• The Journal of the Korea Contents Association
• /
• v.19 no.9
• /
• pp.666-673
• /
• 2019

In the immersive virtual reality environment, we can select and manipulate various virtual objects. in order to select a virtual object, we generally use Ray-casting method that fires a virtual line in user's view and selects an object when the line and the object match, or Cone-casting method that is widely used to select multiple objects at the same time. However, since the virtual objects used in CAD are composed of small and complex objects in detail, when selecting an object in the user's view by existing methods, there occurs a ambiguity problem that needs additional realignment operation even though an object is selected as a group. in this paper, even if a virtual object is composed of several small virtual objects, it calculates the spatial and logical relationship among objects and expands or shrinks desired objects, so that the user can quickly and accurately select a desired object. in order to evaluate the proposed method, performance comparison were performed using Our and Ray-Casting and Cone-Casting methods. Experimental results show that the proposed method has the fastest speed and the highest accuracy when selecting the desired objects.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.74-80
• /
• 2021

This study investigated the process quality level of a K5 gas mask, which recently acquired its operational capability, through statistical process analysis for the mass production stages and their lots. The tensile adhesion strength was the only operating requirement in the manufacturing process of the K5 gas mask. For this purpose, the results of tensile adhesion strength between the lens and facial rubber during the initial and second mass production stages were analyzed using conventional statistical and statistical process analysis methods. The conventional statistical results indicated that the second mass production stage was better than the initial mass production stage. In cases of a control chart and process capability of tensile adhesion strength, the process quality level was also improved by following the mass production stages. The improvement was caused by process stabilization and work skill elevation. These results and methods are expected to be conducted and utilized in the third mass production stage. Moreover, quality improvement of K5 gas mask mass production can be achieved using the Lean 6 sigma procedure, MDAIC (Define, Measure, Analyze, Improve, Control).

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.591-598
• /
• 2021

One of the requirements of metro trains is to stop with precision to ensure that the train can stop precisely at the designated location on the platform. If this is not satisfied, interference with the screen door occurs, causing inconvenience to passengers and delays in operation. In the case of an automatic operated train, the current position is determined by the current speed information of the train, and control is performed by issuing an acceleration/deceleration command. Therefore, accurate control becomes impossible if the error of the speed information is large. In metro railroads, a Precision Stop Marker (PSM) is used to correct the position error, so that the error of stop control can be reduced by correcting the position error at a specific point. On the other hand, because the PSM itself has only position information, it does not compensate for the speed error. This paper proposes a method for performing in-place stop control by estimating the speed with the PSM progress information. The speed can be estimated when the train is operated at a constant deceleration speed, and the target deceleration can be obtained to perform stop control. The feasibility and excellence of the proposed method are shown through a numerical simulation.

• Journal of Convergence for Information Technology
• /
• v.11 no.6
• /
• pp.262-269
• /
• 2021

The Ministry of National Defense is re-establishing the role of the Army in accordance with the defense reform and is promoting the Warrior Platform, a next-generation individual combat system. The Warrior Platform project is divided into three stages and is being promoted. In the first stage, the quality and performance of individual items are improved, in the second stage, items between system development are integrated, and in the third stage, the combat capability is maximized by developing an integrated unit weapon system. In this paper, detailed sub-items for the five essential required competencies (survival, lethality, mobility, sustainability, Communication) that are considered for building an effective warrior platform are presented. We also present a plan that can be used to prepare a specific master plan for the Army's Warrior Platform project by using Analytic Hierarchy Process(AHP) and selecting the priority of the five required capabilities and detailed sub-items for different unit types. As a result of analyzing the priorities of the four types of units with different mission types, we find that there are differences for each unit. These results are expected to be used as useful reference materials for setting the future direction for the development of warrior platform.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.179-186
• /
• 2021

In general, the daily safety inspection activities, which investigate damages in structures and measures the size of the damage, have been relied heavily on the visual inspection so far. Since the probe of the condition and performance of facilities by such personnel is often dependent on the subjective judgment of the investigator, the consistency and repeatability of the probing results may reduce. Particularly, damage located in a difficult-to-reach place depends mainly on experience with the naked eye, and an unsafe method using a ladder has mainly applied when necessary. Therefore, in this study, we tried to propose a way of using images that can reduce the deviation between safety inspection investigators, enhance objectivity, and improve the safety of workers. In this study, we have applied homographic transformation as a method of correcting the image. As a result of analyzing the size of the damage in the corrected image of the test subject, it confirms that the accuracy of measuring the magnitude of the damage can satisfy the target levels of 5.0mm and 0.005m², the target accuracy levels. As a result of the field verification test to which the proposed image correction technique applied, the coefficient of variation of the crack length in the structure decreased from 5.4~7.0% to 0.072~0.12%, and that of the damaged area from 10.9% to 1.6%. It confirms that the measurement accuracy is improved. Therefore, it is expected that this study on the image utilization technique in safety inspection activities can increase the accuracy of damage measurement and improve the reliability of the safety inspection reports and exterior survey drawings.

