• Journal of Advanced Navigation Technology
• /
• v.22 no.6
• /
• pp.530-536
• /
• 2018

Recently, many studies for interoperability of UAV in the NAS has been performed since the application range and demand of UAV are continuously increased. For the interoperation of UAV in the NAS, technical standards and certification system for UAV which is equivalent to the commercial aircraft are required and test and evaluation methodology must be presented by standards. In this paper, qualification test and evaluation methodology aboutfor the UAV navigation system is proposed. For the research, the mission profile and operation environment of UAV were analyzed. Thereafter the test criteria were derived and the test methodology were established. Finally, the simulation and demonstration using test-bed UAV were performed. As a result of the test, it was confirmed that the navigation system of test UAV has a position accuracy about 1.4 meters at 95% confidence level in the entire flight stage.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.3
• /
• pp.11-20
• /
• 2019

In this study, a novel virtual reality (VR) experience environment is proposed for enabling walking adaptation of visually impaired people. The core of proposed VR environment is based on immersive walking interactions and deep learning based braille blocks recognition. To provide a realistic walking experience from the perspective of visually impaired people, a tracker-based walking process is designed for determining the walking state by detecting marching in place, and a controller-based VR white cane is developed that serves as the walking assistance tool for visually impaired people. Additionally, a learning model is developed for conducting comprehensive decision-making by recognizing and responding to braille blocks situated on roads that are followed during the course of directions provided by the VR white cane. Based on the same, a VR application comprising an outdoor urban environment is designed for analyzing the VR walking environment experience. An experimental survey and performance analysis were also conducted for the participants. Obtained results corroborate that the proposed VR walking environment provides a presence of high-level walking experience from the perspective of visually impaired people. Furthermore, the results verify that the proposed learning algorithm and process can recognize braille blocks situated on sidewalks and roadways with high accuracy.

• Journal of IKEEE
• /
• v.23 no.2
• /
• pp.743-746
• /
• 2019

In this paper we proposed a new ocean radiation automatic monitoring system. The proposed system has the following characteristics: First, using NaI + PVT mixed detectors, the response speed is fast and precision analysis is possible. Second, the application of temperature compensation algorithm to scintillator-type sensors does not require additional cooling devices and enables stable operation in the changing ocean environment. Third, since cooling system is not needed, electricity consumption is low, and electricity can be supplied reliably by utilizing solar energy, which can be installed at the observation deck of ocean environment. Fourth, using GPS and wireless communications, accurate location information and real-time data transmission function for measurement areas enables immediate warning response in the event of nuclear accidents such as those involving neighboring countries. The results tested by the authorized testing agency to assess the performance of the proposed system were measured in the range of $5{\mu}Sv/h$ to 15mSv/h, which is the highest level in the world, and the accuracy was determined to be ${\pm}8.1%$, making normal operation below the international standard ${\pm}15%$. The internal environmental grade (waterproof) was achieved, and the rate of variation was measured within 5% at operating temperature of $-20^{\circ}C$ to $50^{\circ}C$ and stability was verified. Since the measured value change rate was measured within 10% after the vibration test, it was confirmed that there will be no change in the measured value due to vibration in the ocean environment caused by waves.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.10a
• /
• pp.277-280
• /
• 2016

Currently, systems for finding missing children are composed of using communication between a QR code and RFID chip, as the use of a smartphone. However, the current systems for finding missing children have limitations in that children can only be found if there are people in the surrounding area; there is an economic burden on parents required to purchase a smartphone for their children; along with difficulties in finding the missing children without the assistance of those in the surrounding area in critical situations such as a kidnapping, due to the limited duration of the battery life. In order to solve such problems, approaches need to be made from two perspectives: having someone in the surrounding area; and absence of anyone in the surrounding area. This thesis is centered on the development of a IoT (Internet of Things) system for finding missing children that combines two methods, namely, the method of finding missing children without a guardian in the surrounding area -within the limited space in which AP is installed by using a beacon and open source hardware being highlighted as the IoT technology - and the method of finding missing children with the smartphone application in which each individual becomes the Access Point (AP). The Main purpose is to provide accurate information of missing children's location for the 2situations and it is found that the accuracy of smartphones APP is 97.7% and security device AP is 91.1%.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.9
• /
• pp.175-181
• /
• 2019

In the case of a conventional single stage decompression regulator used for large depressurization in the hydrogen fuel cell system of a fuel cell electric vehicle (FCEV), problems can arise, such as pulsation, slow response, hydrogen brittleness, leakage, high weight, and high cost due to high decompression. Most of these problems can be overcome easily using two decompression mechanisms (two-stage structures). In addition, a wide outlet-pressure control range can be secured if an electronic solenoid is applied to the second decompression. Accordingly, it is necessary to improve the precision of the outlet pressure of a two-stage pressure-reducing regulator and develop techniques, such as leakage prevention, durability, light weight, and price reduction. Therefore, to improve the outlet pressure accuracy and prevent leakage, the structural part before and after decompression to improve the air tightness were divided and the analysis was carried out assuming that the valve part was closed (open ratio: 0%) after each initial internal pressure application.

