• Proceedings of the Korea Contents Association Conference
• /
• 2011.05a
• /
• pp.415-416
• /
• 2011

This paper describes about a technique of generating music from the natural motion of fish which are detected via the running averaging background subtraction method. The motion of the fish will create musical notes on a background frame which will be analyzed and played by a music playing module that is proposed in this paper called "PhysicX". This module is also capable of interacting with the fishes. in the tank.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.11a
• /
• pp.596-601
• /
• 2001

본 논문에서는 두 대의 카메라로부터 입력받은 인간의 신체와 같은 관절체의 움직임을 Color Marker의 색상 특성을 이용하여 3차원 공간 정보를 인식하는 방법에 관해 기술한다. 이 방법은 인체에 물리적인 장치를 하지 않고 단순히 영상정보만을 이용하여 3차원 정보를 구하였다. 보다 정확한 인체의 특징점을 구하기 위해 환 논문에서는 신체의 관절각에 칼라 마커를 부착하고 칼라 마커틀을 세그먼트하여 관절각의 2차원 정보를 구한 다음 스테레오 기하(Stereo Geometry)를 이용하여 3차원 정보를 계산하고 가상 공간상에서 인간의 움직임을 추적하는 방법을 제안한다. 제안하는 방법은 제스쳐 인식과 3차원 Virtual Reality 인터페이스 시스템 구성 등에 사용될 수 있다.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2008.04a
• /
• pp.135-138
• /
• 2008

Many Victims and property damage are caused in fires. In this paper, an flame detection algorithm is proposed to early alarm fires. The proposed flame detection algorithm is based on 2-stage decision strategy of video processing. The first decision is to check with color distribution of input vidoe. In the second, the candidated region is settled as fire region with activity. As a result of simulation, it is shown that the proposed algorithm is useful for fire recognition.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.4 no.2
• /
• pp.15-19
• /
• 2008

In this paper, we propose a new algorithm of image depth detection using motion estimation and object tracking. In industry, robots are used for automobile, conveyer system, etc. But, these have much necessary time. Thus, in this paper, we develop the efficient method of image depth detection based on motion estimation and object tracking.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.581-583
• /
• 2023

기존의 불 영역 검출을 위한 이미지 전처리 기법은 임계값에 영향을 많이 받기 때문에 사진의 색상 분포에 의존하는 결과가 나온다. 이를 보완하고자 본 논문에서는 화염이 가지고 있는 특유의 비선형적이고 불규칙한 움직임을 추출할 수 있는 방법을 제안한다. 유체 시뮬레이션에서 활용되는 준 라그랑지안(Semi-Lagrangian) 활용하여 유동을 분석하고 이를 기반으로 영상에서 화염의 움직임을 추출한다.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.4
• /
• pp.507-520
• /
• 2001

In this paper, a hybrid algorithm for well detecting both abrupt and gradual scene changes is proposed. This algorithm examines only the candidate intervals for speedup using the binary tree method and skips the intervals that are not candidate. For accuracy, the temporal difference of variance is used to detect the gradual scene changes while the temporal difference of histogram is used to detect the abrupt scene changes. Experimental results show that the proposed hybrid algorithm using the binary tree method works up about 10 times faster that the sequential method and is effective in detecting abrupt scene change and gradual transitions including dissolving and fading.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.06a
• /
• pp.254-257
• /
• 2014

Recently, various applications using real-time face detection have been developed as face recognition technology and hardware grows. While network service is developing and video instruments costs lower, it is needed that smart surveillance camera and service using network camera based on IP and face detection technology. However, videos should be compressed for reducing network bandwidth and storage capacity in surveillance system. As it requires high-level improvement of system performance when all the compressed frames are processed in a face detection program, fast face detection method is needed. In this paper, not only a fast way of algorithm using Haar like features and adaboost learning and motion information but also an application on broadcast system is suggested.

• The KIPS Transactions:PartB
• /
• v.12B no.5 s.101
• /
• pp.561-566
• /
• 2005

This paper presents a visual tracking algorithm for the golf swing motion analysis by using the information of the pixels of video frames and movement of the golf club to solve the problem fixed center point in model based tracking method. The model based tracking method use the polynomial function for trajectory displaying of upswing and downswing. Therefore it is under the hypothesis of the no movement of the center of gravity so this method is not for the amateurs. we proposed method using the information of pixel and movement, we first detected the motion by using the information of pixel in the frames in golf swing motion. Then we extracted the club head and hand by a properties of club shaft that consist of the parallel line and the moved location of club in up-swing and down-swing. In addition, we can extract the center point of user by tracking center point of the line between center of head and both foots. And we made an experiment with data that movement of center point is big. Finally, we can track the real trajectory of club head, hand and center point by using proposed tracking algorithm.

• Journal of the Korean Society for Precision Engineering
• /
• v.27 no.2
• /
• pp.145-152
• /
• 2010

Gait phase detection is important for evaluating the recovery of gait ability in patients with paralysis, and for determining the stimulation timing in FES walking. In this study, three different motion sensors(tilt sensor, gyrosensor and accelerometer) were used to detect gait events(heel strike, HS; toe off, TO) and they were compared one another to determine the most applicable sensor for gait phase detection. Motion sensors were attached on the shank and heel of subjects. Gait phases determined by the characteristics of each sensor's signal were compared with those from FVA. Gait phase detections using three different motion sensors were valid, since they all have reliabilities more than 95%, when compared with FVA. HS and TO were determined by both FVA and motion sensor signals, and the accuracy of detecting HS and TO with motion sensors were assessed by the time differences between FVA and motion sensors. Results show of that the tilt sensor and the gyrosensor could detect gait phase more accurately in normal subjects. Vertical acceleration from the accelerometer could detect HS most accurately in hemiplegic patient group A. The gyrosensor could detect HS and TO most accurately in hemiplegic patient group A and B. Valid error ranges of HS and TO were determined by 3.9 % and 13.6 % in normal subjects, respectively. The detection of TO from all sensor signals was valid in both patient group A and B. However, the vertical acceleration detected HS validly in patient group A and the gyrosensor detected HS validly in patient group B. We could determine the most applicable motion sensors to detect gait phases in hemiplegic patients. However, since hemiplegic patients have much different gait patterns one another, further experimental studies using various simple motion sensors would be required to determine gait events in pathologic gaits.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.5 s.311
• /
• pp.38-47
• /
• 2006

Conventional classification method uses a single classifier based on shape or motion feature. However this method exhibits a weakness if naively used since the classification performance is highly sensitive to the accuracy of moving region to be detected. The detection accuracy, in turn, depends on the condition of the image background. In this paper, we propose to resolve the drawback and thus strengthen the classification reliability by employing a Bayesian decision fusion and by optimally combining the decisions of three classifiers. The first classifier is based on shape information obtained from Fourier descriptors while the second is based on the shape information obtained from image gradients. The third classifier uses motion information. Our experimental results on the classification Performance of human and vehicle with a static camera in various directions confirm a significant improvement and indicate the superiority of the proposed decision fusion method compared to the conventional Majority Voting and Weight Average Score approaches.