• Korean Journal of Remote Sensing
• /
• v.38 no.2
• /
• pp.199-213
• /
• 2022

The unsupervised domain adaptation can solve the impractical issue of repeatedly collecting high-quality training data every year for annual crop classification. This study evaluates the applicability of deep learning-based unsupervised domain adaptation models for crop classification. Three unsupervised domain adaptation models including a deep adaptation network (DAN), a deep reconstruction-classification network, and a domain adversarial neural network (DANN) are quantitatively compared via a crop classification experiment using unmanned aerial vehicle images in Hapcheon-gun and Changnyeong-gun, the major garlic and onion cultivation areas in Korea. As source baseline and target baseline models, convolutional neural networks (CNNs) are additionally applied to evaluate the classification performance of the unsupervised domain adaptation models. The three unsupervised domain adaptation models outperformed the source baseline CNN, but the different classification performances were observed depending on the degree of inconsistency between data distributions in source and target images. The classification accuracy of DAN was higher than that of the other two models when the inconsistency between source and target images was low, whereas DANN has the best classification performance when the inconsistency between source and target images was high. Therefore, the extent to which data distributions of the source and target images match should be considered to select the best unsupervised domain adaptation model to generate reliable classification results.

• Korean Journal of Cognitive Science
• /
• v.34 no.2
• /
• pp.111-151
• /
• 2023

It is reported that binge drinkers show cognitive impairment similar to alcohol use disorder patients. A previous studies using neuropsychological tests and brain imaging techniques to investigate the visual perception of alcohol use disorder patients reported that they had global-local visual perception defects. Although the neurological basis for the global-local visual perception deficit in the heavy drinking group has been presented, there are no studies to date that have investigated the global-local visual perception in the heavy drinking group. This study investigated local-biased visual perception in female college students with binge drinking (BD) using event-related potentials (ERPs). Based on the scores of the Korean version of Alcohol Use Disorder Identification Test and the Alcohol Use Questionnaire, participants were assigned into BD (n=25) and non-BD (n=25) groups. Local-global visual processing was assessed using a local-global paradigm, in which large stimuli (global level) composed of small stimuli (local level) were presented. The stimuli presented at global and local levels were either congruent or incongruent. The behavioral results exhibited that the BD and non-BD groups did not differ in terms of accuracy and response time. In terms of ERPs, the BD and non-BD groups did not show difference in N100, P150 and N200 amplitude. However, the BD group showed significantly smaller P300 amplitude than non-BD group especially in the local condition. In addition, a negative correlation between P300 amplitude and binge drinking score was observed, i.e., severer binge drinking smaller P300 amplitude. The P300 is known to reflect cognitive inhibition and attentional allocation. In the global-local paradigm, the local condition required to attend to local target while ignoring global non-target. Therefore, the present results indicate that female college students with BD do not have local-biased visual processing, instead they seem to have difficulties in inhibition of irrelevant stimuli.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.150-150
• /
• 2011

최근 수자원분야에서 원격탐사(RS) 및 지형정보시스템(GIS)의 비중이 높아짐에 따라 이를 이용하여 1,2 차원 수리분석 향상을 위한 활용이 증가하고 있다. 그러나 현재 사용하고 있는 일반적인 DEM은 하천부분에 대한 고도의 불일치로 인해 수리분석시 정밀한 분석을 하기에는 어려움이 따르므로 이에 대한 개선이 필요하다. 본 연구에서는 현재 사용하고 있는 USGS DEM과 낙동강 측량 자료를 이용하여 낙동강 유역 정밀 DEM 제작 방안을 제시하고자 한다. 측량자료를 이용하여 하천의 DEM을 생성한 후, 자체적으로 개발한 알고리즘을 이용하여 하천 DEM과 USGS DEM을 접합하여 낙동강 유역의 정밀 DEM을 제작한다. 이는 1,2차원 수리모델을 이용하여 분석할 때 결과를 더욱 향상시킬 것으로 기대한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10b
• /
• pp.87-89
• /
• 1998

