• The Journal of the Korea Contents Association
• /
• v.19 no.3
• /
• pp.72-79
• /
• 2019

Autonomous locomotion has two essential functionalities: mapping builds and updates maps by uncertain position information and measured sensor inputs, and localization is to find the positional information with the inaccurate map and the sensor information. In addition, obstacle detection, avoidance, and path designs are necessarily required for autonomous locomotion by combining the probabilistic methods based on uncertain locations. The sensory inputs, which are measured by a metric-based scanner, have difficulties of distinguishing moving obstacles like humans from static objects like walls in given environments. This paper proposes the low resolution grid map combined with reinforcement learning, which is compared with the conventional recognition method for detecting static and moving objects to generate obstacle avoiding path. Finally, the proposed method is verified with experimental results.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.6
• /
• pp.525-534
• /
• 2018

Camera calibration is the process of determining the parameters such as the focal length of a camera, the position of a principal point, and lens distortions. For this purpose, images of checkerboard have been mainly used. When targets were automatically recognized in checkerboard image, the existing studies had limitations in that the user should have a good understanding of the input parameters for recognizing the target or that all checkerboard should appear in the image. In this study, a methodology for automatic target recognition was proposed. In this method, even if only a part of the checkerboard image was captured using rectangles including eight blobs, four each at the central portion and the outer portion of the checkerboard, the index of the target can be automatically assigned. In addition, there is no need for input parameters. In this study, three conditions were used to automatically extract the center point of the checkerboard target: the distortion of black and white pattern, the frequency of edge change, and the ratio of black and white pixels. Also, the direction and numbering of the checkerboard targets were made with blobs. Through experiments on two types of checkerboards, it was possible to automatically recognize checkerboard targets within a minute for 36 images.

• Journal of Ocean Engineering and Technology
• /
• v.33 no.5
• /
• pp.454-461
• /
• 2019

In autonomous interventions using an underwater vehicle with a manipulator, grasping based on target detection and recognition is one of the core technologies. To complete an autonomous grasping task, the vehicle body approaches the target closely and then holds it through operating the end-effector of the manipulator, while the vehicle maintains its position and attitude without unstable motion. For vehicle motion control, it is very important to identify the hydrodynamic parameters of the underwater vehicle, including the propulsion force. This study examined the propulsion characteristics of the autonomous intervention ROV developed by KRISO, because there is a difference between the real exerted force and the expected force. First, the mapping between the input signal and thrusting force for each underwater thruster was obtained through a water tank experiment. Next, the real propulsion forces and moments of the ROV exerted by thrusting forces were directly measured using an F/T (force/torque) sensor attached to the ROV. Finally, the differences between the measured and expected values were confirmed.

• Journal of Advanced Navigation Technology
• /
• v.23 no.4
• /
• pp.296-301
• /
• 2019

Artificial intelligence is closely linked to our real lives, leading innovation in various fields. Especially, as a means of transportation possessing artificial intelligence, autonomous unmanned vehicles are actively researched and are expected to be put into practical use soon. Autonomous cars and autonomous unmanned aerial vehicles are required to equip accurate navigation system so that they can find out their present position and move to their destination. At present, the navigation of transportation that we operate is mostly dependent on GPS. However, GPS is vulnerable to external intereference. In fact, since 2010, North Korea has jammed GPS several times, causing serious disruptions to mobile communications and aircraft operations. Therefore, in order to ensure safety in the operation of the autonomous unmanned vehicles and to prevent serious accidents caused by the intereference, rapid situation judgment and countermeasure are required. In this paper, based on big data and machine learning technology, we propose a countermeasure system for GPS interference that supports decision making by applying John Boyd's OODA loop cycle (detection - direction setting - determination - action).

• Corrosion Science and Technology
• /
• v.19 no.6
• /
• pp.296-301
• /
• 2020

Failure analysis on the welded type 304 pipe used for cooling water piping in the district heating primary side was conducted. Inorganic elements and bacteria in the cooling water and in corrosion products were analyzed, and the weldment was inspected by microscopy and a sensitization test. Corrosion damages were observed in the heat-affected zone, on weld defects such as incomplete fusion or excessive penetration caused by improper welding, or/and at the 6 o'clock position along the pipe axial direction. However, the level of concentration of chloride in the cooling water as low as 80 ppm has been reported to be not enough for even a sensitized type 304 steel, meaning that the additional corrosive factor was required for these corrosion damages. The factor leading to these corrosion damages was drawn to be the metabolisms of the types of bacteria, which is proved by the detection of proton, sulfur containing species, biofilms, and both bacteria and corrosion product analyses.

