• Journal of Navigation and Port Research
• /
• v.46 no.3
• /
• pp.158-167
• /
• 2022

The International Maritime Organization's e-Navigation strategy introduces new technologies to ships to support easier and safer navigation. To implement the e-Navigation strategy, it was necessary to develop a common data model, that could meet various requirements across all aspects of the maritime information service. The International Hydrographic Organization's S-100 Universal Hydrographic Data Model was selected, as the basis for the standardization of maritime data products. Three S-100 based product specifications for weather information, collectively called "S-41X", are currently under development by the NOAA's Ocean Prediction Center, for use in the Electronic Chart Display and Information System (ECDIS). This paper describes a design of a grid based S-413 data model out of three S-41X product specifications. Other S-100 data products, which support the gridded data format, were considered. To verify the data model, an encoding test was conducted, using the Korean Meteorological Adminstration's numerical prediction model results.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.3
• /
• pp.423-432
• /
• 2022

This paper addresses the safety concern that the SOH of batteries in electric vehicles decreases sharply when drivers change or their driving patterns change. Such a change can overload the battery, reduce the battery life, and induce safety issues. This paper aims to present the SOH as the changes on a dashboard of an electric vehicle in real-time in response to user pattern changes. As part of the training process I used battery data among the datasets provided by NASA, and built models incorporating linear regression and ARIMA, and predicted new battery data that contained user changes based on previously trained models. Therefore, as a result of the prediction, the linear regression is better at predicting some changes in SOH based on the user's pattern change if we have more battery datasets with a wide range of independent values. The ARIMA model can be used if we only have battery datasets with SOH data.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.2
• /
• pp.201-210
• /
• 2023

Deep learning-based face verificattion model show high performance and are used in many fields, but there is a possibility the user's face image may be leaked in the process of inputting the face image to the model. Althoughde-identification technology exists as a method for minimizing the exposure of face features, there is a problemin that verification performance decreases when the existing technology is applied. In this paper, after combining the face features of other person, a de-identified face image is created through StyleGAN. In addition, we propose a method of optimizingthe combining ratio of features according to the face verification model using HopSkipJumpAttack. We visualize the images generated by the proposed method to check the de-identification performance, and evaluate the ability to maintain the performance of the face verification model through experiments. That is, face verification can be performed using the de-identified image generated through the proposed method, and leakage of face personal information can be prevented.

• Composites Research
• /
• v.22 no.5
• /
• pp.1-7
• /
• 2009

The vehicle structure of Automated People Mover(APM) made of aluminum honeycomb sandwich with WR580INF4000 glass-fabric epoxy laminate facesheets was evaluated by structural test and finite element analysis. The test of the vehicle structure was conducted according to JIS E 7105. The structural integrity of vehicle structure was evaluated by stress, deflection and natural frequency obtained from dial-gauge and acceleration sensor. And the proposed finite element models were compared with the results of structural test. The results of finite element analysis showed good agreement with those of structural test. Also, in order to improve the stiffness of vehicle structure, the modified underframe model with reinforced side sill was proposed in design stage. The composite vehicle structures with modified underframe model had the improved structural stiffness about 44%.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.6 no.8
• /
• pp.577-584
• /
• 2016

UAV(Unmaned Aerial Vehicle), which is autonomous flight without pilots. Recently, UAV is being applied to various fields such as video recording, aerial photogrammetry. In particular, UAV is getting a lot of attention in the field of space-related information because of it's data acquisition speed and economic feasibility. But analytical study of an unmanned air-side technologies are lacking. In this study, the research of equipment for the unmanned aerial surveys and UAV technologies and trend analysis for generation of terrain model and ortho image effectively were performed. As a result, the ways to improve the utilization field of unmanned aerial surveying and processing of fixed-wing and rotary-wing unmanned aircraft. were suggested. If analytical research on generation of terrain models and ortho image will be performed, production efficiency of the geospatial information industry is expected to be significantly increased.

• Annual Conference of KIPS
• /
• 2020.05a
• /
• pp.504-507
• /
• 2020

ReLU(Rectified Linear Unit) 함수는 제안된 이후로 대부분의 깊은 인공신경망 모델들에서 표준 활성함수로써 지배적으로 사용되었다. 이후에 ReLU 를 대체하기 위해 Leaky ReLU, Swish, Mish 활성함수가 제시되었는데, 이들은 영상 분류 과업에서 기존 ReLU 함수 보다 향상된 성능을 보였다. 따라서 초해상화(Super Resolution) 과업에서도 ReLU 를 다른 활성함수들로 대체하여 성능 향상을 얻을 수 있는지 실험해볼 필요성을 느꼈다. 본 연구에서는 초해상화 과업에서 안정적인 성능을 보이는 EDSR(Enhanced Deep Super-Resolution Network) 모델의 활성함수들을 변경하면서 성능을 비교하였다. 결과적으로 EDSR 의 활성함수를 변경하면서 진행한 실험에서 해상도를 2 배로 변환하는 경우, 기존 활성함수인 ReLU 가 실험에 사용된 다른 활성함수들 보다 비슷하거나 높은 성능을 보였다. 하지만 해상도를 4 배로 변환하는 경우에서는 Leaky ReLU 와 Swish 함수가 기존 ReLU 함수대비 다소 향상된 성능을 보임을 확인하였다. 구체적으로 Leaky ReLU 를 사용했을 때 기존 ReLU 보다 영상의 품질을 정량적으로 평가할 수 있는 PSNR 과 SSIM 평가지표가 평균 0.06%, 0.05%, Swish 를 사용했을 때는 평균 0.06%, 0.03%의 성능 향상을 확인할 수 있었다. 4 배의 해상도를 높이는 초해상화의 경우, Leaky ReLU 와 Swish 가 ReLU 대비 향상된 성능을 보였기 때문에 향후 연구에서는 다른 초해상화 모델에서도 성능 향상을 위해 활성함수를 Leaky ReLU 나 Swish 로 대체하는 비교실험을 수행하는 것도 필요하다고 판단된다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.457-460
• /
• 2023

