• 한국지형학회지
• /
• 제26권3호
• /
• pp.53-67
• /
• 2019

The purpose of this study is to identify the quality of landslide susceptibility in a landslide-prone area (Jinbu-myeon, Gangwon-do, South Korea) by spatial prediction modeling approach and compare the results obtained. For this goal, a landslide inventory map was prepared mainly based on past historical information and aerial photographs analysis (Daum Map, 2008), as well as some field observation. Altogether, 550 landslides were counted at the whole study area. Among them, 182 landslides are debris flow and each group of landslides was constructed in the inventory map separately. Then, the landslide inventory was randomly selected through Excel; 50% landslide was used for model analysis and the remaining 50% was used for validation purpose. Total 12 contributing factors, such as slope, aspect, curvature, topographic wetness index (TWI), elevation, forest type, forest timber diameter, forest crown density, geology, landuse, soil depth, and soil drainage were used in the analysis. Moreover, to find out the co-relation between landslide causative factors and incidents landslide, pixels were divided into several classes and frequency ratio for individual class was extracted. Eventually, six landslide susceptibility maps were constructed using the Bayesian Predictive Discriminant (BPD), Empirical Likelihood Ratio (ELR), and Linear Regression Method (LRM) models based on different category dada. Finally, in the cross validation process, landslide susceptibility map was plotted with a receiver operating characteristic (ROC) curve and calculated the area under the curve (AUC) and tried to extract success rate curve. The result showed that Bayesian, likelihood and linear models were of 85.52%, 85.23%, and 83.49% accuracy respectively for total data. Subsequently, in the category of debris flow landslide, results are little better compare with total data and its contained 86.33%, 85.53% and 84.17% accuracy. It means all three models were reasonable methods for landslide susceptibility analysis. The models have proved to produce reliable predictions for regional spatial planning or land-use planning.

• 한국작물학회:학술대회논문집
• /
• 한국작물학회 2022년도 추계학술대회
• /
• pp.201-201
• /
• 2022

Rice bakanae disease is a serious global threat in major rice-cultivating regions worldwide causing high yield loss. It is caused by the fungal pathogen Fusarium fujikuroi. Varying degree of resistance or susceptibility to bakanae disease had been reported among Korean japonica rice varieties. We developed a modified in vitro bakanae disease bioassay method and tested 31 Korean japonica rice varieties. Nampyeong and Samgwang varieties showed highest resistance while 14 varieties including Junam and Hopum were highly susceptible with 100% mortality rate. We carried out mapping QTLs for bakanae disease resistance with four F2:F3 populations derived from the crosses between Korean japonica rice varieties. The Kompetitive Allele-Specific PCR (KASP) markers developed in our laboratory based on the SNPs detected in Korean japonica rice varieties were used in genotyping F2 plants in the populations. We found four major QTLs on chromosome 1, 4, 6, and 9 with LOD scores of 21.4, 6.9, 6.0, and 60.3, respectively. In addition, we are doing map-based cloning of the QTLs on chromosome 1 and 9 which were found with Junam/Nampyeong F2:F3 population and Junam/Samgwang F2:F3 population, respectively. These QTLs will be very useful in developing bakanae disease resistant high quality rice varieties.

• Journal of Electrochemical Science and Technology
• /
• 제14권4호
• /
• pp.377-387
• /
• 2023

Ni-W/Al₂O₃ nano-composites were electrodeposited on mild steel substrate for mechanical and corrosion resistance applications. This study focused on the preparation of Ni-W/Al₂O₃ nano-composite coating with various quantity of Al₂O₃ incorporations. The addition of Al₂O₃ in the electrolytes were varied from 1-10 g/L in electrolytes and the Al₂O₃ incorporation in Ni-W/Al₂O₃ nano-composite coatings were obtained from 1.82 to 13.86 wt.%. The incorporation of Al₂O₃ in Ni-W alloy matrix influenced the grain size, surface morphology and structural properties were observed. The distributions of Al₂O₃ particle in alloy matrix were confirmed using electron microscopy (FESEM and TEM) and EDAX mapping analysis. The crystal structure informations were studied using X-ray diffraction method and it confirms that the deposits having cubic crystal structure. The better corrosion rate (0.87 mpy) and microhardness (965 HV) properties were obtained for the Ni-W/Al₂O₃ nano-composite coating with 13.86 wt.% of Al₂O₃ incorporations.

