraints. The system enables biped walking to be possible in various environments and with more complicated obstacels. For the purpose, a walking robot should recognize its surrounding environment and determine its movement. In the proposed system, a robot dynamically generates its walking trajectories of each joint by using neural networks when facing new obstacle such as stairs, and it maintains its walking stability by using closed loop fuzzy control system which manipulates the waist joints.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 합동 추계학술대회 논문집 정보 및 제어부문
• /
• pp.495-500
• /
• 2002

Recently, a service robot industry outstands as an up and coming industry of the next generation. Specially, there are so many research in self-steering movement(SSM). In order to implement SSM, robot must effectively recognize all around, detect objects and make a surrounding map with sensors. So, many robots have a sonar and a infrared sensor, etc. But, in these sensors, We only know informations about between the robot and the object as well as resolution faculty is of inferior quality. In this paper, we will introduce new algorithm that recognizes objects around robot and makes a two dimension surrounding map with a omni-direction vision camera and two stereo vision cameras.