• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.848-851
• /
• 2019

이미지 형식으로 등록되는 상표의 특성상 상표의 검색에는 어려움이 따른다. 특허청은 도형 상표의 검색을 용이하게 하기 위해 상표가 포함하고 있는 구성요소에 도형분류코드를 부여한다. 하지만 도형 상표에 포함된 이미지를 확인하고 분류코드를 부여하는 과정은 사람이 직접 수행해야 한다는 어려움이 따른다. 이에 본 논문에서는 딥러닝을 이용하여 자동으로 도형 상표 내 객체를 인식하고 분류코드를 부여하는 방안을 제안한다. DenseNet을 이용하여 중분류를 먼저 예측한 후 각 중분류에 해당하는 Faster R-CNN 모델을 이용하여 세분류 예측을 수행하였다. 성능평가를 통해 비엔나분류 중분류별 평균 74.49%의 예측 정확도를 확인하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06a
• /
• pp.262-264
• /
• 2012

유비쿼터스 지능형 공간에서 사용자의 개입 없이도 자율 실행하는 서비스 플러그 앤 플레이 시스템(Service Plug and Play System)은 기능적인 모델로서 총체적인 시스템의 협력 프로토콜 부재로 그 동시성 및 병렬성을 반영한 실행관리를 하기 어렵다. 이는 시스템을 구성하는 모듈로, 컨텍스트 수집기, 서비스 실행기, 공간관리기, 모바일 제어기 등이 UML의 시퀀스 다이어그램으로 모델링하여 각 기기의 기능과 실행흐름을 독립적으로 표현할 뿐, 기기간의 상호작용, 특히 병렬 및 동시 실행 알고리즘을 설계하고 검증하는 절차의 부재 때문이다. 본 논문에서, 각 기기의 실행 흐름 및 상호작용을 Petri Net을 이용하여 모델링함으로써, 시스템을 구성하는 각 디바이스의 상호작용 및 실행 흐름을 정의 하고 검증한다.

• Journal of Information Management
• /
• v.29 no.2
• /
• pp.1-17
• /
• 1998

The Inno-NET, which is based on the Internet and up-to-date information technology, is a consolidated system to provide business-related information and to solve the difficulties in business activities. In particular, the Inno-NET is an information system to integrate such business information for enterprise as start-up, financing, technology development, recruiting, site acquisition, marketing, trade and investment and so on, in order to help companies solve their problems and make up for their weak points. Accordingly, the Inno-NET will systematically connect supporting agencies and innovative enterprises, thus guaranteeing the exchange of information among them and coordinating the business. The supporting agencies will be included government offices, public organizations, research institutes, universities, and industrial associations all over the ministries. In short, Inno-NET forms a real-time decision making system among government, industries, universities and research institutes. by elt:ictronic networking on the Internet.

• Journal of Biomedical Engineering Research
• /
• v.38 no.1
• /
• pp.25-31
• /
• 2017

Testing TB in chest X-ray images is a typical method to diagnose presence and magnitude of PTB lesion. However, the method has limitation due to inter-reader variability. Therefore, it is essential to overcome this drawback with automatic interpretation. In this study, we propose a novel method for detection of PTB using SegNet, which is a deep learning architecture for semantic pixel wise image labelling. SegNet is composed of a stack of encoders followed by a corresponding decoder stack which feeds into a soft-max classification layer. We modified parameters of SegNet to change the number of classes from 12 to 2 (TB or none-TB) and applied the architecture to automatically interpret chest radiographs. 552 chest X-ray images, provided by The Korean Institute of Tuberculosis, used for training and test and we constructed a receiver operating characteristic (ROC) curve. As a consequence, the area under the curve (AUC) was 90.4% (95% CI:[85.1, 95.7]) with a classification accuracy of 84.3%. A sensitivity was 85.7% and specificity was 82.8% on 431 training images (TB 172, none-TB 259) and 121 test images (TB 63, none-TB 58). This results show that detecting PTB using SegNet is comparable to other PTB detection methods.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.4
• /
• pp.81-88
• /
• 2018