• 사물인터넷융복합논문지
• /
• 제9권6호
• /
• pp.119-124
• /
• 2023

최근 딥러닝은 다양한 분야에서 전통적인 기계학습에 비해 월등히 높은 성능을 보이고 있으며, 패턴인식을 위한 보편적인 방법으로 자리 잡아 가고 있다. 하지만, 이에 비해 정형데이터를 사용하는 분류 문제에서는 여전히 머신러닝 기법이 주류를 이루고 있다. 본 논문에서는 정형데이터를 고차원 텐서로 변환하는 네트워크 모듈을 제안하며, 이 모듈을 보편적인 딥러닝 네트워크와 함께 구성하여 정형데이터의 분류 문제에 적용하였다. 제안된 방법은 4종의 데이터셋을 활용하여 학습 및 검증되었으며, 제안된 방법은 90.22%의 평균 정확도를 달성하여, 최신 딥러닝 모델인 TabNet에 비해 2.55%p 높은 정확도를 보였다. 제안된 방법은 컴퓨터 비전 분야에서 높은 성능을 보이는 다양한 네트워크 구조를 정형데이터에 활용할 수 있다는 점에서 의미가 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권10호
• /
• pp.2768-2787
• /
• 2023

As an important research direction of the application of computer science in the medical field, the automatic generation technology of radiology report has attracted wide attention in the academic community. Because the proportion of normal regions in radiology images is much larger than that of abnormal regions, words describing diseases are often masked by other words, resulting in significant feature loss during the calculation process, which affects the quality of generated reports. In addition, the huge difference between visual features and semantic features causes traditional multi-modal fusion method to fail to generate long narrative structures consisting of multiple sentences, which are required for medical reports. To address these challenges, we propose a multi-feature optimization Transformer (MFOT) for generating radiology reports. In detail, a multi-dimensional mapping attention (MDMA) module is designed to encode the visual grid features from different dimensions to reduce the loss of primary features in the encoding process; a feature pre-fusion (FP) module is constructed to enhance the interaction ability between multi-modal features, so as to generate a reasonably structured radiology report; a detail enhanced attention (DEA) module is proposed to enhance the extraction and utilization of key features and reduce the loss of key features. In conclusion, we evaluate the performance of our proposed model against prevailing mainstream models by utilizing widely-recognized radiology report datasets, namely IU X-Ray and MIMIC-CXR. The experimental outcomes demonstrate that our model achieves SOTA performance on both datasets, compared with the base model, the average improvement of six key indicators is 19.9% and 18.0% respectively. These findings substantiate the efficacy of our model in the domain of automated radiology report generation.

• 대한원격탐사학회지
• /
• 제39권5_2호
• /
• pp.837-847
• /
• 2023

본 연구는 사고현장에서 화학물질로 인해 발생한 사고지점의 피해면적을 식생지수로 산출하는 방안을 연구하였다. 자료수집은 두 종류의 드론을 활용하였으며, 사진측량기법을 적용한 3차원 점군자료에서 피해면적을 산출하였다. 다중분광센서의 분광대역 정보를 활용하여 제작한 정사영상을 토대로 식생지수 영상을 제작하였고, 임계값에 따른 피해면적의 결과로 사고현장에 대한 통계를 분석하였다. 근적외선 밴드 기반의 식생지수 Kappa 값은 0.79, 녹색 밴드 기반의 식생지수는 0.76으로 화학물질사고 조사현장에서 식생지수를 활용한 피해면적 분석 방법을 활용할 수 있을 것으로 판단된다.

• Tribology and Lubricants
• /
• 제39권6호
• /
• pp.256-261
• /
• 2023

This study is conducted to evaluate the durability of superhydrophobic surfaces, with a focus on two aspects: contact angle measurement and self-cleaning-performance analysis. Superhydrophobic copper and aluminum surfaces are fabricated using the immersion method and subjected to a rolling wear test, in which a 2 kg weight is placed on a rolling tester, under loaded conditions. To evaluate their durability, the contact angles of the specimens are measured for each cycle. In addition, the surface deformation of the specimens before and after the test is analyzed through SEM imaging and EDS mapping. The degradation of the self-cleaning performance is evaluated before and after the wear test. The results show that superhydrophobic aluminum is approximately 4.5 times more durable than superhydrophobic copper; the copper and aluminum specimens could endure 21,000 and 4,300 cycles of wear, respectively. The results of the self-cleaning test demonstrate that superhydrophobic aluminum is superior to superhydrophobic copper. After the wear test, the self-cleaning rates of the copper and aluminum specimens decrease to 72.7% and 83.4%, respectively. The relatively minor decrease in the self-cleaning rate of the aluminum specimen, despite the large number of wear cycles, confirms that the superhydrophobic aluminum specimen is more durable than its copper counterpart. This study is expected to aid in evaluating the durability of superhydrophobic surfaces in the future owing to the advantage of performing wear tests on superhydrophobic surfaces without damaging the surface coating.