• Convergence Security Journal
• /
• v.17 no.5
• /
• pp.41-47
• /
• 2017

The Marine Environment Observation Information System supplies oceanographic information elements such as water temperature, chlorophyll, float, etc. based on satellite images to consumers. The data produced by the Korean marine environmental observatories are located in the coastal areas of Korea. But if the range is too far from a particular area of interest, due to distance or spatial constraints, the accuracy and up-to-dateness of the data can not be relied upon. Therefore, it is necessary to perform fusion and complex operation to solve the difference between the field observation and the marine satellite image. In this study, we develop a system that can process marine environmental information in the user's area of interest in the form of layered character (numeric) information considering the readability and light weight rather than the satellite image. In order to intuitively understand satellite image information, we characterize (quantify) the marine environmental information of the area of interest and we process the satellite image band values into layered characters to minimize the absolute amount of transmitted / received data. Also we study modular location-based interest information service method to be able to flexibly extend and connect interested items that diversify various observation fields as well as application technology to serve this.

• Journal of Navigation and Port Research
• /
• v.43 no.1
• /
• pp.23-32
• /
• 2019

In this paper, we proposed a multi-chain Time of Arrival (ToA) positioning method to estimate positions using all received Loran-C signals from multiple chains without constraining to a single chain. Conventionally, we have to choose only one chain among several available chains for position estimation using Loran-C. Therefore, the number of signals to be used for positioning is limited to three to five. In general, if more signals are used for positioning estimation, its performance tends to be improved in terms of accuracy and availability. To validate the proposed method for multi-chain Loran-C, we firstly carried out a static positioning test in land. By analyzing the test results, we confirmed that the proposed method works well under a multi-chain Loran-C scenario. Subsequently, another mobile positioning test was conducted on board a vessel under a practical application scenario. From this second test, we successfully demonstrated that the multi-chain ToA positioning method even in situations where the conventional single-chain Loran-C approach fails for positioning.

• Journal of Navigation and Port Research
• /
• v.43 no.1
• /
• pp.72-78
• /
• 2019

The second-hand ship market provides immediate access to the freight market for shipping investors. When introducing second-hand vessels, the precise estimate of the price is crucial to the decision-making process because it directly affects the burden of capital cost to investors in the future. Previous studies on the second-hand market have mainly focused on the market efficiency. The number of papers on the estimation of second-hand vessel values is very limited. This study proposes an artificial neural network model that has not been attempted in previous studies. Six factors, freight, new-building price, orderbook, scrap price, age and vessel size, that affect the second-hand ship price were identified through literature review. The employed data is 366 real trading records of Panamax second-hand vessels reported to Clarkson between January 2016 and December 2018. Statistical filtering was carried out through correlation analysis and stepwise regression analysis, and three parameters, which are freight, age and size, were selected. Ten-fold cross validation was used to estimate the hyper-parameters of the artificial neural network model. The result of this study confirmed that the performance of the artificial neural network model is better than that of simple stepwise regression analysis. The application of the statistical verification process and artificial neural network model differentiates this paper from others. In addition, it is expected that a scientific model that satisfies both statistical rationality and accuracy of the results will make a contribution to real-life practices.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.419-432
• /
• 2019

Most of deep learning model training was proceeded by supervised learning, which is to train labeling data composed by inputs and corresponding outputs. Labeling data was directly generated manually, so labeling accuracy of data is relatively high. However, it requires heavy efforts in securing data because of cost and time. Additionally, the main goal of supervised learning is to improve detection performance for 'True Positive' data but not to reduce occurrence of 'False Positive' data. In this paper, the occurrence of unpredictable 'False Positive' appears by trained modes with labeling data and 'True Positive' data in monitoring of deep learning-based CCTV accident detection system, which is under operation at a tunnel monitoring center. Those types of 'False Positive' to 'fire' or 'person' objects were frequently taking place for lights of working vehicle, reflecting sunlight at tunnel entrance, long black feature which occurs to the part of lane or car, etc. To solve this problem, a deep learning model was developed by simultaneously training the 'False Positive' data generated in the field and the labeling data. As a result, in comparison with the model that was trained only by the existing labeling data, the re-inference performance with respect to the labeling data was improved. In addition, re-inference of the 'False Positive' data shows that the number of 'False Positive' for the persons were more reduced in case of training model including many 'False Positive' data. By training of the 'False Positive' data, the capability of field application of the deep learning model was improved automatically.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.5
• /
• pp.64-74
• /
• 2019

Although a non-rigid registration has high demands in clinical practice, it has a high computational complexity and it is very difficult for ensuring the accuracy and robustness of registration. This study proposes a method of applying a non-rigid registration to 3D magnetic resonance images of brain in an unsupervised learning environment by using a deep-learning network. A feature vector between two images is produced through the network by receiving both images from two different patients as inputs and it transforms the target image to match the source image by creating a displacement vector field. The network is designed based on a U-Net shape so that feature vectors that consider all global and local differences between two images can be constructed when performing the registration. As a regularization term is added to a loss function, a transformation result similar to that of a real brain movement can be obtained after the application of trilinear interpolation. This method enables a non-rigid registration with a single-pass deformation by only receiving two arbitrary images as inputs through an unsupervised learning. Therefore, it can perform faster than other non-learning-based registration methods that require iterative optimization processes. Our experiment was performed with 3D magnetic resonance images of 50 human brains, and the measurement result of the dice similarity coefficient confirmed an approximately 16% similarity improvement by using our method after the registration. It also showed a similar performance compared with the non-learning-based method, with about 10,000 times speed increase. The proposed method can be used for non-rigid registration of various kinds of medical image data.