• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.923-926
• /
• 2011

본 연구에서는 움직이는 물체가 있는 비디오에서 검출된 전경 영상(실루엣)을 토대로 사람을 추적하고 추적된 사람의 실루엣 형상을 통하여 활동성을 인식하는 실시간 감시 시스템에 적용 가능한 사람의 행동을 인식하고 분석하고자 한다. 전경에서 블랍(사람)을 검출하는 방법은 기존에 연구했던 차영상을 이용하였고, 검출된 블랍을 대상으로 사람임을 판단하고 사람인 경우 검출된 블랍의 실루엣을 이용한 기존의 자세 추정 기법에 추가적으로 4가지 특징들을 추가하여 사람의 행동을 분석한다. 각 파라미터들은 임계치를 통하여 구분하였다. 본 논문에서는 사람의 행동은 크게 네 가지의 경우로 {Standing, Bending/Crawling, Laying down, Sitting} 분류한다. 제안된 특징 파라미터들을 추가한 방법은 기존의 실루엣 기반의 자세 추정 기법만을 사용하는 것보다 좀더 높은 인식율을 보여주었다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.4B
• /
• pp.734-742
• /
• 2000

This paper presents a new volumetric approach to 3D object recognition by using PBSM (part-based superquadric model). The assembly part object can be constructed with the set of volumetric primitives and the relationships between them. We describe volumetric characteristics of the model object with superquadric parameters. In addition, our model base has the relationships between volumetric primitives as well as the surface information : the surface type, the junction type between neighboring surfaces. These surface properties and relationships between parts are effectively used in recognition process. Our integrated method is robust to recognition of the identity, position, and orientation of randomly oriented assembly parts. Furthermore, we can reduce the effects of self-occlusion and non-linear shape changes according to viewpoint. In this paper, we show that our integrated method is robust to recognition of the identity, position, and orientation of randomly oriented assembly parts through experimental results.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.10
• /
• pp.116-122
• /
• 2016

This paper proposes a practical object shape reconstruction method using an underwater imaging sonar. In order to reconstruct the object shape, three methods are utilized. Firstly, the vertical field of view of imaging sonar is modified to narrow angle to reduce an uncertainty of estimated 3D position. The wide vertical field of view makes the incorrect estimation result about the 3D position of the underwater object. Secondly, simple noise filtering and range detection methods are designed to extract a distance from the sonar image. Lastly, a low pass filter is adopted to estimate a probability of voxel occupancy. To demonstrate the proposed methods, object shape reconstruction for three sample objects was performed in a basin and results are explained.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.4
• /
• pp.450-456
• /
• 2014

Humanoid robot is the most suitable robot platform for effective human and robot interaction. However, the robot's complicated body structure containing more than twenty joint actuators makes it difficult to generate stable and elaborate postures using the conventional inverse kinematic method. This paper proposes an alternative approach to generate difficult postures of touching an object placed in front of the foot by the left or right hand with its torso bent forward in single support phase using the fast computational optimization method, particle swarm optimization. The simulated postures are also applied to a commercial humanoid robot platform, which validates the feasibility of the proposed approach.

• Progress in Medical Physics
• /
• v.2 no.2
• /
• pp.161-173
• /
• 1991

A statistical approach was carried out to analyze joint moments acting on the six subjects performing a right-handed seated task. The dynamic task analyzed consisted of moving a hand-held weight of lkg mass back and forth in front of a subject's chest at the shoulder level in an upright seated position. We used experimental data obtained in the Biomechanics Laboratory of the University of Michigan. Based on the acquired data from three trials by each subject, moments were calculated using a 3-dimensional biomechanical model at such articulations as wrist, elbow, shoulder, the third lumbar spine, hip, knee, and ankle joints. The linear correlation and the two way analysis of variance were applied to the calculated joint moments in order to investigate inter-subject and inter-trial varations. The results obtained showed that the largest magnitude and deviation of moment was found at the third lumbar spine, that any linear relationship could not be found between moment and its equivalents attempted in this study, and that the maximum value and deviation of moment acting on each joint were statistically the same for all three trials but those were statistically not the same for all six subjects.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.6
• /
• pp.36-42
• /
• 2009

In the mobile device environment, the context-aware computing has been emerging as a core technology of ubiquitous computing. Compared with a desktop computer, a user interface and resource of mobile device is very limited. Traditional desktop-based user interface has been developed on the basis that a user's activity is static state. In contrast, mobile devices are not able to utilize representative desktop-based interaction mechanisms such as a keyboard and mouse, not only because the activity of a user is dynamic state, but mobile devices have limited resources and small LCD display. In this paper, we introduce an intelligent control system for the mobile device that can utility effectively the limited resource and complement the poor user interface by using an accelerometer being able to sense the physical activity and posture. The proposed system can estimate the user activity, static and dynamic states, and posture watching the PDA at the same time, and the proposed intelligent control system as its application, the backlight ON/OFF on the PDA, is run by the result of the user's behavior.

