• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2024년도 춘계학술발표대회
• /
• pp.794-797
• /
• 2024

Despite advancements in deep learning, existing semantic segmentation models exhibit suboptimal performance under adverse weather conditions, such as fog or rain, whereas they perform well in clear weather conditions. To address this issue, much of the research has focused on making image or feature-level representations weather-independent. However, disentangling the style and content of images remains a challenge. In this work, we propose a novel fine-tuning method, 'freeze-n-update.' We identify a subset of model parameters that are weather-independent and demonstrate that by freezing these parameters and fine-tuning others, segmentation performance can be significantly improved. Experiments on a test dataset confirm both the effectiveness and practicality of our approach.

• 대한조선학회논문집
• /
• 제31권1호
• /
• pp.22-31
• /
• 1994

해양 공간에 설치되어 유체의 흐름과 지형에 변화를 주며, 대상 수산 생물의 유집에 의한 생산 어장의 조성과 치어의 보호 육성을 도모하는 중효한 어장 시설인 인공어초의 기술 개발은 점차 다양화되어지고 있다. 부체부와 계류부로 구성되어 있는 부어초의 개발에 있어서는 부어초의 특성 및 안전상 계류시스템의 설계가 극히 중요한 부분이 된다. 이 논문에서는 부어초의 계류시스템에 많이 사용되어지는 나일론, 폴리에스터와 폴리에틸렌으로 이루어진 인조섬유로우프의 정적시험의 과정과 그 결과로서의 Creep거동 및 하중-인장 곡선을 얻었다. 특히 나일론 로우프의 경우 수중에서 20%내외의 강도 하락(Strength drop)이 관찰되며, Creep 시험 곡선은 대수시간 축(longerithmic time axis)상에서 직선을 유지하고, Flessner's formula와 평행한 결과를 얻었다.

• 한국농공학회지
• /
• 제14권2호
• /
• pp.2593-2602
• /
• 1972

Plastic pipes Wrapped with synthetic filter are recently used for drainage or Collecting of Underground water. But it's use is possible only for small size of diameter less than 300mm, because large size plastic pipes are not readily availabe. For large diameter infiltration gallery, porous concrete pipes are now used, but it's heavy weight brings difficulties in making, moving and setting of the pipes. With it's conventional method of filter setting, fine sands are brought into the pipes to make trouble to lifting pumps and channels Therefore, initial construction cost and maintenance cost become high. To solve-this problem, new method is developed and tested. Small PVC pipes(diameter 14mm) are assembled at the site of construction to newly devised I beam type circls. The size of circular inpiltration gallery is optionally determined by I beam type circle which support small PVC pipes and is made of PVC amterial. This gallery are wrappd with syntheitc filter to prevent sand instruction. In this test, the diameter of 300, 400, 500mm were used. I beam type circles were made with PVC plated with thickness. t=6, 9, 12mm. Water quantity collected through the PVC circulor gallery are measured and the strengths of the gallery. 1. Allowable setting depth of gallery pipe below graund for the case of t=6mm, D=500mm is 2.72m. 2. Collected water quantity depends on soil texture, depth of water grandient of water surface, filter material angle of setting etc. 3. About 126% of water quantity collected from the one gallery pipe measured in two gallery pipes of two parallel installation.

• 한국측량학회지
• /
• 제9권1호
• /
• pp.97-107
• /
• 1991

본 연구에서는 리모트센싱과 GIS의 장점을 살려 하나의 시스템에서 여러 기능을 복합적으로 이용하기 위한 리모트센싱과 GIS의 통합을 시도한 것이다. 래스터와 벡터 데이타의 동시 통합 출력을 위한 중복 알고리즘을 개발하였으며, 리모트센싱과 GIS의 통합결과를 시험적용하기 위하여 인공위성 화상 데이타와 지형도 벡터 그래픽 데이타를 정확하게 통합시켰다. 실제 적용에서는 리모트센싱과 GIS의 주요 적용분야의 주제별 적용에 대한 데이타 모집과 중복 그리고 좌표계 변환을 통하여 대상지역에 다각적으로 적용할 수 있는 기법을 시도함으로써 벡터와 래스터의 중복의 효용성을 입증하고 복합적인 현안분석을 통해서 리모트센싱과 GIS의 복합적용을 위한 새로운 적용 기법을 제시하였다.

• 펄프종이기술
• /
• 제39권1호
• /
• pp.30-37
• /
• 2007

The seed paper was used in farm field recently for a sound young plant. The most of seed paper are made of synthetic non-woven sheet. Therefore, it is very difficult to bio-degrade in soil and is very hard to have some special function, for example keeping herbicide and/or insecticide activity because of its lack of chemical acceptability. The purpose of this research is manufacture of seedling paper which have a function of herbicide activity from waste paper. The fiber properties from waste paper were remarkably improved by fine removal with washing and/or flotation process. The paper-making ability for seed paper was enhanced with enzyme treatment of secondary fibers. The paper for seedling must have a good bio-degradation ability in soils. The absorption amount of chemical like as dithiopyr was increased remarkably in enzyme treated base paper. The embossing treatment of base paper was very effective for seed attachment and chemicals retention. And also, the developed seed paper showed a good penetration property of young root through embossed paper.