• The korean journal of orthodontics
• /
• v.51 no.2
• /
• pp.77-85
• /
• 2021

Objective: To evaluate the accuracy of a multi-stage convolutional neural network (CNN) model-based automated identification system for posteroanterior (PA) cephalometric landmarks. Methods: The multi-stage CNN model was implemented with a personal computer. A total of 430 PA-cephalograms synthesized from cone-beam computed tomography scans (CBCT-PA) were selected as samples. Twenty-three landmarks used for Tweemac analysis were manually identified on all CBCT-PA images by a single examiner. Intra-examiner reproducibility was confirmed by repeating the identification on 85 randomly selected images, which were subsequently set as test data, with a two-week interval before training. For initial learning stage of the multi-stage CNN model, the data from 345 of 430 CBCT-PA images were used, after which the multi-stage CNN model was tested with previous 85 images. The first manual identification on these 85 images was set as a truth ground. The mean radial error (MRE) and successful detection rate (SDR) were calculated to evaluate the errors in manual identification and artificial intelligence (AI) prediction. Results: The AI showed an average MRE of 2.23 ± 2.02 mm with an SDR of 60.88% for errors of 2 mm or lower. However, in a comparison of the repetitive task, the AI predicted landmarks at the same position, while the MRE for the repeated manual identification was 1.31 ± 0.94 mm. Conclusions: Automated identification for CBCT-synthesized PA cephalometric landmarks did not sufficiently achieve the clinically favorable error range of less than 2 mm. However, AI landmark identification on PA cephalograms showed better consistency than manual identification.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.6
• /
• pp.15-29
• /
• 2018

This paper proposes an enhanced object tracking algorithm to compensate the lack of temporal information in existing particle swarm optimization based object trackers using the trajectory of the target object. The proposed scheme also enables the tracking and documentation of the location of an online updated set of distractions. Based on the trajectories information and the distraction set, a rule based approach with adaptive parameters is utilized for occlusion detection and determination of the target position. Compare to existing algorithms, the proposed approach provides more comprehensive use of available information and does not require manual adjustment of threshold values. Moreover, an effective weight adjustment function is proposed to alleviate the diversity loss and pre-mature convergence problem in particle swarm optimization. The proposed weight function ensures particles to search thoroughly in the frame before convergence to an optimum solution. In the existence of multiple objects with similar feature composition, this algorithm is tested to significantly reduce convergence to nearby distractions compared to the other existing swarm intelligence based object trackers.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.11
• /
• pp.1384-1390
• /
• 2019

This paper proposes a technological solution for improving the recognition direction of target objects for automatic vision inspection by machine vision. This paper proposes a technological solution for improving the recognition direction of target objects for automatic vision inspection by machine vision. This enables the automatic machine vision inspection to detect the image of the inspection object regardless of the position and orientation of the object, eliminating the need for a separate inspection jig and improving the automation level of the inspection process. This study develops the technology and method that can be applied to the wire harness manufacturing process as the inspection object and present the result of real system. The results of the system implementation was evaluated by the accredited institution. This includes successful measurement in the accuracy, detection recognition, reproducibility and positioning success rate, and achievement the goal in ten kinds of color discrimination ability, inspection time within one second and four automatic mode setting, etc.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.12
• /
• pp.1013-1020
• /
• 2020

Spark-induced breakdown spectroscopy (SIBS) utilizes an electric spark to induce a strong plasma for collecting atomic emissions. This study analyses the potential for usinga compact SIBS instead of conventional laser-induced breakdown spectroscopy (LIBS) in discriminating rocks and soils for planetary missions. Targeting bulky solids using SIBS has not been successful in the past, and therefore a series of optimizations of electrode positioning and electrode materials were performed in this work. The limit of detection (LOD) was enhanced up to four times compared to when LIBS was used, showing a change from 78 to 20 ppm from LIBS to SIBS. Because of the higher energy of plasma generated, the signal intensity by SIBS was higher than LIBS in three orders of magnitude with the same spectrometer setup. Changing the electrode material and locating the optimum position of the electrodes were considered for optimizing the current SIBS setup being tested for samples of planetary origin.

• Structural Engineering and Mechanics
• /
• v.85 no.2
• /
• pp.197-206
• /
• 2023

The deterioration caused by chloride penetration and carbonation plays a significant role in a concrete structure in a marine environment. The chloride corrosion in some marine concrete structures is invisible but can be dangerous in a sudden collapse. Therefore, as a novelty, this research investigates the ability of a non-destructive damage detection method named the Power Spectral Density (PSD) to diagnose damages caused only by chloride ions in concrete structures. Furthermore, the accuracy of this method in estimating the amount of annual damage caused by chloride in various parts and positions exposed to seawater was investigated. For this purpose, the RC Arosa bridge in Spain, which connects the island to the mainland via seawater, was numerically modeled and analyzed. As the first step, each element's bridge position was calculated, along with the chloride corrosion percentage in the reinforcements. The next step predicted the existence, location, and timing of damage to the entire concrete part of the bridge based on the amount of rebar corrosion each year. The PSD method was used to monitor the annual loss of reinforcement cross-section area, changes in dynamic characteristics such as stiffness and mass, and each year of the bridge structure's life using sensitivity equations and the linear least squares algorithm. This study showed that using different approaches to the PSD method based on rebar chloride corrosion and assuming 10% errors in software analysis can help predict the location and almost exact amount of damage zones over time.