• The KIPS Transactions:PartB
• /
• v.14B no.3 s.113
• /
• pp.171-182
• /
• 2007

We propose a simple method for tracking a nonparameterized subject contour in a single video stream with a moving camera and changing background. Then we present a method to eliminate the tracked contour object by replacing with the background scene we get from other frame. First we track the object using LOD (Level-of-Detail) canny edge maps, then we generate background of each image frame and replace the tracked object in a scene by a background image from other frame that is not occluded by the tracked object. Our tracking method is based on level-of-detail (LOD) modified Canny edge maps and graph-based routing operations on the LOD maps. We get more edge pixels along LOD hierarchy. Our accurate tracking is based on reducing effects from irrelevant edges by selecting the stronger edge pixels, thereby relying on the current frame edge pixel as much as possible. The first frame background scene is determined by camera motion, camera movement between two image frames, and other background scenes are computed from the previous background scenes. The computed background scenes are used to eliminate the tracked object from the scene. In order to remove the tracked object, we generate approximated background for the first frame. Background images for subsequent frames are based on the first frame background or previous frame images. This approach is based on computing camera motion. Our experimental results show that our method works nice for moderate camera movement with small object shape changes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.10B
• /
• pp.1937-1945
• /
• 1999

In paper, we propose a 3D polygonal mesh simplification technique based on vertex clustering. The proposed method differentiates the size of each cluster according to the local property of a 3D object. We determine the size of clusters by considering the normal vector of triangles and the vertex distribution. The subdivisions of cluster are represented by octree. In this paper, we use the Harsdorff distance between the original mesh and the simplified one as a meaningful error value. Because proposed method adaptively determine the size of cluster according to the local property of the mesh, it has smaller error as compared with the previous methods and represent the small regions on detail. Also it can generate a multiresolution model and selectively refine the local regions.

• Journal of Broadcast Engineering
• /
• v.1 no.2
• /
• pp.176-187
• /
• 1996

This paper deals with the accurate estimation of 3- D pose (position and orientation) of a moving object with reference to the world frame (or robot base frame), based on a sequence of stereo images taken by cameras mounted on the end - effector of a robot manipulator. This work is an extension of the previous work[1]. Emphasis is given to the 3-D pose estimation relative to the world (or robot base) frame under the presence of not only the measurement noise in 2 - D images[ 1] but also the camera position errors due to the random noise involved in joint angles of a robot manipulator. To this end, a new set of discrete linear Kalman filter equations is derived, based on the following: 1) the orientation error of the object frame due to measurement noise in 2 - D images is modeled with reference to the camera frame by analyzing the noise propagation through 3- D reconstruction; 2) an extended Jacobian matrix is formulated by combining the result of 1) and the orientation error of the end-effector frame due to joint angle errors through robot differential kinematics; and 3) the rotational motion of an object, which is nonlinear in nature, is linearized based on quaternions. Motion parameters are computed from the estimated quaternions based on the iterated least-squares method. Simulation results show the significant reduction of estimation errors and also demonstrate an accurate convergence of the actual motion parameters to the true values.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.10
• /
• pp.4629-4635
• /
• 2012

This study was aimed at the relationship between peak EMG amplitude on low back muscles acting on L5/S1 and load center of gravity, trunk lateral bending while lifting an object. Musculoskeletal disorders including low back pain can occur even when handling heavy objects only once as well as when doing non-heavy materials repeatedly. 11 male subjects with average 23 age were required to lift a 15.8kg object symmetrically three times. Peak EMG amplitudes on 6 muscles related with L5/S1 were recorded and analyzed. The lifting conditions consisted of lifting symmetric load with no trunk lateral bending, asymmetric load with no trunk lateral bending, and asymmetric load with trunk lateral bending to the load center of gravity within an object. The results showed that peak EMG amplitude on back muscles contralateral to load center of gravity was observed greater in comparison with the symmetric load. Also, in case of lifting asymmetric load the posture with trunk lateral bending increased peak EMG amplitude on muscles contralateral to load center of gravity more than with no trunk lateral bending. This research can be used as one administrative intervention in order to reduce the low back pain incidence with suggesting workers that they keep the trunk not bending to load center of gravity if possible when lifting a heavy asymmetric object.