• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.1
• /
• pp.8-16
• /
• 2016

In the field of computer vision and robotics, bin picking is an important application area in which object pose estimation is necessary. Different approaches, such as 2D feature tracking and 3D surface reconstruction, have been introduced to estimate the object pose accurately. We propose a new approach where we can use both 2D image features and 3D surface information to identify the target object and estimate its pose accurately. First, we introduce a label detection technique using Maximally Stable Extremal Regions (MSERs) where the label detection results are used to identify the target objects separately. Then, the 2D image features on the detected label areas are utilized to generate 3D surface information. Finally, we calculate the 3D position and the orientation of the target objects using the information of the 3D surface.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2871-2877
• /
• 2014

Real-time communication and diversification of multimedia application services using high-capacity data over the Internet demand high speed connection, and real-time and high-speed access internet service such as multimedia make Internet traffic increase rapidly. Meanwhile, QoS (Quality of Service) of internet service is not satisfied. In this situation, ISP (Internet Service Provider) has been required to improve service quality and extend network according to the user's needs. This request includes issue of extending network focused on increasing router and the number of routing table as well as simply extending bandwidth.

• Journal of the East Asian Society of Dietary Life
• /
• v.24 no.4
• /
• pp.457-464
• /
• 2014

This study investigated the effects and interrelationship on sensual behavior and wine information sources in selection attributes of wine. Selection attributes of wine were categorized into four variables: sensual behavior, recommend information, label information, and wine information. The study showed that for sensual behavior variable, "taste" was the most influential item as compared to "food harmony", "mood harmony" and "partner's choice". For recommendation information, label information and wine information, "specialist", "grape varieties" and "brand" had the most significant effects. The study performed factor analysis on sensual behavior and wine information sources. The cumulative variance was 74.197%, implying that all four variables incorporated appropriate items. In the reliability analysis, all four variables showed Cronbach's ${\alpha}$ values above 0.6. In the analysis of the causal relations using a structural model, the effects of customers' sensual behavior on wine information sources was further investigated. The model verified that taste, food harmony, mood harmony and partner's choice, which are items of sensual behavior, had significant impacts when choosing wine. Sensual behavior influenced all wine information sources, which customers utilize in decision-making. Among these sources, sensual behavior had the biggest effects on recommendation information, followed by wine information and label information.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37C no.12
• /
• pp.1195-1210
• /
• 2012

In this paper, we propose a new signaling protocol as an extension of Generalized Multi-Protocol Label Switching (GMPLS) based PCE protocol for control carrier ethernet. The proposed protocol is not only compatible with the GMPLS PCE standard protocol, but also provides the topology discovery and configuration. In order to verify interoperability, we implement the proposed protocol as well as the system integrating functions including commercial system for testbed. In addition, we have simulated topology discovery test for proposed protocol performance.

• Journal of the Institute of Convergence Signal Processing
• /
• v.16 no.1
• /
• pp.1-8
• /
• 2015

In many image processing tasks, connected component labeling (CCL) is performed to extract regions of interest. CCL was usually done in a sequential fashion when image resolution was relatively low and there are small number of input channels. As image resolution gets higher up to HD or Full HD and as the number of input channels increases, sequential CCL is too time-consuming to be used in real time applications. To cope with this situation, parallel CCL framework was introduced where multiple cores are utilized simultaneously. Several parallel CCL methods have been proposed in the literature. Among them are NSZ label equivalence (NSZ-LE) method[1], modified 8 directional label selection (M8DLS) method[2], and HYBRID1 method[3]. Soh [3] showed that HYBRID1 outperforms NSZ-LE and M8DLS, and argued that HYBRID1 is by far the best. In this paper we propose an improved hybrid parallel CCL algorithm termed as HYBRID2 that hybridizes M8DLS with label backtracking (LB) and show that it runs around 20% faster than HYBRID1 for various kinds of images.

