• 한국시뮬레이션학회논문지
• /
• 제30권2호
• /
• pp.41-48
• /
• 2021

농촌 인구의 감소와 고령화가 지속되면서 농업 생상성 향상의 중요성이 높아지고 있는 가운데, 농작물 품질에 대한 조기 예측은 농업 생산성 및 수익성 향상에 중요한 역할을 할 수 있다. 최근 CNN 기반의 딥러닝 기술 및 전이 학습을 활용하여 농작물의 질병을 분류하거나 수확량을 예측하는 연구가 활발하게 진행되고 있지만, 수확 후 농작물의 품질을 식재단계에서 조기에 예측하는 연구는 찾아보기 힘들다. 본 연구에서는 건강 기능성 식품으로 주목받고 있는 새싹삼을 대상으로, 수확 후 새싹삼의 품질을 식재단계에서 조기에 예측하는 모델을 제안한다. 이를 위하여 묘삼의 이미지를 촬영한 후 수경재배를 통해 새싹삼을 재배하였고, 수확 후 새싹삼의 품질을 분류하여 실험 데이터를 수집하였다. 다수의 CNN 기반의 사전 학습된 모델을 활용하여 새싹삼 조기 품질 예측 모델을 구축하고, 수집된 데이터를 이용하여 각 모델의 학습 및 예측 성능을 비교 분석하였다. 분석 결과 모든 예측 모델에서 80% 이상의 예측 정확도를 보였으며, 특히 ResNet152V2 기반의 예측 모델에서 가장 높은 정확도를 보였다. 본 연구를 통해 인력에 의존하던 기존의 묘삼 선별 작업을 자동화하여 새싹삼의 품질을 높이고 생산량을 증대시켜 농가의 수익창출에 기여할 수 있을 것으로 기대된다.

• 국제학술발표논문집
• /
• The 5th International Conference on Construction Engineering and Project Management
• /
• pp.596-601
• /
• 2013

In the current pavement construction practice, the state agencies traditionally determine the quality of the as-constructed pavement mix based on individual mixture material parameters (e.g., air voids, cement or asphalt content, aggregate gradation, etc.) and consider these parameters as key variables to influence payment schedule to the contractors and the present and future quality of the as-constructed mixture. A set of empirically pre-determined pay adjustment schedule for each parameter that was differently developed and being used by the individual agencies is then applied to a given project, in order to judge whether each parameter conforms to the designated specifications and consequently the contractor may either be rewarded or penalized in accordance with the payment schedule. With an improved quality assurance system, the Performance Related Specification, the individual parameters are not utilized as a direct judgment factor; rather, they become independent variables within a performance prediction function which is directly used to predict the performance. The quantified performance based on the prediction model is then applied to evaluate the pavement quality. This paper presents the brief history of the quality assurance in asphalt pavement construction including the Performance Related Specifications, statistical performance models in terms of fatigue and rutting distresses, as an example of the performance prediction models, and envisions the possibilities as to how this Performance Related Specification could be utilized in other infrastructures construction quality assurance.

• 한국식품저장유통학회지
• /
• 제4권2호
• /
• pp.123-130
• /
• 1997

To develop the prediction program for quality change of Citrus unshiu during marketing, we examined the quality characteristics of Citrus unshiu stored at experimental refrigerator set to 4, 8, 12 and 16$^{\circ}C$ for 2 months. According to the storage temperature the changes of quality characteristics were different respectively, but it was most severe during 16$^{\circ}C$ storage. Activation energy and Q10 value were 6683.16 cal/mol K and 1.53 respectively. The determination coefficient of regression equation of pH, acidity and vitamin C by surface response analysis were over 0.85. Using these regression equation, we developed the prediction program for the change of pH, acidity and vitamin C contents. The calculated values and experimental values of pH, acidity and vitamin C contents for short-term storage of Citrus unshiu were coincided well.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2011년도 춘계학술대회 논문집
• /
• pp.707-708
• /
• 2011

• 한국환경과학회지
• /
• 제18권6호
• /
• pp.683-690
• /
• 2009

The water quality of reservoir can be controled by water quality prediction model because it can not only grasping the present water state but also predicting the water quality in future. In this study, WASP model is used to predict the water quality of Chungju reguration reservoir. This model has some special option which predicts the pollutant outflow phenomenon caused by the contamination sources. So this model is widely used because that can present the scientific basis in this field. This model can help the managers make the right choice of water quality policy. Environmental grade of Chungju reguration reservoir is in III,IV grade which is in bad condition comparatively. The water contamination will be in poor as the year passes. When considering T-N, T-P which are the nutrient to control eutrophication, the concentrated administration about contamination sources is in urgent.

