• The Journal of the Korea Contents Association
• /
• v.19 no.12
• /
• pp.164-172
• /
• 2019

In this paper, we implement a wheeled mobile robot that accurately and autonomously finds the optimal route from the starting point to the destination point based on computer vision in a complex indoor environment. We get a number of waypoints from the starting point to get the best route to the target through deep reinforcement learning. However, in the case of autonomous driving, the majority of cases do not reach their destination accurately due to external factors such as surface curvature and foreign objects. Therefore, we propose an algorithm to deepen the waypoints and destinations included in the planned route and then correct the route through the waypoint recognition while driving to reach the planned destination. We built an autonomous wheeled mobile robot controlled by Arduino and equipped with Raspberry Pi and Pycamera and tested the planned route in the indoor environment using the proposed algorithm through real-time linkage with the server in the OSX environment.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.2
• /
• pp.449-454
• /
• 2020

This paper proposes a machine learning-based emotion analysis system that detects a user's depression through their SNS posts. We first made a list of keywords related to depression in Korean, then used these to create a training data by crawling Twitter data - 1,297 positive and 1,032 negative tweets in total. Lastly, to identify the best machine learning model for text-based depression detection purposes, we compared RNN, LSTM, and GRU in terms of performance. Our experiment results verified that the GRU model had the accuracy of 92.2%, which is 2~4% higher than other models. We expect that the finding of this paper can be used to prevent depression by analyzing the users' SNS posts.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.1
• /
• pp.151-159
• /
• 2013

It is well known that disease of human and economic traits of livestock are affected a lot by gene combination effect rather than a single gene effect. But existing methods have disadvantages such as heavy computing, many expenses and long time. In order to overcome those drawbacks, SNPHarvester was developed to find the main gene combinations among the many genes. In this paper, we used the superior gene combination which are related to the quality of the Korean beef cattle among sets of SNPs by SNPHarvester, and identified the superior genotypes using radial graph which can enhance various qualities of Korean beef among selected SNP combinations.

• Clean Technology
• /
• v.20 no.1
• /
• pp.88-94
• /
• 2014

During the past decades much attention has been paid to the desulfurization of diesel oil which is important as a source for the fuel cells to prevent the sulfur poisoning of both diesel steam reforming catalyst and electrode of fuel cell. Although alternative desulfurization techniques have been investigated, desulfurization for ultra-low sulfur diesel (ULSD) is still challenged. Therefore, this research focuses on the desulfurization of commercial ULSD for the application to molten carbonate fuel cell (MCFC). Herein, the performances of several kinds of commercial adsorbents based on activated carbons, zeolites, and metal oxides for desulfurization of ULSD were screened. The results showed that metal oxides based materials can feasibly reduce sulfur concentration in ULSD to a level of 0.1 ppmw while activated carbons and zeolites did not reach this level at current conditions.

• Journal of Digital Contents Society
• /
• v.15 no.5
• /
• pp.651-661
• /
• 2014

The complexity to search a specific content over TV platform is drastically increasing. Based on previous studies from computer data management system, we propose a new method that helps users to search and select content effectively. In general, contents over computers are represented by spatial metaphor, which replicates our physical environment and value systems about space. We suggest 3D lattice structure to construct information space for TV platform. Users can infer relevance between contents via special clue in information space, so as to select content more easily. Also, they can search contents through its temporal property that also represented in space. We make full use of our natural capability that can reduce additional overload to learning new interface. The results of this study can be significant and heuristic contributions, as they can be applied to diverse service areas utilizing video contents.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.8 no.4
• /
• pp.43-54
• /
• 2004

A visualization study was carried out to investigate the effects of phase difference qualitatively by examining wake pattern on the phase difference of a dragonfly type wing model. The model was built with scaled-up, flapping wings composed of a paired wing with fore- and hind-wings in tandem that mimick the wing form of a dragonfly. The present study was conducted by using the smoke-wire technique and an electronic device below the tandem wings was mounted to find the exact wing position angles. Uncertainties in wing position angle are about $\pm$$1.0^{\cire}$ and instantaneous wing positional angle varies from $-16.5^{\cire}$ to $+22.8^{\cire}$. The tests were made at phase differences between the fore-wing and hind-wing at $0^{\cire}$, $90^{\cire}$, $180^{\cire}$ and $270^{\cire}$. The results show that Karman vortex structures were produced at phase differences of $90^{\cire}$, $180^{\cire}$ and $270^{\cire}$, but Karman vortex structures were not observed at the phase difference of $0^{\cire}$.

