• Journal of Internet Computing and Services
• /
• v.7 no.5
• /
• pp.135-145
• /
• 2006

This paper presents a novel approach to real-time recognition of 3D environment and objects for intelligent robots. First. we establish the three fundamental principles that humans use for recognizing and interacting with the environment. These principles have led to the development of an integrated approach to real-time 3D recognition and modeling, as follows: 1) It starts with a rapid but approximate characterization of the geometric configuration of workspace by identifying global plane features. 2) It quickly recognizes known objects in environment and replaces them by their models in database based on 3D registration. 3) It models the geometric details on the fly adaptively to the need of the given task based on a multi-resolution octree representation. SIFT features with their 3D position data, referred to here as stereo-sis SIFT, are used extensively, together with point clouds, for fast extraction of global plane features, for fast recognition of objects, for fast registration of scenes, as well as for overcoming incomplete and noisy nature of point clouds. The experimental results show the feasibility of real-time and behavior-oriented 3D modeling of workspace for robotic manipulative tasks.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.30 no.1
• /
• pp.46-53
• /
• 2010

A Compton X-ray backscatter technique has been developed to quantitatively assess impact damage in quasi-isotropic laminated composites made by a drop-weight tester. X-ray backscatter imaging system with a slit-type camera is constructed to obtain a cross-sectional profile of impact-damaged laminated composites from the electron-density variation of the cross section. A nonlinear scattering model based on Boltsman equation is introduced to compute Compton X-ray backscattering field for the defect assessment. An adaptive filter is also used to reduce noises from many sources including quantum noise and irregular distributions of fibers and matrix in composites. Delaminations masked or distorted by the first delamination are detected and characterized effectively by the Compton X-ray backscatter technique, both in width and location, by application of error minimization algorithm.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.2
• /
• pp.239-245
• /
• 2020

Recently, BIM (Building Information Modeling) are widely being utilized in Construction industry. However, most structures that have been constructed in the past do not have BIM. For structures without BIM, the use of SfM (Structure from Motion) techniques in the 2D image obtained from the camera allows the generation of 3D model point cloud data and BIM to be established. However, since these generated point cloud data do not contain semantic information, it is necessary to manually classify what elements of the structure. Therefore, in this study, deep learning was applied to automate the process of classifying structural components. In the establishment of deep learning network, Inception-ResNet-v2 of CNN (Convolutional Neural Network) structure was used, and the components of bridge structure were learned through transfer learning. As a result of classifying components using the data collected to verify the developed system, the components of the bridge were classified with an accuracy of 96.13 %.

• Journal of Broadcast Engineering
• /
• v.7 no.1
• /
• pp.21-27
• /
• 2002

This paper presents a mews event reenactment system (NERS), which generates virtual character animations in a quick and convenient manner. Thus, NERS can be used to produce computer graphics(CG) scenes of news events that are hard to photograph, such as fire, traffic accident, cases of murder, and so on. By using plenty of captured motion data and CG model data this system produces an appropriate animation of virtual characters straightforwardly without any motion capturing device and actors in the authoring stage. NERS is designed to be capable of making virtual characters move along user-defined paths, stitching motions smoothly and modifyingthe positions and of the articulations of a virtual character in a specific frame. Therefore a virtual character can be controlled precisely so as to interact with the virtual environments and other characters. NERS provides both an interactive and script-based (MEL: Maya Embedded Language) interface so that user can this system in a convenient way. This system has been implemented as a plug-in of commercial CG tool, Maya (Alias/wavefront), in order to make use of its advanced functions

• Journal of KIISE:Software and Applications
• /
• v.33 no.10
• /
• pp.851-861
• /
• 2006

The paper proposes a novel unsupervised video object segmentation algorithm for image sequences with low depth-of-field (DOF), which is a popular photographic technique enabling to represent the intention of photographer by giving a clear focus only on an object-of-interest (OOI). The proposed algorithm largely consists of two modules. The first module automatically extracts OOIs from the first frame by separating sharply focused OOIs from other out-of-focused foreground or background objects. The second module tracks OOIs for the rest of the video sequence, aimed at running the system in real-time, or at least, semi-real-time. The experimental results indicate that the proposed algorithm provides an effective tool, which can be a basis of applications, such as video analysis for virtual reality, immersive video system, photo-realistic video scene generation and video indexing systems.