• 한국물환경학회지
• /
• 제30권4호
• /
• pp.409-417
• /
• 2014

Spatial prediction methods have been useful to determine the variability of water quality in space and time due to difficulties in collecting spatial data across extensive spaces such as watershed. This study compares two kriging methods in predicting BOD concentration on the unmonitored sites in the Geum River Watershed and to assess its predictive performance by leave-one-out cross validation. This study has shown that cokriging method can make better predictions of BOD concentration than ordinary kriging method across the Geum River Watershed. Challenges for the application of cokriging on the spatial prediction of surface water quality involve the comparison of network-distance-based relationship and euclidean-distance-based relationship for the improvement in the predictive performance.

• Journal of Animal Science and Technology
• /
• 제47권6호
• /
• pp.1095-1100
• /
• 2005

본 시험은 초음파진단기를 이용한 한우의 도체육질 예측기법의 확립과 예측율 향상을 목적으로 실시하였다. 도체육질 등급별 초음파 표준화상과 초음파 육질 index를 활용한 decision tree 분석을 이용하여 24개월령 거세한우 66두의 도체육질 등급을 예측하였으며, 그 결과를 요약하면 다음과 같다. 초음파 육질 index를 이용한 의사결정나무 분석 결과, 초음파 화상 내 늑골의 선명도가 육질 등급에 가장 큰 영향을 미치는 것으로 나타났다. 거세한우 66두를 대상으로 초음파 육질 index에 의한 도체육질 등급의 예측율은 86.4%를 나타났으며, 도체 등급별 초음파 표준화상을 이용한 78.8%에 비해 7.6%의 예측율 향상을 보였다.

• 산업경영시스템학회지
• /
• 제42권4호
• /
• pp.203-210
• /
• 2019

Recently, a number of researchers have produced research and reports in order to forecast more exactly air quality such as particulate matter and odor. However, such research mainly focuses on the atmospheric diffusion models that have been used for the air quality prediction in environmental engineering area. Even though it has various merits, it has some limitation in that it uses very limited spatial attributes such as geographical attributes. Thus, we propose the new approach to forecast an air quality using a deep learning based ensemble model combining temporal and spatial predictor. The temporal predictor employs the RNN LSTM and the spatial predictor is based on the geographically weighted regression model. The ensemble model also uses the RNN LSTM that combines two models with stacking structure. The ensemble model is capable of inferring the air quality of the areas without air quality monitoring station, and even forecasting future air quality. We installed the IoT sensors measuring PM2.5, PM10, H2S, NH3, VOC at the 8 stations in Jeonju in order to gather air quality data. The numerical results showed that our new model has very exact prediction capability with comparison to the real measured data. It implies that the spatial attributes should be considered to more exact air quality prediction.

• 한국농공학회논문집
• /
• 제49권4호
• /
• pp.13-22
• /
• 2007

The uncertainty in water quality model predictions is inevitably high due to natural stochasticity, model uncertainty, and parameter uncertainty. An integrated modeling system under uncertainty was described and demonstrated for use in watershed management and receiving-water quality prediction. A watershed model (HSPF), a receiving water quality model (WASP), and a wetland model (NPS-WET) were incorporated into an integrated modeling system (modified-BASINS) and applied to the Hwaseong Reservoir watershed. Reservoir water quality was predicted using the calibrated integrated modeling system, and the deterministic integrated modeling output was useful for estimating mean water quality given future watershed conditions and assessing the spatial distribution of pollutant loads. A Monte Carlo simulation was used to investigate the effect of various uncertainties on output prediction. Without pollution control measures in the watershed, the concentrations of total nitrogen (T-N) and total phosphorous (T-P) in the Hwaseong Reservoir, considering uncertainty, would be less than about 4.8 and 0.26 mg 4.8 and 0.26 mg $L^{-1}$, respectively, with 95% confidence. The effects of two watershed management practices, a wastewater treatment plant (WWTP) and a constructed wetland (WETLAND), were evaluated. The combined scenario (WWTP + WETLAND) was the most effective at improving reservoir water quality, bringing concentrations of T-N and T-P in the Hwaseong Reservoir to less than 3.54 and 0.15 mg ${L^{-1}$, 26.7 and 42.9% improvements, respectively, with 95% confidence. Overall, the Monte Carlo simulation in the integrated modeling system was practical for estimating uncertainty and reliable in water quality prediction. The approach described here may allow decisions to be made based on probability and level of risk, and its application is recommended.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제16권11호
• /
• pp.1116-1120
• /
• 2010

본 논문에서 우리는 추천 시스템을 위한 두 개의 정제된 이웃선정 알고리즘을 제시하고, 또한 아이템의 속성정보가 어떻게 고품질의 예측을 위해 사용될 수 있는지를 보인다. 정제된 이웃선정 알고리즘은 가상 이웃과 대체 이웃을 각각 사용하여 이행적 유사도를 기반으로 한 이웃선정 방법을 적용한다. 실험 결과는 본 논문에서 제안한 알고리즘을 적용한 추천 시스템이 다른 시스템에 비해 보다 우수한 성능을 가짐을 보여준다. 이러한 제안 시스템은 예측 품질의 저하 없이 대규모 데이터셋 문제 및 초기 참여자 문제를 극복할 수 있게 한다.