• Journal of Power System Engineering
• /
• v.18 no.6
• /
• pp.70-76
• /
• 2014

Surely fuel economy(F.E.) label is needed to meet consumer's right to know. Korea has developed and adopted F.E. correction equation similar to the feeling, because consumer complain about F.E. label higher than they feel. Recently, through continuous research and development high-efficiency & performance green car like HEV, PHEV and EV are being sold. In this situation, it is needed to know whether the current equation can reflect a part of their improved technic or not. In this paper, to review current equation through test using hybrid vehicles on 5-cycle SOC, correction equation and F.E were discussed. The result show HEV didn't meet the SOC error standard on US06. Also, HEV has bigger F.E. difference between FTP-75 and 5-cycle than conventional vehicles. However, the correction equation that include HEV almost same with current one.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.04a
• /
• pp.686-691
• /
• 2000

최근 들어 인터넷의 사용이 급격히 증가함에 따라 인터넷의 백본망에는 패킷 전달의 고속성, 확장성 뿐만 아니라 QOS 보장을 요구하게 되었다. 효과적인 인터넷 백본망을 구현하기 위해 본 논문에서는 Multi-Protocol Label Switching(MPLS)를 고려하고 MPLS 에서 인터넷 Quality Of Service(QOS) 서비스 모델 중 백본망에 적합한 Differentiated Service(DS) 모델을 그대로 제공하기 위해 MPLS 와 DS 의 연동에 대한 문제를 다룬다. 우선 MPLS 에서 QOS 제공을 위한 기능요소를 정성적으로 분석한다. 그리고 두 DS 망 가운데 ATM 기반의 MPLS 망을 두고 있는 경우를 가정하여, 종단대 종단간 DS 제공을 위하여 DS 망과 MPLS 망에서의 시그널링 과정을 제안한다. 즉, DS 의 BB 로부터 SLA 를 위한 신호를 MPLS 의 입구 라우터에서 해석하여 DS Type-Length-Value(TLV)를 만들고 이를 이용하여 Constraint-based Label Distribution Protocol(CR-LDP) 시그널링을 통해 마다 Label-Only-Inferred-PSC LSP(L-LSP)를 설립하여 MPLS 에서 DS QOS 를 그대로 제공한다. 또한 입구 DS BB 로부터의 정보를 MPLS 망을 통해 BB 에게 보낸다.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.545-550
• /
• 2022

In Deep Learning method, it is well known that it requires large amount of data to train the deep neural network. And it also requires the labeling of each data to fully train the neural network, which means that experts should spend lots of time to provide the labeling. To alleviate the problem of time-consuming labeling process, some methods have been suggested such as weak-supervised method, one-shot learning, self-supervised, suggestive learning, and so on. In this manuscript, those methods are analyzed and its possible future direction of the research is suggested.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.12
• /
• pp.555-560
• /
• 2021

Recently, as people's interest in facial skin beauty has increased, research on skin disease recognition for facial skin beauty is being conducted by using deep learning. These studies recognized a variety of skin diseases, including acne. Existing studies can recognize only the single skin diseases, but skin diseases that occur on the face can enact in a more diverse and complex manner. Therefore, in this paper, complex skin diseases such as acne, blackheads, freckles, age spots, normal skin, and whiteheads are identified using the Inception-ResNet V2 deep learning mode with multi-label classification. The accuracy was 98.8%, hamming loss was 0.003, and precision, recall, F1-Score achieved 96.6% or more for each single class.

• Korean Journal of Clinical Pharmacy
• /
• v.23 no.4
• /
• pp.327-333
• /
• 2013

Background: Antihistamine and anti-allergy medications are widely used during pregnancy. Reading label information is one of the easiest ways to get safety information. But there are content gaps among countries. Objective: To compare the risk level and the recommendation level of antihistamine/anti-allergy drug's label information in pregnant women among Korea, the USA, the UK, and Japan. Method: Study drugs of antihistamine/anti-allergy medications were selected according to Korea drug classification codes. Based on the label information of selected product, risk level was classified into 5 categories as follows: 'Definite', 'Probable', 'Possible', and 'Unlikely', 'Unclassified' according to the level of evidence. Recommendation level was classified into 4 categories as follows: 'Contraindicated', 'Cautious', 'Compatible', and 'Unclassified'. Frequency and proportion were presented according to the each category. To estimate agreement of each category among 4 countries, percent agreement and kappa (k) coefficient were calculated. Results: Total 13 drug ingredients were selected for antihistamine/anti-allergy medications. In risk level, Korea (46%) and Japan (69%) were mostly classified in the category of 'Unclassified', but 'Unlikely' category was more frequent in the UK (62%) and the USA (46%). In recommendation level, the proportion of 'Contraindicated' was highest in Korea (46%) compared to other countries. In contrast, the category of 'Cautious' was 77%-85% in the USA, the UK, and Japan. The percent agreement for risk level was highest in the USA-UK (54%). The recommendation level of Korea-USA showed lowest agreement for percent agreement (46%) and kappa coefficient (k=0.02). Conclusion: We confirmed the differences among safety information provided by four different countries. 'Contraindicated' was more likely in Korea compared with other countries.