• Safety and Health at Work
• /
• 제15권1호
• /
• pp.24-32
• /
• 2024

Background: Learning from incidents for accident prevention is a two-stage process, involving the investigation of past accidents to identify the causal factors, followed by the identification and implementation of remedial measures to address the identified causal factors. The focus of past research has been on the identification of causal factors, with limited focus on the identification and implementation of remedial measures. This research begins to contribute to this gap. The motivation for the research is twofold. First, previous analyses show the recurring nature of accidents within the Ghanaian mining industry, and the causal factors also remain the same. This raises questions on the nature and effectiveness of remedial measures identified to address the causes of past accidents. Secondly, without identifying and implementing remedial measures, the full benefits of accident investigations will not be achieved. Hence, this study aims to assess the nature of remedial measures proposed to address investigation causal factors. Method: The study adopted SMARTER from business studies with the addition of HMW (H - Hierarchical, M - Mapping, and W - Weighting of causal factors) to analyse the recommendations from 500 individual investigation reports across seven different mines in Ghana. Results: The individual and the work environment (79%) were mostly the focused during the search for causes, with limited focus on organisational factors (21%). Forty eight percentage of the recommendations were administrative, focussing on fixing the problem in the immediate affected area or department of the victim(s). Most recommendations (70.4%) were support activities that only enhance the effectiveness of control but do not prevent/mitigate the failure directly. Across all the mines, there was no focus on evaluating the performance of remedial measures after their implementation. Conclusion: Identifying sharp-end causes leads to proposing weak recommendations which fail to address latent organisational conditions. The study proposed a guide for effective planning and implementation of remedial actions.

• 자원환경지질
• /
• 제46권4호
• /
• pp.301-312
• /
• 2013

산사태는 지형, 지질, 임상, 토양 등과 같은 다양한 요인들이 복합적으로 작용하여 발생한다. 따라서 산사태 발생위치와 산사태 유발 요인 사이의 상관관계를 파악할 수 있는 다양한 분석 기법이 사용되고 있으며 본 연구에서는 산사태 위험지역을 정량적으로 예측할 수 있는 효과적인 기법을 제안하고자 퍼지관계 기법과 인공신경망 기법을 이용하여 포항지역의 산사태 취약성을 분석하였다. 취약성 분석을 위해 먼저 산사태 위치를 파악하여 현황도를 작성하였으며, 산사태 발생과 관련 있는 11개의 요인들에 대한 공간 데이터베이스를 구축하였다. 퍼지관계 기법에서는 cosine amplitude method를 이용해 각 요인 별 퍼지 소속 함수 값을 획득하고 퍼지관계 함수 연산을 이용하여 취약성도를 작성하였다. 인공신경망 기법에서는 오류 역전파 알고리즘을 이용하여 산사태와 관련 요인들 간의 상대적 가중치를 결정하고 취약성도를 작성하였다. 두 기법으로 도출된 산사태 취약성도의 ROC(Receiver Operating Characteristic)와 AUC(Area Under the Curve)를 통한 검증 결과는 82.18%와 87.4%로 나타났다. 퍼지 관계 및 인공신경망 기법 모두 높은 예측 정확도를 보여 취약성 분석 기법으로서의 적용 가능성이 있는 것으로 분석되었다. 한편 본 연구지역의 경우 인공신경망 기법이 퍼지관계 기법에 비해 좀 더 나은 예측 정확도를 보이는 것으로 분석되었다.

• 한국지리정보학회지
• /
• 제8권3호
• /
• pp.96-106
• /
• 2005

원격탐사와 GIS 기법는 시공간 측면에서 매우 귀중한 정보를 제공할 수 있으며, 홍수와 같은 재해 발생시 홍수발생 지역에 대한 매핑, 모니터링 빚 재해지역 관리 등에 있어 매우 유용한 정보를 제공할 수 있다. 지난 2000년 메콩강 유역의 Tonle Sap호에서 발생한 홍수에 의해 많은 피해가 발생하였으며, 특히 7월과 10월 사이에 두 차례의 홍수 피크가 기록되었다. 본 연구에서는 홍수피해에 대한 정량적인 분석을 위해 ISODATA와 세크멘테이션 기법을 이용하여 Landsat ETM+와 RADARSAT 영상을 분석하였다. 그러나, 영상으로부터 분석된 범람면적이 구름과 복잡한 지표피복물 등으로 인해 실제 홍수피해 상황을 정확히 반영하지 못했다. 따라서, 이러한 문제점을 해결하고자 GIS 기능 중 비용-거리 (cost-distance) 기법을 이용하여 홍수범람 면적을 분석하였으며, 분석결과는 수치표고자료(DEM)와 중첩하여 홍수량을 계산하였다. 계산된 홍수량은 수리모형인 MIKE 11의 모델링 결과와 비교하였다. 계산결과 영상 내에 많은 구름이 존재하는 Landsat ETM+ 영상의 경우와 복잡한 지표피복이나 시스템 변수 등의 영향으로 홍수피해 지역을 정확히 분류하기 어려운 RADARSAT 영상에서도 좋은 결과를 얻을 수 있었다.