• The Journal of Korea Robotics Society
• /
• v.11 no.3
• /
• pp.156-162
• /
• 2016

For the underwater localization, acoustic sensor systems are widely used due to greater penetration properties of acoustic signals in underwater environments. On the other hand, the good penetration property causes multipath and interference effects in structured environment too. To overcome this demerit, a localization method using the attenuation of electro-magnetic(EM) waves was proposed in several literatures, in which distance estimation and 2D-localization experiments show remarkable results. However, in 3D-localization application, the estimation difficulties increase due to the nonuniform (doughnut like) radiation pattern of an omni-directional antenna related to the depth direction. For solving this problem, we added a depth sensor for improving underwater 3D-localization with the EM wave method. A micro scale pressure sensor is located in the mobile node antenna, and the depth data from the pressure sensor is calibrated by the curve fitting algorithm. We adapted the depth(z) data to 3D EM wave pattern model for the error reduction of the localization. Finally, some experiments were executed for 3D localization with the fast calculation and less errors.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.56-58
• /
• 2006

Camera calibration in machine vision is the process of determining the intrinsic cameara parameters and the three-dimensional (3D) position and orientation of the camera frame relative to a certain world coordinate system. On the other hand, Takagi-Sugeno-Kang (TSK) fuzzy system is a very popular fuzzy system and approximates any nonlinear function to arbitrary accuracy with only a small number of fuzzy rules. It demonstrates not only nonlinear behavior but also transparent structure. In this paper, we present a novel and simple technique for camera calibration for machine vision using TSK fuzzy model. The proposed method divides the world into some regions according to camera view and uses the clustered 3D geometric knowledge. TSK fuzzy system is employed to estimate the camera parameters by combining partial information into complete 3D information. The experiments are performed to verify the proposed camera calibration.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.6
• /
• pp.33-38
• /
• 2016

Understanding of current features of collapsed buildings, terrain, and other infrastructures is a critical issue for disaster site managers. On the other hand, a comprehensive site investigation of current location of survivors buried under the remains of a building is a difficult task for disaster managers due to the difficulties in acquiring the various information on the disaster sites. To overcome these circumstances, such as large disaster sites and limited capability of rescue workers, this study makes use of a drone (unmanned aerial vehicle) to effectively obtain current image data from large disaster areas. The framework of 3D model reconstruction of disaster sites using aerial imagery acquired by drones was also presented. The proposed methodology is expected to assist fire fighters and workers on disaster sites in making a rapid and accurate identification of the survivors under collapsed buildings.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.55-58
• /
• 2008

Designers, who design industry products, use CAD(Computer Aided Design) tools for making new design and looking around virtual 3D models. Hand-drawings and sketches show only one viewpoint limiting 3D perception. However, CAD system that provides automation and multiple view points, can help to save time and cost. Accordingly, we developed Augmented Reality(AR) and Rapid Prototyping(RP) based product design system that is interactive and realistic This AR based design system utilize mockups that are made of urethane and styrofoam where as users change 3D model's color, texture and user interface. These interactive ways help to evaluate design more instinctively.

• Advances in aircraft and spacecraft science
• /
• v.4 no.3
• /
• pp.219-235
• /
• 2017

The large container ships and fast patrol boats are complex marine structures. Therefore, their global mechanical behaviour has long been modeled mostly by refined beam theories. Important issues of cross section warping and bending-torsion coupling have been addressed by introducing special functions in these theories with inherent assumptions and thus compromising their robustness. The 3D solid Finite Element (FE) models, on the other hand, are accurate enough but pose high computational cost. In this work, different marine vessel structures have been analysed using the well-known Carrera Unified Formulation (CUF). According to CUF, the governing equations (and consequently the finite element arrays) are written in terms of fundamental nuclei that do not depend on the problem characteristics and the approximation order. Thus, refined models can be developed in an automatic manner. In the present work, a particular class of 1D CUF models that was initially devised for the analysis of aircraft structures has been employed for the analysis of marine structures. This class, which was called Component-Wise (CW), allows one to model complex 3D features, such as inclined hull walls, floors and girders in the form of components. Realistic ship geometries were used to demonstrate the efficacy of the CUF approach. With the same level of accuracy achieved, 1D CUF beam elements require far less number of Degrees of Freedom (DoFs) compared to a 3D solid FE solution.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.2
• /
• pp.655-666
• /
• 2014

