• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.7
• /
• pp.700-710
• /
• 2009

A computer vision technique of estimating the location of an unmanned ground vehicle is proposed. Identifying the location of the unmaned vehicle is very important task for automatic navigation of the vehicle. Conventional positioning sensors may fail to work properly in some real situations due to internal and external interferences. Given a DSM(Digital Surface Map), location of the vehicle can be estimated by the registration of the DSM and multiview range images obtained at the vehicle. Registration of the DSM and range images yields the 3D transformation from the coordinates of the range sensor to the reference coordinates of the DSM. To estimate the vehicle position, we first register a range image to the DSM coarsely and then refine the result. For coarse registration, we employ a fast random sample matching method. After the initial position is estimated and refined, all subsequent range images are registered by applying a pair-wise registration technique between range images. To reduce the accumulation error of pair-wise registration, we periodically refine the registration between range images and the DSM. Virtual environment is established to perform several experiments using a virtual vehicle. Range images are created based on the DSM by modeling a real 3D sensor. The vehicle moves along three different path while acquiring range images. Experimental results show that registration error is about under 1.3m in average.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.3
• /
• pp.35-42
• /
• 2021

In this paper, we propose a deep learning-based person re-identification method using a three-dimensional RGB-Depth Xtion2 camera considering joint coordinates and dynamic features(velocity, acceleration). The main idea of the proposed identification methodology is to easily extract gait data such as joint coordinates, dynamic features with an RGB-D camera and automatically identify gait patterns through a self-designed one-dimensional convolutional neural network classifier(1D-ConvNet). The accuracy was measured based on the F1 Score, and the influence was measured by comparing the accuracy with the classifier model (JC) that did not consider dynamic characteristics. As a result, our proposed classifier model in the case of considering the dynamic characteristics(JCSpeed) showed about 8% higher F1-Score than JC.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 2004.07a
• /
• pp.229-229
• /
• 2004

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.26 no.4
• /
• pp.423-431
• /
• 2008

The control surveying which aims at identifying the coordinate system of satellite images with that of ground is a repeatedly performed essential process to produce digital ortho - photos and it acts as the main factor to increase the production cost of the photos by duplicated budgets and redundant works when executing the projects for acquiring basic geographical information from high density satellite images. During the experimentation, an application system was established for producing a stereo image chip by the analysis of DPPDB file structure, the stereo image chip was produced with SPOT and IKONOS images, the analysis of 3D modeling accuracy was performed to secure the required accuracy and to present the optimal number and deployment of the control points, and a 3D modeling was performed for new SPOT images and lastly, 3D ground coordinates were extracted by the observation of the same points through the overlapping with the new images. As the results of the research, it is proved that the stereo image chip can be used as the ground controls through the accuracy analysis between the coordinates of the images and the ground, close results were obtained between the coordinates by the ground survey and those by the 3D modeling using new images and the observation of the same points, positional changes were not found during observing the same points, and the research presented the methodology for improving the process of the control survey by showing the availability of the image controls on the stereo image chip instead of the ground controls.

• The Journal of Korean Society for Radiation Therapy
• /
• v.15 no.1
• /
• pp.35-40
• /
• 2003

I. Purpose Confirming an error to be able to break out in a method to move couch manually while operator sees the skin marks on patient in case of curing head who got 2 targets adjoined, so we analyze coordinates price of couch, evaluate reproducibility and precision of change movements between targets. II. Materials and Methods In radiotherapy, for confirming errors in manual movements by operators by exchanging between two targets to treat patient head, we read coordinates price(vertical, longitudinal, lateral three directions of couch) shown on a monitor of LINAC( CL 2100, Varian, USA) in order to evaluate accuracy about the length that moved in time for moving couch manually. After reading movement length of coordinates recorded in three directions of all treatment, we compared distance between targets recorded in RTP(Pinnacle, ADAC, USA) with reading coordinates price of couch, setting actually done the same patient for ten times, coordinates were recorded, treated for evaluating averages and degrees of errors and standard deviations. III. Results In method to confirm skin marks of patient by operators' view and to move couch manually, average standard deviations of movements between two targets are vertical 1.4mm, longitudinal 0.9mm, lateral 2.2mm in each direction. As for the error in straight dimension, it is about 3.6mm averages and 5.1mm maximum. The average of errors in each directions was vertical 1mm, longitudinal 0.7mm, lateral 2.7mm. The greatest error broke out in lateral direction with $25\%$ of all cases ; to exceed an error average. IV. Conclusions If operators moved manually couch for changing target points, errors about 3.6mm average degrees occur. It is important that operators confirm the errors prices of actual couch coordinates for asking a correct movement between the targets adjoined each other ; in case of treatment demanding high precision like 3D conformal therapy or IMRT. Therefore, if we apply couch coordinates confirmation to reproducibility and to precision evaluation of treatment, it's expected that we can execute high-quality radiotherapy.