• Smart Structures and Systems
• /
• 제33권3호
• /
• pp.227-241
• /
• 2024

Safety and structural integrity of civil structures, like bridges and buildings, can be substantially enhanced by employing appropriate structural health monitoring (SHM) techniques for timely diagnosis of incipient damages. The information gathered from health monitoring of important infrastructure helps in making informed decisions on their maintenance. This ensures smooth, uninterrupted operation of the civil infrastructure and also cuts down the overall maintenance cost. With an early warning system, SHM can protect human life during major structural failures. A real-time online damage localization technique is proposed using only the vibration measurements in this paper. The concept of the 'Degree of Scatter' (DoS) of the vibration measurements is used to generate a spatial profile, and fractal dimension theory is used for damage detection and localization in the proposed two-phase algorithm. Further, it ensures robustness against environmental and operational variability (EoV). The proposed method works only with output-only responses and does not require correlated finite element models. Investigations are carried out to test the presented algorithm, using the synthetic data generated from a simply supported beam, a 25-storey shear building model, and also experimental data obtained from the lab-level experiments on a steel I-beam and a ten-storey framed structure. The investigations suggest that the proposed damage localization algorithm is capable of isolating the influence of the confounding factors associated with EoV while detecting and localizing damage even with noisy measurements.

• 경영정보학연구
• /
• 제18권2호
• /
• pp.173-198
• /
• 2016

본 연구의 목적은 비즈니스 인텔리전스 연구 관점에서 기업부실화 예측 성능을 향상키시는 것이다. 이를 위해 본 연구는 기존 연구들에서 미흡하게 다루어졌던 1) 데이터셋을 구성하는 과정에서 발생하는 바이어스 문제, 2) 거시경제위험 요소의 미반영 문제, 3) 데이터 불균형 문제, 4) 서술적 바이어스 문제를 다루어 경기순환국면을 반영한 기업부실화 예측 프레임워크를 제안하고, 이를 바탕으로 기업부실화 예측 모델을 개발하였다. 본 연구에서는 경기순환국면별로 각각의 데이터셋을 구성하고, 각 데이터셋에서 의사결정나무, 인공신경망 등 단일 분류기부터 앙상블 기법까지 다양한 데이터마이닝 알고리즘을 적용하여 실험하였다. 또한 본 연구는 데이터불균형 문제를 해결하기 위해, 오버샘플링 기법인 SMOTE(synthetic minority over-sampling technique) 기법을 통해 초기 데이터 불균형 상태에서부터 표본비율을 1:1까지 변화시켜 가며, 기업부실화 예측 모델을 개발하는 실험을 하였고, 예측 모델의 변수 선정 시에 선행연구를 바탕으로 재무비율을 추출하고, 여기서 파생된 IT 산출물인 재무상태변동성과 산업수준상태변동성을 예측 모델에 삽입하였다. 마지막으로, 본 연구는 각 순환국면에서 만들어진 기업부실화 예측 모델의 예측 성능 비교와 경기 확장기와 수축기에서의 기업부실화 예측 모델의 유용성에 대해 논의하였다. 본 연구는 비즈니스 인텔리전스 연구 측면에서 기존 연구에서 미흡하게 다루어졌던 4가지 문제점을 검토하고, 이를 해결할 프레임워크를 제안함으로써 기존 연구 대비 기업부실화 예측률을 10% 이상 향상시켰다는 점에서 연구의 의의를 찾을 수 있다.

• 품질경영학회지
• /
• 제51권2호
• /
• pp.283-296
• /
• 2023

Purpose: The purpose of this study is to actually implement and verify whether welding defects can be detected in real time by utilizing deep learning AI solutions in the welding process of electric vehicle hairpin winding motors. Methods: AI's function and technological elements using synthetic neural network were applied to existing electric vehicle hairpin winding motor laser welding process by making special hardware for detecting electric vehicle hairpin motor laser welding defect. Results: As a result of the test applied to the welding process of the electric vehicle hairpin winding motor, it was confirmed that defects in the welding part were detected in real time. The accuracy of detection of welds was achieved at 0.99 based on mAP@95, and the accuracy of detection of defective parts was 1.18 based on FB-Score 1.5, which fell short of the target, so it will be supplemented by introducing additional lighting and camera settings and enhancement techniques in the future. Conclusion: This study is significant in that it improves the welding quality of hairpin winding motors of electric vehicles by applying domestic artificial intelligence solutions to laser welding operations of hairpin winding motors of electric vehicles. Defects of a manufacturing line can be corrected immediately through automatic welding inspection after laser welding of an electric vehicle hairpin winding motor, thus reducing waste throughput caused by welding failure in the final stage, reducing input costs and increasing product production.

