• 한국정밀공학회지
• /
• 제16권12호
• /
• pp.133-142
• /
• 1999

We introduce the algorithms of 2-D and 3-D position estimation using 2-D vision sensors. The sensors used in this research issue red laser slit light to the body. So, it is very convenient to obtain the coordinates of corner point or edge in sensor coordinate. Since the measured points are normally not fixed in the body coordinate, the additional conditions, that corner lines or edges are straight and fixed in the body coordinate, are used to find out the position and orientation of the body. In the case of 2-D motional body, we can find the solution analytically. But in the case of 3-D motional body, linearization technique and least mean squares method are used because of hard nonlinearity.

• 한국해양공학회지
• /
• 제21권4호
• /
• pp.45-54
• /
• 2007

In the context of auto-landing containers from a container ship to a truck or automatic guided vehicle and vice versa, this research investigates three schemes, one in Part I and two in Part II, for measuring the absolute position of a container. Coordinate transformations between the reference-coordinate, sensor-coordinate, and body-coordinate systems are briefly discussed. The scheme explored in Part I aims the use of three laser-slit sensors, which are relatively inexpensive. In this case, nine nonlinear equations are formulated for six unknown variables (three for orientation and three for position), so a closed-form solution is not available. Instead, an approximate solution through linearization was derived. An advantage of the method in Part I is its ability to measure an absolute position in 3D space, while a disadvantage is the computation time required to obtain pseudo-inverses and the approximate nature of the obtained solution. Numerical examples are provided.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 추계학술대회 논문집
• /
• pp.296-300
• /
• 1996

This paper presents 3D position estimation algorithm with the images of 2D vision sensors which issues Red Laser Slit light and recieves the line images. Since the sensor usually measures 2D position of corner(or edge) of a body and the measured point is not fixed in the body, the additional information of the corner(or edge) is used. That is, corner(or edge) line is straight and fixed in the body. For the body which moves in a plane, the Transformation matrix between the body coordinate and the reference coordinate is analytically found. For the 3D motion body, linearization technique and least mean squares method are used.

• 한국멀티미디어학회논문지
• /
• 제10권6호
• /
• pp.716-725
• /
• 2007

In this paper we describe a motion tracking algorithm for 3D human animation using stereo vision system. This allows us to extract the motion data of the end effectors of human body by following the movement through segmentation process in HIS or RGB color model, and then blob analysis is used to detect robust shape. When two hands or two foots are crossed at any position and become disjointed, an adaptive algorithm is presented to recognize whether it is left or right one. And the real motion is the 3-D coordinate motion. A mono image data is a data of 2D coordinate. This data doesn't acquire distance from a camera. By stereo vision like human vision, we can acquire a data of 3D motion such as left, right motion from bottom and distance of objects from camera. This requests a depth value including x axis and y axis coordinate in mono image for transforming 3D coordinate. This depth value(z axis) is calculated by disparity of stereo vision by using only end-effectors of images. The position of the inner joints is calculated and 3D character can be visualized using inverse kinematics.

• Journal of international Conference on Electrical Machines and Systems
• /
• 제3권4호
• /
• pp.458-464
• /
• 2014

This paper proposes the rotor position and rotor speed estimation for an interior permanent magnet synchronous machines (IPMSM) using Unscented Kalman Filter (UKF) in alpha-beta coordinate system. Conventional algorithms using UKF are based on the simple observer model of IPMSM in d-q coordinate system. Rotor acceleration is neglected within the sampling step. An expansion of the observer model in an alpha-beta coordinate system with the consideration of the rotor speed variation provides the improved rotor position and speed estimation. The results show good stability concerning the expansion of observer model for the IPMSM.