Recently, as the appearance and exterior design of the construction structure are highlighted, the irregularly shaped structures are increasing in a construction facility. Many softwares provide a quantity take-off function of 3D object under BIM environment, however, they are focused on the limited function based on the solid modeling method. Because the vast geometric information of the curved surface is difficult to extract in the 3D objects that consist of major changes in vertical section shape as the irregularly shaped structures, it is difficult to express a 3D object as a solid model. On the other hand, the irregularly shaped structures can be expressed in relatively free in the surface model because the surface model consists of points, lines and surfaces. Accordingly, the surface modeling method is suitable for the modeling of large irregularly shaped structures. This study suggests a quantity take-off algorithm for the irregularly shaped structures using the surface modeling approach that is beneficial in the design work of structures. Some case projects are used for verifying the accuracy of the proposed method.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.3
• /
• pp.199-204
• /
• 2013

An implementation method of real-time numeral recognizer based on gesture is presented in this paper for various information devices. The proposed algorithm steadily captures the motion of a hand on 3D open space with the Kinect sensor. The captured hand motion is simplified with PCA, in order to preserve the trace consistency and to minimize the trace variations due to noises and size changes. In addition, we also propose a new HMM using both the gradient and the positional features of the simplified hand stroke. As the result, the proposed algorithm has robust characteristics to the variations of the size and speed of hand motion. The recognition rate is increased up to 30%, because of this combined model. Experimental results showed that the proposed algorithm gives a high recognition rate about 98%.

• Journal of the Korea Computer Graphics Society
• /
• v.27 no.2
• /
• pp.1-11
• /
• 2021

In a virtual environment, a 3D workspace and 3D interaction are possible, but most virtual reality applications use a 2D color picker. This paper implements a 3D color picker based on 3D color space in a virtual environment, and compares color selection performance with the existing 2D color picker. The 3D color picker is intuitive by using the 3D color space as it is, and it can position the 3D pointer at a specific point in the color space using a controller, which is a virtual reality device, so a user can select a color in one step. On the other hand, the 2D color picker has the advantage of being familiar with existing users who work with colors in a computer environment, but has a disadvantage that requires several steps of user interaction since it has to set color properties through 2D interfaces. Based on user experiments, we confirmed the usefulness of a 3D color picker in addition to a 2D color picker in a virtual environment, and it was possible to perform natural 3D work in a virtual environment using the 3D color picker.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.6
• /
• pp.407-412
• /
• 2015

In model-based tracking, an accurate 3D model of a target object or scene is mostly assumed to be known or given in advance, but the accuracy of the model should be guaranteed for accurate pose estimation. In many application domains, on the other hand, end users are not highly distracted by tracking errors from certain levels of modeling errors. In this paper, we examine perceptual tracking errors, which are predominantly caused by modeling errors, on subjective evaluation and compare them to computational tracking errors. We also discuss the tolerance of modeling errors by analyzing their permissible ranges.

• Journal of Platform Technology
• /
• v.12 no.1
• /
• pp.3-11
• /
• 2024

Deaf people who use sign language as their first language sometimes have difficulty communicating because they do not know spoken Korean. Deaf people are also members of society, so we must support to create a society where everyone can live together. In this paper, we present a method to increase the recognition rate of Korean sign language using a CNN model. When the original image was used as input to the CNN model, the accuracy was 0.96, and when the image corresponding to the skin area in the YCbCr color space was used as input, the accuracy was 0.72. It was confirmed that inserting the original image itself would lead to better results. In other studies, the accuracy of the combined Conv1d and LSTM model was 0.92, and the accuracy of the AlexNet model was 0.92. The CNN model proposed in this paper is 0.96 and is proven to be helpful in recognizing Korean sign language.