본 연구는 시설작물의 병충해 진단을 위해 딥러닝 모델을 응용한 인공지능 서비스 앱, Dr. Vegetable을 제안하고자 한다. 농업 현장에서 숙련된 농부는 한눈에 농작물의 병충해를 판단할 수 있지만 미숙련된 농부는 병충해 피해를 발견하더라도 그 종류와 해결 방법을 찾아내기가 매우 어렵다. 또한 아무리 숙련된 농부라고 할지라도 육안검사만으로 병충해를 조기에 발견하는 것은 쉽지 않다. 한편 시설작물의 경우 병충해에 의한 연쇄피해가 발생할 우려가 있으므로 병충해의 조기 발견 및 방제가 매우 중요하다. 즉, 농부의 경험에 따른 농작물 병해충 진단은 정확성을 장담할 수 없으며 비용과 시간적인 측면에서 위험성이 높다고 할 수 있다. 본 논문에서는 YOLOv5를 활용하여 상추, 고추, 토마토 등 농작물의 병충해를 진단하는 인공지능 서비스를 제안한다. 특히 한국지능정보사회진흥원이 운영하고 있는 AI 통합 플랫폼인 AI 허브에서 제공하는 노지 작물 질병 및 해충 진단 이미지를 사용하여 딥러닝 모델을 학습하였다. 본 연구를 통해 개발된 모바일 어플리케이션을 이용하여 실제 시설농장에서 병충해 진단 서비스를 적용한 결과 약 86%의 정확도, F1 Score 0.84, 그리고 0.98의 mAP 값을 얻을 수 있었다. 본 연구에서 개발한 병충해 진단 딥러닝 모델을 다양한 조도에서 강인하게 동작하도록 개선한다면 농업 현장에서 널리 활용될 수 있을 것으로 기대한다.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.13-20
• /
• 2024

This paper introduces a method for generating model images that can identify specific cylindrical medicine containers in videos and investigates data collection techniques. Previous research had separated object detection from specific object recognition, making it challenging to apply automated image stitching. A significant issue was that the coordinate-based object detection method included extraneous information from outside the object area during the image stitching process. To overcome these challenges, this study applies the newly released YOLOv8 (You Only Look Once) segmentation technique to vertically rotating pill bottles video and employs the ORB (Oriented FAST and Rotated BRIEF) feature matching algorithm to automate model image generation. The research findings demonstrate that applying segmentation techniques improves recognition accuracy when identifying specific pill bottles. The model images created with the feature matching algorithm could accurately identify the specific pill bottles.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.20 no.1
• /
• pp.113-119
• /
• 2014

Recently, R.O.K. Navy is increasing re-usability of design data and application of M&S(Modeling and Simulation) through the establishment of collaborative product development environment focused on Naval Ship Product Model(NSPM). As a result, the reliability of the result of design is getting better, and furthermore, a study to improve quality of construction through simulation of production/operation is in progress. Accordingly, the database construction of design data and the DB(Database) quality become important, but there was not research related to those or it was just initial state. This paper conducted research about system of the quality verification process of shape elements which compose NSPM based on the quality verification guideline of NSPM as the result of the precedent study. The hull surface was limited as verification object. The study to verify two things that application of basic drawing by the cad model of hull surface, and whether there is error in the geometric quality of cad model was progressed. To achieve this goal, the verification criteria and algorithm were defined and the prototype system which is based on was developed.

• The Journal of Internal Korean Medicine
• /
• v.34 no.4
• /
• pp.384-404
• /
• 2013

Objectives : Effects of Bogijetong-tang (BJT) on peripheral nerve regeneration have been reported in a previous study on BJT but additional study on a damaged peripheral neuropathy where its damage level is physically and chemically more severe was needed. Plus, this study was conducted because there haven't been any studies for BJT on central nerve regeneration. Methods : In order to check the effect on central nerve regeneration, the study on cerebellum cells was started and the sciatic nerve was used to observe the effects on a peripheral nerve which was severely damaged both physically and chemically. Nerve recovery effects were observed by analyzing target proteins such as phospho-extracellular signal-regulated kinase, ${\beta}1$ integrin, neurofilament 200, growth-associated protein-43, cyclin-dependent kinase 1, phospho-vimentin, phospho-Smad, and caspase 3. Results : The significant changes of target protein in cerebellum neurons have been observed. The changes of index protein on the axon regeneration and the nerve recovery in the sciatic nerve have been observed and the effects on cell protection were observed, as well. Conclusions : This study confirmed that BJT made a significant influence on nerve protection and recovery of a damaged peripheral neuropathy and it also made a possibility of its regeneration in a damaged central nerve injury.