• 자원환경지질
• /
• 제54권1호
• /
• pp.21-33
• /
• 2021

국내 오염시나리오별 안정화 효율과 경제성이 뛰어난 안정화제를 선택하여 적용할 수 있도록, 국내외에서 연구된 대표적인 안정화제를 대상으로 국내 중금속 오염 현장 부지 특성별 중금속 안정화 효율이 높은 안정화제 순위를 결정하였다. 총 5종류의 오염시나리오를 가정하여 각각 해당되는 국내 오염부지 토양을 확보하였다. 국내외 활용도와 안정화 효율 연구 결과, 오염특성별 부지 시나리오에 적용 가능성 등을 고려하여 기존에 연구되었던 안정화제 13가지를 선정하였다. 선정한 오염 토양과 안정화제의 오염 가능성과 현장 적용 가능성을 평가하기 위하여 XRD/XRF 분석, 독성용출시험과 인공강우용출시험 등을 실시하였다. 부지 오염시나리오를 대표하는 5종류 오염 토양에 대하여 선정된 13종의 안정화제에 의한 비소, 수은, 납, 6가 크롬, 아연, 니켈, 구리 등 총 8종의 중금속(반금속인 비소 포함) 용출 저감 효과를 규명하는 용출 배치실험을 수행하였다. 총 5개 오염 토양에 대하여 13개 안정화제 주입 비율 3%, 5%, 7% 적용 시, 각 중금속(비소 포함)에 대한 중금속 용출 저감 효율이 안정화제를 주입하지 않은 토양 대비 20% 이상을 나타내는 안정화제 중에서 저감 효율이 높은 순위부터 5개 안정화제(Top 5)를 선택하였다. 각 안정화제에 대하여 안정화제 주입비율, 중금속 종류, 부지별 조건에 따라 수행된 배치실험 결과에 대하여 Top 5에 해당하는 총 횟수를 합산하여, 다양한 국내 부지 오염시나리오에 적용할 수 있는 안정화제의 순위를 결정하였다. 5개 오염 토양에 대하여 8개 중금속 항목별 용출 저감 효율이 20% 이상인 경우, 가장 안정화 효율이 높은 순위는 광산배수처리 슬러지(mine drainage treatment sludge), 산화철, 생석회, 소석회-석회석, 황화철, 바이오차 순으로 나타났다. 위 안정화제들에 대하여 안정화제의 효율대비 단가를 산정한 결과, 광산배수처리 슬러지, 석회석, 제강슬래그(비소의 경우), 생석회, 소석회 순으로 경제성이 높게 나타나 현장 적용성이 뛰어난 것으로 밝혀졌다.

• 한국작물학회:학술대회논문집
• /
• 한국작물학회 2022년도 추계학술대회
• /
• pp.53-53
• /
• 2022

Modem agricultural production needs to provide sustainable management practices that are eco-friendly and low cost. Plant extracts are a cost-effective and environmentally friendly alternative to synthetic plant growth regulators. This study was therefore carried out to investigate the effects of various plant extracts produced using different extraction methods on the vegetative growth of rice under laboratory and greenhouse conditions. For this study, seventeen plant extracts were made from plant species such as leaves of M. arvense, C. asiatica, M. oleifera, V. radiata, V. unguiculate, P. guajava, A. vera, and A. tuberosum, aboveground plant parts of C. rotundus, M. sativa, and P. frutescens, roots of R. undulatum, tubers of A. sativum, leaves and stems of G. max (cv. Taegwang) as well as rice straw and hulls (cv. Hopyeong). As a test crop, we applied these extracts to rice plants. For the purpose of making our extracts, some plant materials and species were collected in fields and others were purchased from Chonnam Hanyaknonghyup Cooperation (South Korea). Leaves, roots, and aboveground plant parts of plant species were dried, ground, extracted (water, boiling water and ethanol) and fermented. Rice growth promotion effects were determined using plant extracts at 0, 0.05, 0.1, 0.5, and 1% concentrations under petri dish conditions. Seven selected plant extracts were applied to rice seeds with soil drench application or seedling at 3-4 leaf stages with soil and foliar applications under greenhouse conditions. For comparison with extracts, we used urea at 0.6%. Of the 17 water extracts used in this study, 10 extracts reduced rice growth, but the other 7 extracts (P. guajava, A. vera, A. tuberosum, M. sativa, A. sativum, and G. max) increased growth by 40-60% on compared to the control in Petri dish bioassay. Thus, these 7 extracts were selected for further study. Under greenhouse conditions, rice growth also increased by 20-40% when the same 7 extracts were applied to rice seeds using soil drench application. Furthermore, at the 3-4 leaf stage rice growth also increased 30-80% or 30-60% when the same 7 extracts were applied using soil and foliar applications. Overall, the 7 extracts produced higher rates of growth promotion when soil drench application was used than when foliar application was used. In the case of boiling water and ethanol extracts, rice growth increased only 20% in response to both soil drench and foliar application of the same 7 extracts. Rice growth promotion was greater when extracts were produced using water extraction method than boiling water and ethanol extraction methods. Most notably, the 7 water extracts used in this study produced higher rates of growth promotion than urea at 0.6% which is typically used for crop growth promotion. Overall, the 7 water extracts when applied using soil drenching method can be used as effective growth promotors of rice in organic agriculture.