• 한국산업정보학회논문지
• /
• 제17권3호
• /
• pp.51-57
• /
• 2012

팩맨(Pacman)은 세계적으로 사랑을 받는 클래식 게임의 종결자로 알려져 있다. 본 논문에서는 클래식 게임인 팩맨을 키보드나 마우스를 사용하지 않고 간단한 손동작만으로 게임을 플레이 할 수 있도록 한다. 즉, 손의 좌표 중심점을 이용하여 방향키를 대신할 수 있도록 한 모션을 이용한 게임이다. 또한 캠을 이용하여 MFC 다이얼로그에 영상을 받아 손에 대한 포인터를 추출하여 손의 움직임에 따라 몬스터의 움직임도 바뀌도록 한다. 본 논문에서 피부색 추출을 위해 RGB 영상을 YCbCbr영상으로 변환하고, 보다 좋은 영상을 얻기 위하여 곱 연산과 하이브리드 미디언 필터링을 이용하였다. 그리고 손 영역의 무게 중심을 구하여 이를 중심으로 손 영역의 움직임을 구하는 방법을 사용하였다.

• 디지털콘텐츠학회 논문지
• /
• 제13권3호
• /
• pp.271-278
• /
• 2012

적외선 기반 깊이 센서는 최근 저렴해진 가격으로 인해 소비자용으로 널리 보급되고 있으며 원래 목적을 넘어서 방송용 가상 스튜디오 제스처 인식을 포함한 다양한 분야로까지 적용 범위를 확대하고 있다. 그러나 이러한 방송 스튜디오 환경에서는 깊이 센서와 간섭을 일으키는 적외선이 다량 방출되어 올바른 깊이 정보의 포착이 불가능한 문제가 발생한다. 본 논문에서는 특정 적외선 파장대를 사용하는 깊이 센서가 적외선 방출 광원 하에서 간섭이 발생하는 원리에 대해 분석하고 깊이 센서의 올바른 작동을 보장하기 위한 필터링 기법을 논의한다. 또한 여러 차단 주파수대를 가지는 통과 필터를 적용하는 실험방법과 그 결과를 제시하며 올바른 대역 통과 필터를 적용함으로써 조명에서 방출되는 적외선을 차단하고 효과적으로 깊이 정보를 포착할 수 있다는 사실을 실험적으로 증명한다.

• 센서학회지
• /
• 제25권3호
• /
• pp.196-201
• /
• 2016

Rehabilitation exercises are the treatments designed to help patients who are in the process of recovery from injury or illness to restore their body functions back to the original status. However, many patients suffering from chronic diseases have found difficulties visiting hospitals for the rehabilitation program due to lack of transportation, cost of the program, their own busy schedules, etc. Also, the program usually contains a few medical check-ups which can cause patients to feel uncomfortable. In this paper, we develop a remote rehabilitation system with bio-signals by a stereo camera. A Kinect stereo camera manufactured by Microsoft corporation was used to recognize the body movement of a patient by using its infrared(IR) camera. Also, we detect the chest area of a user from the skeleton data and process to gain respiratory status. ROI coordinates are created on a user's face to detect photoplethysmography(PPG) signals to calculate heart rate values from its color sensor. Finally, rehabilitation exercises and bio-signal detecting features are combined into a Windows application for the cost effective and high performance remote rehabilitation system.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2016년도 추계학술발표대회
• /
• pp.510-513
• /
• 2016

Detecting of the object in image processing is substantial but it depends on the object itself and the environment. An object can be detected either by its shape or color. Color is an essential for pattern recognition and computer vision. It is an attractive feature because of its simplicity and its robustness to scale changes and to detect the positions of the object. Generally, color of an object depends on its characteristics of the perceiving eye and brain. Physically, objects can be said to have color because of the light leaving their surfaces. Here, we conducted experiment in the aquarium fish tank. Different color of fish robots are mimic the natural swim of fish. Unfortunately, in the underwater medium, the colors are modified by attenuation and difficult to identify the color for moving objects. We consider the fish motion as a moving object and coordinates are found at every instinct of the aquarium to detect the position of the fish robot using OpenCV color detection. In this paper, we proposed to identify the position of the fish robot by their color and use the position data to control the fish robot gathering in one point in the fish tank through serial communication using RF module. It was verified by the performance test of detecting the position of the fish robot.

• 음성과학
• /
• 제11권2호
• /
• pp.193-210
• /
• 2004

In this paper, we develop a real time lip-synch system that activates 2-D avatar's lip motion in synch with an incoming speech utterance. To realize the 'real time' operation of the system, we contain the processing time by invoking merge and split procedures performing coarse-to-fine phoneme classification. At each stage of phoneme classification, we apply the support vector machine (SVM) to reduce the computational load while retraining the desired accuracy. The coarse-to-fine phoneme classification is accomplished via two stages of feature extraction: first, each speech frame is acoustically analyzed for 3 classes of lip opening using Mel Frequency Cepstral Coefficients (MFCC) as a feature; secondly, each frame is further refined in classification for detailed lip shape using formant information. We implemented the system with 2-D lip animation that shows the effectiveness of the proposed two-stage procedure in accomplishing a real-time lip-synch task. It was observed that the method of using phoneme merging and SVM achieved about twice faster speed in recognition than the method employing the Hidden Markov Model (HMM). A typical latency time per a single frame observed for our method was in the order of 18.22 milliseconds while an HMM method applied under identical conditions resulted about 30.67 milliseconds.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2011년도 추계학술대회
• /
• pp.154-157
• /
• 2011

