• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.6C
• /
• pp.593-603
• /
• 2009

In this paper, we propose a new integrated computer vision system designed to track multiple human beings and extract their silhouette with an active stereo camera. The proposed system consists of three modules: detection, tracking and silhouette extraction. Detection was performed by camera ego-motion compensation and disparity segmentation. For tracking, we present an efficient mean shift based tracking method in which the tracking objects are characterized as disparity weighted color histograms. The silhouette was obtained by two-step segmentation. A trimap is estimated in advance and then this was effectively incorporated into the graph cut framework for fine segmentation. The proposed system was evaluated with respect to ground truth data and it was shown to detect and track multiple people very well and also produce high quality silhouettes. The proposed system can assist in gesture and gait recognition in field of Human-Robot Interaction (HRI).

• Journal of KIISE:Software and Applications
• /
• v.30 no.5_6
• /
• pp.497-502
• /
• 2003

Content-based video copy detection is a complementary approach to watermarking. As opposed to watermarking, which relies on inserting a distinct pattern into the video stream, video copy detection techniques match content-based signatures to detect copies of video. Existing typical content-based copy detection schemes have relied on image matching which is based on key frame detection. This paper proposes a motion directional histogram, which is quantized and accumulated the direction of motion, for video copy detection. The video clip is represented by a motion directional histogram as a 1-dimensional graph. This method is suitable for real time indexing and counting the TV CF verification that is high motion video clips.

• Korean Journal of Applied Biomechanics
• /
• v.14 no.3
• /
• pp.301-311
• /
• 2004

The purpose of this study was to verify the usefulness of the Mathcad program as a tool for the studying and teaching the sport biomechanics. A projectile motion was analyzed because it is the one of the most popular motion in sports activities. A 3 dimensional CG data for the high jump bar clear phase was used to calculate the initial velocity vector of the CG. Linear regression function and other functions such as cubic spline and derivative of Mathcad were used to calculate this vector. Finally, the approach angle to the bar and peak jump height was calculated. Programming in Mathcad was relatively easy compare to traditional computer language such as Fortran and C, because of the unique documentation method of Mathcad. Additionally the 2 and 3 dimensional graph function was very easy and useful to describe the mechanical data. If the use of Mathcad program is more popular in the field of sport biomechanics, it could greatly contribute to overcome the limit of research caused by the lack of proper programming ability.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.20 no.11
• /
• pp.1052-1057
• /
• 2010

Dynamic stability of rotating blade considering gravity effect is investigated in this paper. Equations of motion for the beam is derived by employing hybrid deformation variable method and transformed into dimensionless form. The present modeling method is verified by RecurDyn. Stability diagrams are presented to show the influence of the configuration of the beam and angular velocity on the dynamic stability by applying Floquet's theory. Since the natural frequencies are varied when the blade has rotating motion, it is found that relatively large unstable regions exist approximately 1.1 times as high as the first bending natural frequency and half of the sum of first and second bending natural frequency.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.2
• /
• pp.17-23
• /
• 2016

NUI is a motion interface. It uses the body of the user without the use of HID device such as a mouse and keyboard to control the device. In this paper, we use a Pi Camera and sensors connected to it with small embedded board Raspberry Pi. We are using the OpenCV algorithms optimized for image recognition and computer vision compared with traditional HID equipment and to implement a more human-friendly and intuitive interface NUI devices. comparison operation detects motion, it proposed a more advanced motion sensors and recognition systems fused connected to the Raspberry Pi.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.1
• /
• pp.53-60
• /
• 2010

We propose a motion planning method for autonomous mobile robots. In order to minimize traveling time, a smooth path and a time optimal velocity profile should be generated under kinematic and dynamic constraints. In this paper, we develop an effective and practical method to generate a good solution with lower computation time. The initial path is obtained from voronoi diagram by Dijkstra's algorithm. Then the path is improved by changing the graph and path simultaneously. We apply the dynamic programming algorithm into the stage of improvement. Simulation results are presented to verify the performance of the proposed method.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.8 no.6
• /
• pp.111-117
• /
• 2008

In this paper, we describes the boxing motion recognition which is used in the part of games, animation. To recognize the boxing motion, we have used two algorithms, one is principle component analysis, the other is dynamic time warping algorithm. PCA is the simplest of the true eigenvector-based multivariate analyses and often used to reduce multidimensional data sets to lower dimensions for analysis. DTW is an algorithm for measuring similarity between two sequences which may vary in time or speed. We introduce and compare PCA and DTW algorithms respectively. We implemented the recognition of boxing motion on the motion capture system which is developed in out research, and depict the system also. The motion graph will be created by boxing motion data which is acquired from motion capture system, and will be normalized in a process. The result has implemented in the motion recognition system with five actors, and showed the performance of the recognition.

• Journal of Korean Institute of Industrial Engineers
• /
• v.32 no.2
• /
• pp.82-90
• /
• 2006

Reconfigurability of machine tools is one of the critical factors to realize the responsive manufacturing systems to satisfy the mass customization production. This paper presents the methods to model the reconfigurable machine tools (RMTs) on Internet in response to change in the machining requirements. Specifically, the structure and motion model of machine tools using module combination rules and connectivity graph are developed. And we developed Internet-based simulator for designing of RMTs (ISD-RMT). In response to the user requirements, various structures of RMTs can be derived using the module combination rules and connectivity graph relationships. In addition, the user can verify generating structures through the control and simulation procedures.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.16 no.5
• /
• pp.128-133
• /
• 2008

In this paper, a new technique was suggested to estimate vehicle speed for the traffic accident reconstruction, and accident investigators can estimate initial vehicle speed based on this suggested technique. Turning tests with several vehicle speeds were executed and compared with the motion of the vehicle and the shape of the tire marks. A new method for estimating the coefficient of friction is suggested by using the longitudinal and lateral components of tire marks. And also, a speed calculation graph is suggested to estimate vehicle speed for traffic accident reconstruction.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.803-806
• /
• 2000

The correspondence problem is known to be difficult to solve because false positives and false negatives almost always exist in real image sequences. In this paper, we propose a robust feature tracking algorithm considering incomplete trajectories such as entering and/or vanishing trajectories. We solve the correspondence problem as the optimal graph search problem, by considering false feature points and by properly reflecting motion characteristics. The proposed algorithm finds a local optimal correspondence so that the effect of false feature points can be minimized in the decision process. The time complexity of the proposed graph search algorithm is given by O(mn) in the best case and O(m$^2$n) in the worst case, where m and n are the number of feature points in two consecutive frames. The proposed algorithm can find trajectories correctly and robustly, which has been shown by experimental results.