• 대한의용생체공학회:의공학회지
• /
• 제12권1호
• /
• pp.49-56
• /
• 1991

In this paper, a method for camera position estimation in gasher using elechoendoscopic image sequence is proposed. In orders to obtain proper image sequences, the gasser in divided into three sections. It Is presented thats camera position modeling for 3D information extvac lion and image distortion due to the endoscopic lenses is corrected. The feature points are represented with respect to the reference coordinate system below 10 percents error rate. The faster distortion correction algorithm is proposed in this paper. This algorithm uses error table which is faster than coordinate transform method using n -th order polynomials.

• 대한의용생체공학회:의공학회지
• /
• 제3권1호
• /
• pp.55-58
• /
• 1982

In this paper, a method for camera position estimation in gaster using elechoendoscopic image sequence is proposed. In order to obtain proper image sequences, the gaster in divided into three sections. It is presented that camera position modeling for 3D information extraction and image distortion due to the endoscopic lenses is corrected.The feature points are represented with respect to the reference coordinate system belpw 10 percents error rate. The faster distortion correction algorithm is proposed in this paper. This algorithm uses error table which is faster than coordinate transform method using n-th order polynomials.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 B
• /
• pp.1248-1250
• /
• 1996

The objective of camera calibration is to obtain the correlation between camera image coordinate and 3-D real world coordinate. Most calibration methods are based on the camera model which consists of physical parameters of the camera like position, orientation, focal length, etc and in this case camera calibration means the process of computing those parameters. In this research, we suggest a new approach which must be very efficient because the artificial neural network(ANN) model implicitly contains all the physical parameters, some of which are very difficult to be estimated by the existing calibration methods. Implicit camera calibration which means the process of calibrating a camera without explicitly computing its physical parameters can be used for both 3-D measurement and generation of image coordinates. As training each calibration points having different height, we can find the perspective projection point. The point can be used for reconstruction 3-D real world coordinate having arbitrary height and image coordinate of arbitrary 3-D real world coordinate. Experimental comparison of our method with well-known Tsai's 2 stage method is made to verify the effectiveness of the proposed method.

• 한국인터넷방송통신학회논문지
• /
• 제13권1호
• /
• pp.243-249
• /
• 2013

폐기물 환경 시설에서 폐기물 운반 차량의 폐기물 매립장 내 이동 경로와 폐기물 투척 위치를 추적하여 기록함으로써 향 후 폐기물 매립지 재사용을 위한 이력관리를 효율적으로 진행할 수 있다. 이들 폐기물 매립 위치 정보들을 3D 로 매핑하고 위치를 추출하여 최종 매립위치와 매립물 기록을 추출하기 위한 새로운 연구 방법이 다양한 실험조건하에서 구현되었다. 본 연구에서는 쓰레기 매립장 내에서의 차량의 이동경로와 매립시점을 판단할 수 있는 방법을 제시한다. 현재의 영상과 배경영상의 차를 구하여 차량의 이동경로를 추적하고 이동경로의 형태로부터 정차 여부를 판단할 수 있으며, 정차시의 영상과 배출구의 개방영상과의 비교를 통해 폐기물의 매립시점을 검출할 수 있도록 한다. 감시 카메라의 설치정보를 이용하며 매립지 평면상의 좌표와 화면좌표를 변환할 수 있는 방법을 제시한다.

• 제어로봇시스템학회논문지
• /
• 제12권8호
• /
• pp.824-833
• /
• 2006

In this paper, the image modelling of road's lane markings is established using view frustum(VF) model. From this model, a measurement system of lane markings and obstacles is proposed. The system also involve the real time processing of the 3D position coordinate and the distance data from the camera to the points on the 3D world coordinate by virtue of the camera calibration. In order to reduce their measurement error, an useful algorithm for which analyze the geometric variations due to traveling vehicle's fluctuation using VF model is proposed. In experiments, without correction, for instance, the $0.4^{\circ}$ of pitching rotation gives the error of $0.4{\sim}0.6m$ at the distance of 10m, but the more far distance cause exponentially the more error. We con finned that this algorithm can be reduced less than 0.1m of error at the same condition.