In this paper, we present some problems with the Groupware Messenger functionality based on dot net 2.0 and implement a new design structure to solve them. They include memory leakage, slow processing, and client window memory crash. These problems resulted in the inconvenience of using instant messaging and the inefficient handling of office tasks. Therefore, in this paper, instant messaging functionality is implemented according to a new design architecture. The new system upgrades dot net 4.5 for clients and deploys the new features based on MQTT for the messenger server. We verify that the memory leak problem and client window memory crash issues have been eliminated on the system with the new messenger functionality. We measure the amount of time it takes to bind data to a set of messages and evaluate the performance, compared to a given system. Through this comparative evaluation, we can see that the new system is more reliable and performing.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.132-133
• /
• 2021

XR(가상 및 확장현실)에서의 리얼타임 그래픽은 조명이 제 역할을 하지 못하면 인물이 공간감이 느껴지지 않거나 스킨톤이 맞지 않아 이질감을 초래한다. 본 논문에서는 Art-Net 시스템을 활용하여 실제 조명과 가상조명에서의 배경 및 인물의 컬러를 동기화하는 방법을 제안하였다. 제안된 Art-Net 시스템을 통해 인물의 공간감을 향상 시킬 수 있을 뿐만 아니라 보는 이의 몰입감을 높였다.

• Korean Journal of Materials Research
• /
• v.12 no.12
• /
• pp.955-961
• /
• 2002

2.7vol%TiNi/6061 Al composites with TiNi shape memory alloy as reinforcement were fabricated by vacuum hot press. It was investigated by OM, SEM, EPMA and XRD analysis for the behavior of diffusion layer formation on various heat treatment condition. Thickness of diffusion layer was increased proportionally according to heat treatment time. The layer was formed by the mutual diffusion of TiNi and Al. The diffusion rate from TiNi fiber to Al matrix was faster than that of reverse diffusion path. The more diffused layer was formed in Al matrix. The diffusion at interface layer was consisted of $A1_3$Ti, $Al_3$Ni analyzed by EPMA, XRD results.

• 한국가스학회:학술대회논문집
• /
• 2001.10a
• /
• pp.150-153
• /
• 2001

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.56 no.2
• /
• pp.147-154
• /
• 2020

The questionnaire survey was carried out to collect basic data to identify the cause of the risk that did not appear in the current data provided by the fishermen's occupational accidents of the National Federation of Fisheries Cooperative (NFFC) to the onsite specialist, such as a fishing master, chief engineer and fisher in stow net fishing vessel from June to July 2018. The results are as follows. A total of 134 cases were surveyed, including 53 cases in Yeosu, 44 cases in Mokpo, 30 cases in Boryeong and 7 cases in Gunsan. Approximately 60% of the respondents on board the stow net fishing vessel were more than 20 years old, and the boarding experience was seasoned and suitable for the respondents. In the 4M analysis by safety accident factor during fishing work, the mechanical factors included 51 cases (25.0%) of decrepit equipment or poor maintenance and 49 cases (24.0%) of systemic error of fishing equipment. The environmental factors exhibited 71 cases (33.5%) of poor communication by mixed manning of foreign fishermen and 63 cases (29.7%) of poor environment such as heavy weather etc. The human factors showed 78 cases (37.3%) of personnel's carelessness and 45 cases (21.5%) of unskillfulness of operating net and haulers. The management factors exhibited 59 cases (32.1%) of work practice of poor safety precautions and 56 cases (30.4%) of inadequacy of education on hazard factor. The results are expected to contribute to the creation of a safe operating environment for stow net fishery, such as a fisher boarding a stow net fishing vessel, enabling the analysis of major causes of working safety accidents by cause.

• The Korean Journal of Applied Statistics
• /
• v.23 no.1
• /
• pp.207-218
• /
• 2010

Generalized additive model(GAM) is the statistical model that resolves most of the problems existing in the traditional linear regression model. However, overfitting phenomenon can be aroused without applying any method to reduce the number of independent variables. Therefore, variable selection methods in generalized additive model are needed. Recently, Lasso related methods are popular for variable selection in regression analysis. In this research, we consider Group Lasso and Elastic net models for variable selection in GAM and propose an algorithm for finding solutions. We compare the proposed methods via Monte Carlo simulation and applying auto insurance data in the fiscal year 2005. lt is shown that the proposed methods result in the better performance.