• Journal of IKEEE
• /
• v.25 no.2
• /
• pp.273-279
• /
• 2021

In this paper, we propose an algorithm that shortens the learning time by performing individual learning using partitioning the deep learning structure. The proposed algorithm consists of four processes: network classification origin setting process, feature vector extraction process, feature noise removal process, and class classification process. First, in the process of setting the network classification starting point, the division starting point of the network structure for effective feature vector extraction is set. Second, in the feature vector extraction process, feature vectors are extracted without additional learning using the weights previously learned. Third, in the feature noise removal process, the extracted feature vector is received and the output value of each class is learned to remove noise from the data. Fourth, in the class classification process, the noise-removed feature vector is input to the multi-layer perceptron structure, and the result is output and learned. To evaluate the performance of the proposed algorithm, we experimented with the Extended Yale B face database. As a result of the experiment, in the case of the time required for one-time learning, the proposed algorithm reduced 40.7% based on the existing algorithm. In addition, the number of learning up to the target recognition rate was shortened compared with the existing algorithm. Through the experimental results, it was confirmed that the one-time learning time and the total learning time were reduced and improved over the existing algorithm.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.1
• /
• pp.28-36
• /
• 2022

Currently, the automation and intelligence of the shipbuilding industry have improved its work production capacity and cost competitiveness, but the reduction rate of safety accidents among industrial site workers is still low and the damage caused by safety accidents is very serious, so there is a need for improvement according to the workplace. This research aims to demonstrate the connectivity between smart safety helmets in the demonstration area to verify the effectiveness along with the development of smart helmets for worker protection and environmental safety in shipyards. For efficient communication between workers, impact noise of over 95dB was confirmed in the workplace, and noise reduction was required. To solve this problem, the filtering performance was compared and analyzed using the Butterworth, Chebyshev, and elliptic algorithms. The connectivity test and noise reduction method between smart helmets proposed in this study will increase the usability and safety of the field through the development of advanced smart helmets tailored to the shipbuilding workplace in the future.

• Journal of the Society of Disaster Information
• /
• v.18 no.1
• /
• pp.10-18
• /
• 2022

Purpose: For smooth performance of flood analysis due to heavy rain disasters at energy storage facilities in the Incheon area, field surveys, observational surveys, and pre-established reports and drawings were analyzed. Through the field survey, the characteristics of pipelines and rivers that have not been identified so far were investigated, and based on this, the input data of the SWMM model selected for inundation analysis was constructed. Method: In order to determine the critical duration through the probability flood analysis according to the calculation of the probability rainfall intensity by recurrence period and duration, it is necessary to calculate the probability rainfall intensity for an arbitrary duration by frequency, so the research results of the Ministry of Land, Transport and Maritime Affairs were utilized. Result: Based on this, the probability of rainfall by frequency and duration was extracted, the critical duration was determined through flood analysis, and the rainfall amount suggested in the disaster prevention performance target was applied to enable site safety review. Conclusion: The critical duration of the base was found to be a relatively short duration of 30 minutes due to the very gentle slope of the watershed. In general, if the critical duration is less than 30 minutes, even if flooding occurs, the scale of inundation is not large.

• Korean Journal of Construction Engineering and Management
• /
• v.23 no.4
• /
• pp.3-14
• /
• 2022

Autonomous equipment requires a large amount of data from various environments. However, it takes a lot of time and cost for an experiment in a real construction sites, which are difficulties in data collection and processing. Therefore, this study aims to develop a virtual environment for autonomous tower cranes technology development and validation. The authors defined automation functions and operation conditions of tower cranes with three performance criteria: operational design domain, object and event detection and response, and minimum functional conditions. Afterward, this study developed a virtual environment for learning and validation for autonomous functions such as recognition, decision making, and control using the Unity game engine. Validation was conducted by construction industry experts with a fidelity which is the representative matrix for virtual environment assessment. Through the virtual environment platform developed in this study, it will be possible to reduce the cost and time for data collection and technology development. Also, it is also expected to contribute to autonomous driving for not only tower cranes but also other construction equipment.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.3
• /
• pp.239-247
• /
• 2022

Cities are becoming more complex due to rapid industrialization and population growth in modern times. In particular, urban areas are rapidly changing due to housing site development, reconstruction, and demolition. Thus accurate road information is necessary for various purposes, such as High Definition Map for autonomous car driving. In the case of the Republic of Korea, accurate spatial information can be generated by making a map through the existing map production process. However, targeting a large area is limited due to time and money. Road, one of the map elements, is a hub and essential means of transportation that provides many different resources for human civilization. Therefore, it is essential to update road information accurately and quickly. This study uses Semantic Segmentation algorithms Such as LinkNet, D-LinkNet, and NL-LinkNet to extract roads from drone images and then apply hyperparameter optimization to models with the highest performance. As a result, the LinkNet model using pre-trained ResNet-34 as the encoder achieved 85.125 mIoU. Subsequent studies should focus on comparing the results of this study with those of studies using state-of-the-art object detection algorithms or semi-supervised learning-based Semantic Segmentation techniques. The results of this study can be applied to improve the speed of the existing map update process.