• Journal of IKEEE
• /
• v.13 no.2
• /
• pp.181-185
• /
• 2009

In this paper, we are concerned with a 3D line segment extraction method by area-based stereo matching technique. The main idea is based on line fitting of elevation data on 2D line coordinates of ortho-image. Elevation data and ortho-image can be obtained by well-known area-based stereo matching technique. In order to use elevation in line fitting, the elevation itself should be reliable. To measure the reliability of elevation, in this paper, we employ the concept of self-consistency. We test the effectiveness of the proposed method with a quantitative accuracy analysis using synthetic images generated from Avenches data set of Ascona aerial images. Experimental results indicate that our method generates 3D line segments almost 7.5 times more accurate than raw elevations obtained by area-based method.

• Structural Engineering and Mechanics
• /
• v.55 no.4
• /
• pp.785-799
• /
• 2015

A three-dimensional (3-D) method of analysis is presented for determining the free vibration frequencies of hyperboloidal shells free at the top edge and clamped at the bottom edge like a hyperboloidal cooling tower by the Ritz method based upon the circular cylindrical coordinate system instead of related 3-D shell coordinates which are normal and tangent to the shell midsurface. The Legendre polynomials are used as admissible displacements. Convergence to four-digit exactitude is demonstrated. Natural frequencies from the present 3-D analysis are also compared with those of straight beams with circular cross section, complete (not truncated) conical shells, and circular cylindrical shells as special cases of hyperboloidal shells from the classical beam theory, 2-D thin shell theory, and other 3-D methods.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.229-230
• /
• 2022

본 논문에서는 CNN Object Detection과 더불어 3D 카메라 기반 디지털 좌표 인식 기술을 제안한다. 이 기술은 3D Depth Camera인 Intel 사의 Realsense D455를 이용해 대상을 감지하고 분류하며 대상의 위치를 파악한다. 또한 이 기술은 기존의 Depth Camera 내장 거리와는 다르게 좌표를 인식하여 좌표간의 거리까지 계산이 가능하다. 또한 Tensorflow SSD 구조와의 메모리 공유를 통해 시스템의 자원 낭비를 줄이며, 속도를 높이는 멀티쓰레드를 탑재했다. 본 기술을 통해 좌표간의 거리를 계산함으로써 스포츠, 심리, 놀이, 산업 등 다양한 환경에서 활용할 수 있다.

• Journal of Korean Society for Geospatial Information Science
• /
• v.12 no.4 s.31
• /
• pp.61-68
• /
• 2004

In this paper, we suggest an automatic technique to obtain 3-D road information in complex urban areas using road pavement markings. This method is composed of following three main steps. The first step is extracting the pavement markings from aerial images, the second one is matching the same pavement markings in two aerial images, and the last one is obtaining the 3-D coordinates of those using EOP(exterior orientation parameters) of aerial images. Here, we focus on the first and second step because the last step can be performed by using the well hewn collinearity condition equation. We used geometric properties and spatial relationships of the pavement markings to extract the lane line markings on the images and extracted arrow lane markings additionally using template matching. And then, we obtained 3-D coordinates of the road using relational matching for the pavement markings. In order to evaluate the accuracy of extraction, we did a visual inspection and compared the result of this technique with those measured by digital photogrammetric system.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.1624-1629
• /
• 2008

The governing equations in generalized curvilinear coordinates for a 3D pulsatile flow are the Incompressible Navier-Stokes (INS) equations with the artificial dissipative terms and continuity equation discretized using a second-order accurate, finite volume method on the nonstaggered computational grid. This method adopts a dual or pseudo time-stepping Artificial Compressibility (AC) method integrated in pseudo-time. The computational technique implements the implicit approximate factorization method of the Beam and Warming method (1978), which is the extension of the Alternate Direction Implicit (ADI) method. The algorithm yields practically identical velocity profiles and secondary flows that are in excellent overall agreement with an experimental measurement (Rindt & Steenhoven, 1991).