정보 기술의 발달로 문헌정보의 양이 증가하고 있지만 정확하고 빠른 정보를 지역적으로 분산된 서지정보 데이터베이스에서 찾기에 이미 개발된 방식으로는 시간이 많이 걸린다. 현재 사용중인 텔넷(telnet)을 통한 방식이나. 웹/CGI 기술을 이용한 방식은 문헌정보가 존재하는 각각의 서버를 사용자가 한번에 하나씩 접속하여 검색을 해야 하므로 검색시간이 많이 걸릴 뿐 아니라, 사용자는 각 서버마다 상이한 검색방법을 익혀야 한다. 최근에 이러한 단점들을 개선하기 위해 한곳에서 종합적인 서지 및 본문정보검색을 하는 통합 검색시스템이 중앙 집중형 시스템을 모델로 하여 구축되고 있으나 이 방식 역시, 자료의 불일치성, 시스템 공장과 과부하등 여러 문제점들을 내포하고 있다. 따라서 본 논문에서는 이러한 문제들의 해결책으로 분산 서지정보 통합검색시스템 DBIRS를 제안하고 구현한다. DBIRS는 웹 환경하에서 CORBA 기술과 Java언어를 이용함으로서, 단일 사용자 인터페이스로 병렬 복수질의를 수행하여 동시에 각 도서관에서 자료를 찾을 수 있다. 또한 객체를 기반으로 하는 표준 모델을 사용하여 소프트웨어의 개발과 확장을 가능하게 하고, 향후 멀티미디어 데이터나 웹과 같은 비 정형 데이터 시스템과의 통합을 쉽게 해준다.

• Journal of Food Hygiene and Safety
• /
• v.38 no.6
• /
• pp.496-507
• /
• 2023

This study investigated species identification and labeling compliance of 48 shrimp products sold in the Korean online markets. Species identification was conducted using the standard DNA barcoding method, using the cytochrome c oxidase subunit I gene. The obtained sequences were compared with those deposited in the NCBI GenBank and BOLD Systems databases. Additionally, phylogenetic analysis was performed to further verify the identified shrimp species. Consequently, 16 shrimp species were identified, including Penaeus vannamei, Pandalus borealis, Palaemon gravieri, Leptochela gracilis, Penaeus monodon, Pleoticus muelleri, Metapenaeopsis dalei, Euphausia pacifica, Lebbeus groenlandicus, Trachypenaeus curvirostris, Argis lar, Metanephrops thomsoni, Metapenaeopsis barbata, Alpheus japonicus, Penaeus chinensis, and Mierspenaeopsis hardwickii. The most prevalent species was Penaeus vannamei, found in 45.8% of the analyzed products. A significant mislabeling rate of 72.9% was found; however, upon excluding generic names such as shrimp, the mislabeling rate reduced to 10.4%. The mislabeling rate was higher in highly-processed products (89.3%) compared with that in minimally-processed products (50%). No correlation was found between the country of origin and mislabeling rate. The results of this study provide crucial data for future monitoring of shrimp products and improving the labeling of shrimp species in Korea.

• Land and Housing Review
• /
• v.6 no.4
• /
• pp.239-239
• /
• 2015

최근 국내 주택시장에서의 관심은 수요와 공급의 불일치 해소이다. 이를 위해서는 정확한 수요권역 설정에 의한 사업지구별 수요예측이 필수적이다. 본 연구에서는 주택시장권역 설정 시 선행연구에서 사용하는 가구이동자료 기준의 수요권역과 청약자 주소지에 의한 실제 수요권역의 차이를 비교분석 하였다. 수도권 3개 택지개발지구를 대상으로 한 실증분석 결과, 청약자료에 의한 실제 수요권역은 가구이동자료 기준의 수요권역에 비해 확대되는 것으로 나타나 기존의 가구이동자료 기준에 의한 수요권역 설정의 보완이 필요한 것으로 판단된다. 특히 사업지구와 서울 도심과의 거리, 핵심 수요권과 배후수요권의 수요구성비와 같은 입지경쟁력, 개발규모(신도시급 여부) 등의 변수가 정확한 수요권역의 설정을 위해 우선적으로 보완되어야 할 변수로 나타났다. 또한 동일 사업지구에 있어서도 임대/분양과 같은 주택유형, 주택의 규모에 따라 수요권역의 차이를 보였다. 아울러 부동산경기 역시 수요권역의 범위형성에 영향을 미치고 있으며 정확한 수요권역 분석을 위해서는 주택가격, 사업지구 입지경쟁력 등이 함께 복합적으로 고려되어야 할 필요성이 확인되었다.