• Korean journal of applied entomology
• /
• v.23 no.2 s.59
• /
• pp.126-131
• /
• 1984

Twenty one fungi were detected from 26 rice seed samples obtained from Yeongnam Crop Experiment Station, Honam Crop Experiment Station, Chungnam Provincial Office of Rural Development and farmers of Chungnam Province. Gerlachia oryzae was one of frequently detected fungi with $1.0\~45.0\%$ of detection ratio from 22 rice seed samples. The deep-freezing method was selected as the most suitable for routine seed health testing when the blotter method, deep-freezing method and agar plate method were evaluated for efficiency of detecting seed-borne G. oryzae from rice seeds. Our data obtained by seed component plating technique suggested that G. oryzae was present not only on chaff, in endosperm and in seed coat, but also in the embryo. Seed-borne G. oryzae caused seed rot, seedling blight and symptom of brownish discoloration on coleoptile primary and 2nd leaf when the infected seeds were sown in agar of test tube or in soil.

• Spatial Information Research
• /
• v.15 no.3
• /
• pp.279-290
• /
• 2007

The purpose of this study is to find the policy implications for carrying out the 3D GIS in local governments. For this, we made up questions to the local officials related to GIS in local governments and analyzed the questions by the hierarchy of local government. As the result of this, we found that the recognition about 3D GIS concept and the necessity of 3D GIS development was shown at the high rate without the hierarchy, but 3D GIS construction and propulsion was differently shown by the hierarchy. In conclusion, we found it was necessary to develop the 3D GIS with the different approach by the hierarchy of the local government. Especially, the 3D GIS applications has to be focussed on the improvements of the existing GIS in local government. In carrying out the 3D GIS, the support programs considered the local government's peculiarity should have to accompany.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.3
• /
• pp.35-44
• /
• 2011

Ontologies are playing an important role in semantic web which is emerging as a next stage of the web revolution because various kinds of metadata are described in ontologies. Correspondingly, many query languages like SPARQL, RDQL etc. have been proposed for querying these ontologies. But users have to know the structures and resource names of ontologies completely to get search results even if they have expertise on complex formal logic and syntax of the query languages. Especially, casual users do not know the resource names and may use different words from resource names when they write their query language. This vocabulary gap problem have to be solved to raise the success rate. In this paper, an approach for translating user's search words to corresponding resource names has been proposed. This approach uses semantic similarity between user created search words and ontology resource names.

• 한국IT서비스학회:학술대회논문집
• /
• 2008.05a
• /
• pp.425-430
• /
• 2008

USN의 발전으로 인해 인간 중심의 정보화 사회에서 사물 간에도 정보들을 유기적으로 결합하여 활용하는 유비쿼터스 컴퓨팅 사회로 급격히 변모하고 있다. 이러한 유비쿼터스 컴퓨팅 인프라를 성공적으로 구축하기 위해서는 센서노드에 대한 기술의 발전과 함께 USN 미들웨어에 대한 연구가 필요하다. USN의 핵심기술 중 하나인 USN 미들웨어는 시스템 소프트웨어로써 USN 응용에서 USN 계층의 자원과 데이타를 쉽게 활용할 수 있도록 다양한 기본 서비스를 제공한다. USN 디렉토리 서비스는 이와 같은 USN 미들웨어의 핵심 컴포넌트로서 USN 응용이 필요로 하는 USN 자원에 대한 정보를 검색하는 기능을 수행한다. 분산된 USN 자원에 대한 메타데이타의 효율적인 검색은 분산된 환경에서의 USN 디렉토리 서비스의 핵심요소이며 최근 이에 대한 연구가 보이고 있다. 분산 USN 디렉토리 서비스에서의 메타데이타 검색 모델은 크게 중앙 집중식(Centralized) 모델, 계층식(Hierarchical) 모델, P2P(Peer-to-Peer) 모델이 있다. 그러나 현재 메타데이타 검색 모델에 대한 연구는 원하는 메타데이타를 찾기 위한 서버 검색 회수만을 대상으로 하고 있으며 실제 단말 간 응답시간에 영향을 미치는 지연시간은 고려하지 않고 있다. 본 논문은 실제 단말 간 응답시간에 영향을 미치는 검색회수와 지연시간을 고려한 하이브리드 모델을 제안하고, 하이브리드 모델이 계층 모델에서 검색 시 불필요한 검색횟수의 증가와 P2P 모델의 광범위한 지역에서 응답시간이 증가하게 되는 단점을 개선할 수 있음을 보인다.