인간의 동작 인식은 가상 현실 시스템 및 게임 등에 적용할 수 있는 컴퓨터 비전 분야의 요소 기술 중 하나로써, 최근까지 그 연구과 활발히 진행되고 있다. 본 논문에서는 빠르고 정확한 동작 인식을 위해, 실루엣과 모션 특징이 결합된 방법을 제안한다. 제안하는 방법은 전역 특징을 이용한 후보 동작 선정 및 지역 특징을 이용한 검증 2 단계로 구성된다. 전역 특징은 Motion History Image의 Hu 모멘트를 이용해 계산되며, 후보 동작의 선정은 이들의 통계치를 이용해 결정한다. 한정된 후보 동작들 중 입력 동작을 정확히 인식하기 위해, 공간 및 방향성 비닝 기법으로 추출된 광류와 실루엣 특징을 지역 특징으로 이용한다. 최종 인식 결과는 Hu 모멘트 통계치와의 유사도 및 지역 특징의 학습을 통해 생성된 Support Vector Machine의 결과를 고려하여 결정된다. 제안하는 방법의 성능을 평가하기 위해, 실세계에서 사용 빈도가 높으며 동작의 변화가 큰 13 개의 제스처를 선정하여 데이터 셋을 구성하였다. 실험 결과 제안하는 방법의 연산 시간은 50 ms, 인식 정확도는 95%임을 확인하였다.

• 센서학회지
• /
• 제24권3호
• /
• pp.175-180
• /
• 2015

Worldwide, studies on intelligent vehicles for the convenience of drivers have been actively conducted as the number of cars has increased. However, vehicle convenience enabled by buttons lowers the concentration on driving and hence poses as a huge threat to the safety of the driver. The use of one of the convenient features, impaired driving auxiliary equipment, is limited because of its complex usage, and this device also hinders the front view of the driver. This paper proposes a vehicle-control device for controlling the convenient features as well as changes in speed and direction using gestures and motions of the driver. This device consists of an ultrasonic sensor for recognizing movement, an arduino for accepting signal control functions and servo and DC motors apply to various vehicle parts. Firstly, the vehicle-control device was designed using a 3D CAD program known as Solid-works based on the size of the steering wheel. Then, through simulations, a suitable length for minimizing the absorbent between ultrasonic sensors was confirmed using a program known as COMSOL Multiphysics. Finally, simulation results were verified through experiments, and the optimal size of the device was identified through the number of errors.

• Journal of Electrical Engineering and Technology
• /
• 제7권1호
• /
• pp.115-123
• /
• 2012

A video analysis system used to detect events in video streams generally has several processes, including object detection, object trajectories analysis, and recognition of the trajectories by comparison with an a priori trained model. However, these processes do not work well in a complex environment that has many occlusions, mirror effects, and/or shadow effects. We propose a new approach to a context-aware video surveillance system to detect predefined contexts in video streams. The proposed system consists of two modules: a feature extractor and a context recognizer. The feature extractor calculates the moving energy that represents the amount of moving objects in a video stream and the stationary energy that represents the amount of still objects in a video stream. We represent situations and events as motion changes and stationary energy in video streams. The context recognizer determines whether predefined contexts are included in video streams using the extracted moving and stationary energies from a feature extractor. To train each context model and recognize predefined contexts in video streams, we propose and use a new ensemble classifier based on the AdaBoost algorithm, DAdaBoost, which is one of the most famous ensemble classifier algorithms. Our proposed approach is expected to be a robust method in more complex environments that have a mirror effect and/or a shadow effect.

• 한국통신학회논문지
• /
• 제25권3B호
• /
• pp.529-540
• /
• 2000

본 논문에서는 비트율이 매우 한정된 응용에 적용되는 초저속 동영상 부호화 표준인 H.236plus에 기반한 의미기반 율-제어 기법을 제안한다. 기존의 율-제어 기법들은 의미기반이 아닌 단순히 프레임에 대해 매크로 블록 단위로 최적의 양자화 파라미터를 설정해 정해진 비트율을 조정하나, 본 논문에서는 부호화 효율을 높이기 위해 의미 영역인식 및 우선권 부여를 통해 영역별로 최적의 양자화 파라미터를 설정함으로써 제약된 비트 내에서 주관적인 화질의 향상을 보이는 기법을 제안한다. 제안하고자하는 기법의 핵심은 부호화 시 배경에 해당하는 영역에 배정되는 비트량을 할당해 전경(얼굴) 영역에 비트 자원을 더 많이 배정함으로써 주관적인 화질을 보호, 개선하기 위한 것이다.

• 한국의류학회지
• /
• 제42권3호
• /
• pp.474-485
• /
• 2018

Smart sensors and connected devices have changed the concept of garments along with IT technology convergent garments that transform the performance of basic functions. Various types of products have been researched and developed due to the increased interest in smart clothing; in addition, studies based on physical and mechanical properties have also been actively studied to improve accuracy and reliability. This study represents a basic study for the development of smart textiles based on motion recognition for the surfing practice of beginners interested in IT convergence type. A physical durability evaluation of conductive yarn according to sewing method was later carried out. This study is a conditional specimen sewn with cotton lower thread and 100mm pattern length based on the results of previous studies. The durability of the conductive yarn according to the sewing method was evaluated according to the sewing method. Durability was evaluated by two kinds of repeated strain and abrasion tests. The specimen with applied cotton in a lower thread zigzag pattern 2mm stitch size 100mm stitch length was shown to have the most suitable durability for smart textile.