• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.4
• /
• pp.321-329
• /
• 2015

This paper addresses a new algorithm for a stereo visual odometry to measure the ego-motion of a vehicle. The new algorithm introduces an inlier grouping method based on Delaunay triangulation and vanishing point computation. Most visual odometry algorithms rely on RANSAC in choosing inliers. Those algorithms fluctuate largely in processing time between images and have different accuracy depending on the iteration number and the level of outliers. On the other hand, the new approach reduces the fluctuation in the processing time while providing accuracy corresponding to the RANSAC-based approaches.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2000.03b
• /
• pp.297-302
• /
• 2000

This paper reports the application of GPS (Global Positioning System) technique into the geotechnical information system. Typical example of such application can be the database of logging data. Other examples can be found from the surveying for road work, excavation work, landslide mapping and image processing of slope face. This paper also reports the enhancement method of the accuracy. The results from this study confirms that GPS could be a powerful tool for the future geotechnical investigation works in Korea.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.18 no.1
• /
• pp.19-26
• /
• 2010

To expand the application area of global navigation satellite systems, precision landing is one of the most critical area to be solved. For the development and validation of the precision landing system, many aspects need to be analyzed including the system architecture, signal characteristics, atmospheric delay, communication delay, accuracy, integrity, and availability. Among them, the signal characteristics analysis requires the processing of measurements collected by real-flight experiments. This paper presents the processing results of the real measurements collected by a flight and landing experiment. To process and analyze the data, double differencing position-domain hatch filter is utilized. Accuracy of the proposed filter is evaluated utilizing reference trajectory generated by commercial software. Finally, by comparing with conventional range domain characteristics of position domain filter is analyzed.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.6
• /
• pp.448-455
• /
• 2003

A system design technique of automatic thickness measurement and defect inspection system, which measures the thickness of the ALC(Autoclaved Lightweight Concrete) block and inspects the defect on a realtime basis is proposed. The image processing system was established with a CCD camera, an image grabber, and a personal computer without using assembled measurement equipment. The image obtained by this system was analyzed by a devised algorithm, specially designed for the enhanced measurement accuracy. For the realization of the proposed algorithm, the preprocessing method that can be applied to overcome uneven lighting environment, an enhanced edge decision method using 8 edge-pairs with irregular and rough surface, the unit length decision method in uneven condition with rocking objects, and the curvature calibration method of camera using a constructed grid are developed. The experimental results, show that the required measurement accuracy specification is sufficiently satisfied using our proposed method.

• Journal of Information Processing Systems
• /
• v.14 no.2
• /
• pp.418-434
• /
• 2018

With the increasing use of the Internet and electronic documents, automatic text categorization becomes imperative. Several machine learning algorithms have been proposed for text categorization. The k-nearest neighbor algorithm (kNN) is known to be one of the best state of the art classifiers when used for text categorization. However, kNN suffers from limitations such as high computation when classifying new instances. Instance selection techniques have emerged as highly competitive methods to improve kNN through data reduction. However previous works have evaluated those approaches only on structured datasets. In addition, their performance has not been examined over the text categorization domain where the dimensionality and size of the dataset is very high. Motivated by these observations, this paper investigates and analyzes the impact of instance selection on kNN-based text categorization in terms of various aspects such as classification accuracy, classification efficiency, and data reduction.

• Journal of Information Processing Systems
• /
• v.13 no.2
• /
• pp.256-267
• /
• 2017

An active research area in computer vision, stereo matching is aimed at obtaining three-dimensional (3D) information from a stereo image pair captured by a stereo camera. To extract accurate 3D information, a number of studies have examined stereo matching algorithms that employ adaptive support weight. Among them, the adaptive census transform (ACT) algorithm has yielded a relatively strong matching capability. The drawbacks of the ACT, however, are that it produces low matching accuracy at the border of an object and is vulnerable to noise. To mitigate these drawbacks, this paper proposes and analyzes the features of an improved stereo matching algorithm that not only enhances matching accuracy but also is also robust to noise. The proposed algorithm, based on the ACT, adopts the truncated absolute difference and the multiple sparse windows method. The experimental results show that compared to the ACT, the proposed algorithm reduces the average error rate of depth maps on Middlebury dataset images by as much as 2% and that is has a strong robustness to noise.

• Journal of the Korean Society for Precision Engineering
• /
• v.31 no.1
• /
• pp.75-81
• /
• 2014

Laser beam machining has been known as efficient for glass micromachining. It is usually used the ultra-short pulsed laser which is time-consuming and uneconomic process. In order to use economic and powerful long pulsed laser, indirect processing called laser-induced backside wet etching (LIBWE) is good alternative method. In this paper, micromachining of glass using Nd:YAG laser with nanosecond pulsed beam has been attempted. In order to improve shape accuracy, combined processing with magnetic stirrer has been widely used. Magnetic stirrer acts to circulate the solution and remove the bubble but it is not suitable for deep hole machining. To get better effect, ultrasonic vibration was applied for improving shape accuracy.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2001.10a
• /
• pp.85-90
• /
• 2001

Recent general Press brake has many problems in cutting high accurate products in the progress of industry. Previous hand-operated press brake needs many pre-processing works to adjust bending ang1e and marking-off works to calculate bending length. Also, the hand-operating work makes many geometric errors and has difficulty for variety-mass production. To solve these problems, this paper proposes Computer Numerical Control (CNC) general press brake and development of servo-control system based on database for reduction of geometric errors and pre-processing work time and high accuracy bending work.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.13 no.4
• /
• pp.30-36
• /
• 2004

The development of LD(Laser Diode) tester and its control system based on the graphical programming language(LabVIEW) is addressed. The ill tester is used to check the optic power and the optic spectrum of the LD Chip. The emitter size of LD chip and the diameter of the Detector(optic fiber and photo diode) are very small, therefore the test device needs high accuracy. But each motion part of the test device could not accomplish high accuracy due to the limit of the mechanical performance. So, an image processing with machine vision is proposed to compensate for the error. By adopting our method we can reduce the error of position within $\pm$5$\mu\textrm{m}$.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.1383-1388
• /
• 2007

Accuracy and processing time are very important factors when the desired shape is fabricated with Selective Laser Sintering (SLS), one of Solid Freeform Fabrication (SFF) systems. In a conventional SLS process, laser spot size is fixed during laser exposing on the sliced figure. Therefore, it is difficult to accurately and rapidly fabricate the desired shape. In this paper, to deal with those problems an SFF system having ability of changing spot size is developed. The system provides high accuracy and optimal processing time. Specifically, a variable beam expander is employed to adjust spot size for different figures on a sliced shape. Finally, Design and performance estimation of the SFF system employing a variable beam expander are achieved and the mechanism will be addressed to measure the real spot size generated from the variable beam expander.