• Environmental and Resource Economics Review
• /
• v.29 no.2
• /
• pp.113-144
• /
• 2020

Throughout various data sources, it is widely observed that air quality in South Korea has become improved. Koreans, however, insist that their health status and economic burden due to worsened air quality get degenerated. This study aims to explain the mismatch between perception and measured data, air pollution-led medical expenses, and consumption behaviors in the economics perspectives. First, we demonstrated data-driven evidence of mismatch in the perceived severity of air pollution and its enhancement in measured data. Second, using the health demand model, we theoretically derived and empirically showed a co-rising relation between air pollution severity and medical expenses. Last, we analyzed that the perception led to increased defensive expenditures in consumption. This result implies the possibility of overestimation in air pollution impacts on socioeconomic losses and its possible reverse interpretation from increased social benefit after improved air quality. Our results recommend policy consideration to strengthen air quality standards, to support socially vulnerable groups regarding defensive expenditures, and to improve the accessibility and credibility of air pollution information.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.9 no.1
• /
• pp.46-55
• /
• 2006

The purpose of this study is a development of an automatic mosaicing for applying to large number of airborne multispectral images, which reduces manual operation by human. 2436 airborne multispectral images were acquired from DuncanTech MS4100 camera with three bands; green, red and near infrared. LIDAR(LIght Detection And Ranging) data and GPS/INS(global positioning system/inertial navigation system) data were collected with the multispectral images. First, the multispectral images were converted to image patterns by unsupervised classification. Their patterns were compared with those of adjacent images to derive relative spatial position between images. Relative spatial positions were derived for 80% of the whole images. Second, it accomplished an automatic mosaicing using GPS/INS data and unsupervised classification. Since the time of GPS/INS data did not synchronized the time of readout images, synchronized GPS/INS data with the time of readout image were selected in consecutive data by comparing unsupervised classified images. This method realized mosaicing automatically for 96% images and RMSE (root mean square error) for the spatial precision of mosaiced images was only 1.44 m by validation with LIDAR data.

• The KIPS Transactions:PartD
• /
• v.12D no.4 s.100
• /
• pp.657-662
• /
• 2005

The Non-Sampling Error can happen any time by means of the intended or unintended error by the interviewer or respondent, but it is very difficult to find the error in survey database because it can hardly be computed mathematically and systematically. Until now, we have found it accidentally through the simple relation between the items or through the inspection from the random field. Therefore we introduced an heuristic methodology that can detect the interviewer's error by statistical decision-making or data mining techniques with a case study. It will be helpful so as to improve the statistical duality and provide efficient field management for the supervisor.

• Journal of Korean Society for Geospatial Information Science
• /
• v.23 no.1
• /
• pp.15-22
• /
• 2015

To create a wind resource map, we need a contour map, a roughness map and wind data. We need a land cover map for the roughness map of these data. A land cover map represents the area showing similar characteristics after color indexing based on the scientific method. The features of land cover is classified by Remote sensing technique. In this study, we verified the application of the NDVI technique is reasonable after we created the wind resource map using roughness maps by unsupervised classification and NDVI technique. As a result, the wind resource map using the NDVI technique showed a 60% accordance rate and difference in class less than one. From the results, The NDVI technique is found alternative to create roughness maps by the unsupervised classification.