• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.8 no.4
• /
• pp.67-73
• /
• 2008

In the Markov network which models disparity map with the Markov Random Field(MRF), the belief propagation algorithm is operated by message passing between nodes corresponding to each pixels. Belief propagation algorithm required much iteration for accurate result. In this paper, we propose the stereo matching algorithm using belief propagation in multi-resolution domain. Multi-resolution method based on wavelet or lifting can reduce the search area, therefore this algorithm can generate disparity map with fast speed.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.9
• /
• pp.76-84
• /
• 1994

This paper presents a new high-performance VLSI architecture and VLSI implementation for full-search block matching algorithm. The proposed VLSI architecture has the feature of two directional parallel and pipeline processing, thereby reducing the PE idle time at which the direction of block matching operation within the search area is changed. Therfore, the proposed architecture is faster than the existing architectures under the same clock frequency. Based on HSPICE circuit simulation, it is verified that the implemented procesing element is operated successfully within 13 ns for 75 MHz operation.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.243-243
• /
• 2000

This paper introduces a visual inspecting and monitoring system based on an image processing technique. We propose an image processing method for analyzing landslide movement in real time. The method adopts Laplacian of Gaussian operator to extract linear features for the captured images and uses a linear matching algorithm to distinguish the matching error for those features. When the algorithm is processed, motion parameters such as displacement area and its direction are computed. Once movement is recognized, displacements are estimated graphically with statistical amount in the image plane. The simulation results are shown us to verify the effectiveness of the developed method.

• The Journal of Korea Robotics Society
• /
• v.7 no.2
• /
• pp.142-149
• /
• 2012

The camera has limitations of poor visibility in underwater environment due to the limited light source and medium noise of the environment. However, its usefulness in close range has been proved in many studies, especially for navigation. Thus, in this paper, vision-based object detection and tracking techniques using artificial objects for underwater robots have been studied. We employed template matching and mean shift algorithms for the object detection and tracking methods. Also, we propose the weighted correlation coefficient of adaptive threshold -based and color-region-aided approaches to enhance the object detection performance in various illumination conditions. The color information is incorporated into the template matched area and the features of the template are used to robustly calculate correlation coefficients. And the objects are recognized using multi-template matching approach. Finally, the water basin experiments have been conducted to demonstrate the performance of the proposed techniques using an underwater robot platform yShark made by KORDI.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.10
• /
• pp.983-988
• /
• 2011

This paper presents a robust method for ground plane detection in vision-based applications based on a monocular sequence of images with a non-stationary camera. The proposed method, which is based on the reliable estimation of the homography between two frames taken from the sequence, aims at designing a practical system to detect road surface from traffic scenes. The homography is computed using a feature matching approach, which often gives rise to inaccurate matches or undesirable matches from out of the ground plane. Hence, the proposed homography estimation minimizes the effects from erroneous feature matching by the evaluation of the difference between the predicted and the observed matrices. The method is successfully demonstrated for the detection of road surface performed on experiments to fill an information void area taken place from geometric transformation applied to captured images by an in-vehicle camera system.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.90-93
• /
• 2009

The use of physics based deformation methods is continuously increasing in computer graphics area such as game and simulation. Many researchers have worked on this method. However, relatively few researchers have considered the development of the user interaction to the 3D objects. This research proposes a physics-based deformation technique using AR (Augmented Reality) environments to enhance user immersion and the effectiveness of the deformation. In the AR circumstances, the physics based deformation should be accomplished in realtime. In the proposed method, we combine RBF (Radial Basis Function) [1] and LSM (Lattice Shape Matching) [2, 3] and apply it to polygonal models for real-time user interaction. The dynamics of the LSM is also calculated to trace the movement of each lattice. Finally these algorithms are implemented in AR environments.

• Economic and Environmental Geology
• /
• v.26 no.4
• /
• pp.541-549
• /
• 1993

The main goal of this paper is to determine hydraulic properties and to predict drawdown for the efficient and stable development of groundwater in the Daejong-Chun area, North of Kyungsang-Do. Based on geological survey and analysis of well logging data conducted in 1991, it is found that the type of aquifer of this area is considered to be an anisotropic unconfined aquifer with saturated thickness of 19.8 m. In order to characterize this aquifer pumping test was conducted, and the resulting drawdown data were utilized for the analysis by applying both type curve matching technique and semi-log straight line method. As a result, the average specific yield of this aquifer is estimated as 32.3%, and the average ratio of $K_H$ to $K_V$ is only 2.7, which means that gravitational effect is not significant factor for this type of aquifer. For the validation of the estimated hydraulic properties, the analytical model which was developed with Newton-Raphson iteration procedure in this study, was employed to generate the drawdown. And, the resulting drawdown was compared against actual drawdown data and it shows the excellent matches. The actual drawdown data for 9 hours of pumping were used for history matching purposes and relatively satisfactory matches were achieved in this match. Then, the model was run by using the tuned parameters that are obtained during history matching stage, and the drawdown was predicted for the next 30 years of pumping with $3,000m^3/day$ of constant pumping rate. Its result indicates that the drawdown was stabilized as 1.41 m from 20 days with $3,000m^3/day$ of constant pumping rate, which is the required amount of water to be safely supplied to this area.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.29B no.7
• /
• pp.62-69
• /
• 1992

The conventional BMA is performed with the fixed block size. For better performance at low bitrate, the block size is required to be large in relatively stationary area, while small in moving area. Thus, in this paper, a video coding technique using variable block size model is proposed. It decides the block size based on the degree of local motion defined by the local mean and variance of blocks. Computer simulation shows that the proposed method gives comparable performance to the conventional one with less bits required for motion vector coding.

• Proceedings of the IEEK Conference
• /
• 2001.06b
• /
• pp.373-376
• /
• 2001

This paper has been studied a real-time video frame data processing control that used the linear systolic array for motion estimation. The proposed data control processing provides to the input data into the multiple processor array unit(MPAU) from search area and reference block data. The proposed data control architecture has based on two slice band for input data processing. And it has no required external control logic blocks for input data as like reference block or search area data.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.11a
• /
• pp.141-144
• /
• 1997

This paper presents a method of navigation sign recognition in indoor environments using a fuzzy inference for an autonomous mobile robot. In order to adapt to image deformation of a navigation sign resulted from variations of view-points and distances, a multi-labeled template matching(MLTM) method and a dynamic area search method(DASM) are proposed. The DASM is proposed to detect correct feature points among incorrect feature points. Finally sugeno-style fuzzy inference are adopted for recognizing the navigation sign.