• Journal of the Institute of Convergence Signal Processing
• /
• v.6 no.4
• /
• pp.182-190
• /
• 2005

In the paper, we propose an intelligent automatic parking assist system. To realize an automatic parking, first, the prospective parking position and the location of a vehicle should be recognized. Second, the system should compute a path which introduces the parking position precisely with avoiding any obstacles. Third, the handle should be controlled so that the vehicle moves through the path. To calculate the location of the vehicle and its surroundings, the system applies the camera image method to transforming input images to the plane map. It also uses the inertial navigation method which recognizes the position and the direction of a moving vehicle by using a kinematic model of the vehicle. To generate a path of the vehicle, the simple path method and the Bezier spline method are tested. The divided arc method which generates multiple paths is also tested. We apply a method which makes the system choose the best path with multiple objective functions. We introduce the virtual road method, as a solution for the problem of mechanical time delay, to have the vehicle followed the designated path.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.3
• /
• pp.57-63
• /
• 2008

As diverse memory cards based on NAND flash memory are getting popularity with consumer electronics such as digital camera, camcorder and MP3 player the compatibility problems between a newly developed memory card and existent host systems have become a main obstacle to time-to-market delivery of product. The common practice for memory card compatibility test is to use a real host system as a test bed. As an improved solution, an FPGA-based prototyping board can be used for emulating host systems. However, the above approaches require a long set-up time and have limitations in representing various host and device systems. In this paper, we propose a co-validation environment for compatibility test between memory card and host system using formal modeling based on Esterel language and co-simulation methodology. Finally, we demonstrate the usefulness of the proposed environment with a case study of real memory card development.

• Atmosphere
• /
• v.29 no.5
• /
• pp.537-550
• /
• 2019

To analyze the observation environment of solar radiation stations operated by the Korea Meteorological Administration (KMA), we analyzed the skyline, Sky View Factor (SVF), and solar radiation due to the surrounding topography and artificial structures using a Digital Elevation Model (DEM), 3D camera, and solar radiation model. Solar energy shielding of 25 km around the station was analyzed using 10 m resolution DEM data and the skyline elevation and SVF were analyzed by the surrounding environment using the image captured by the 3D camera. The solar radiation model was used to assess the contribution of the environment to solar radiation. Because the skyline elevation retrieved from the DEM is different from the actual environment, it is compared with the results obtained from the 3D camera. From the skyline and SVF calculations, it was observed that some stations were shielded by the surrounding environment at sunrise and sunset. The topographic effect of 3D camera is therefore more than 20 times higher than that of DEM throughout the year for monthly accumulated solar radiation. Due to relatively low solar radiation in winter, the solar radiation shielding is large in winter. Also, for the annual accumulated solar radiation, the difference of the global solar radiation calculated using the 3D camera was 176.70 MJ (solar radiation with 7 days; suppose daily accumulated solar radiation 26 MJ) on an average and a maximum of 439.90 MJ (solar radiation with 17.5 days).

• Journal of Biomedical Engineering Research
• /
• v.25 no.1
• /
• pp.5-10
• /
• 2004

Body movement is one of the important factors in sleep analysis. In this study, a thermopile detector with four channels was implemented as a non-contacting detector of body movement in sleep. Using a thermopile mathematical model and several frames of thermal images, the possibility of detecting body movement was evaluated. Instant body movement signals were evaluated for the upper, lower, and entire body using the Haar wavelet. This decomposition shows the points in time when the upper-body or lower-body movement occurred and the level of body movement. Additionally, partial body movement was decomposed in head-only, whole body, and leg-only movement using the ANFIS algorithm. Finally, three subject's data were evaluated for 60 minutes, and the detection rates of instant and partial body movement, on average, were 96.3％ and 89.2％, respectively.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.1
• /
• pp.57-63
• /
• 2012

TIP technology enables navigation of the internal contents of images by extracting geometry information from two-dimensional drawing or a picture and generating three-dimensional effects from extracted information. The technology can be applied to a variety of practical fields including game, entertainment, education, public relations and so on. This paper proposes extended application of TIP technology and realization method for smart devices using OpenGL ES Library for Android platform. Considering problems associated with a foreground object extraction, the proposed method uses vanishing points chosen by the user to facilitate more realistic scene configuration. Then, method acquires three-dimensional background model using OpenGL ES Library, develops three-dimensional virtual space and enables image navigation via camera viewpoint conversion. The experimental image is made on Android 2.1 and OpenGL ES 1.0 using the image taken on devices built on the Android platform. Thus, the proposed technology can be implemented to various smart devices built on the Android platform at lower cost and in less time.