• Title/Summary/Keyword: 안개 예측

Search Result 44, Processing Time 0.03 seconds

No-Reference Visibility Prediction Model of Foggy Images Using Perceptual Fog-Aware Statistical Features (시지각적 통계 특성을 활용한 안개 영상의 가시성 예측 모델)

  • Choi, Lark Kwon;You, Jaehee;Bovik, Alan C.
    • Journal of the Institute of Electronics and Information Engineers
    • /
    • v.51 no.4
    • /
    • pp.131-143
    • /
    • 2014
  • We propose a no-reference perceptual fog density and visibility prediction model in a single foggy scene based on natural scene statistics (NSS) and perceptual "fog aware" statistical features. Unlike previous studies, the proposed model predicts fog density without multiple foggy images, without salient objects in a scene including lane markings or traffic signs, without supplementary geographical information using an onboard camera, and without training on human-rated judgments. The proposed fog density and visibility predictor makes use of only measurable deviations from statistical regularities observed in natural foggy and fog-free images. Perceptual "fog aware" statistical features are derived from a corpus of natural foggy and fog-free images by using a spatial NSS model and observed fog characteristics including low contrast, faint color, and shifted luminance. The proposed model not only predicts perceptual fog density for the entire image but also provides local fog density for each patch size. To evaluate the performance of the proposed model against human judgments regarding fog visibility, we executed a human subjective study using a variety of 100 foggy images. Results show that the predicted fog density of the model correlates well with human judgments. The proposed model is a new fog density assessment work based on human visual perceptions. We hope that the proposed model will provide fertile ground for future research not only to enhance the visibility of foggy scenes but also to accurately evaluate the performance of defog algorithms.

Single Color Image Based on Fog Degree Measurement (Single Color Image의 안개 정도 측정 방법)

  • Lee, Geun-Min;Kim, won-ha
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • 2017.06a
    • /
    • pp.260-263
    • /
    • 2017
  • 본 논문은 single image에서 측정한 빛 전달량 값과 local contrast 값을 사용하여 안개 량을 수치화하는 방법을 제안한다. 제안하는 방법은 빛 전달량 값을 사용하여 안개로 예측되는 지역을 추정하고, 추정된 안개 예측지역의 넓이와 해당 지역의 local contrast 크기의 범위를 사용하여 안개 정도를 수치화 한다. single image에서 측정 가능한 안개 의 물리적 특성들을 고려하였기 때문에 기존의 안개 검출 알고리즘들이 구분하지 못했던 영상들에서도 안개 량을 정확하게 측정하였다. 실제 빛의 산란 정도를 측정하는 감광 계수 측정계를 사용하여 측정한 안개 량과 제안하는 방법의 수치를 비교했을 때, 다양한 환경과 물체를 포함한 영상들에서 95%이상의 정확도로 안개 정도를 수치화 하였다. 또한 빛 전달량 추정 과정에서 local contrast 값을 추출하여 사용하기 때문에 기존의 빛 전달량을 측정하는 방법에서 복잡도를 거의 증가시키지 않는다.

  • PDF

Developing a regional fog prediction model using tree-based machine-learning techniques and automated visibility observations (시정계 자료와 기계학습 기법을 이용한 지역 안개예측 모형 개발)

  • Kim, Daeha
    • Journal of Korea Water Resources Association
    • /
    • v.54 no.12
    • /
    • pp.1255-1263
    • /
    • 2021
  • While it could become an alternative water resource, fog could undermine traffic safety and operational performance of infrastructures. To reduce such adverse impacts, it is necessary to have spatially continuous fog risk information. In this work, tree-based machine-learning models were developed in order to quantify fog risks with routine meteorological observations alone. The Extreme Gradient Boosting (XGB), Light Gradient Boosting (LGB), and Random Forests (RF) were chosen for the regional fog models using operational weather and visibility observations within the Jeollabuk-do province. Results showed that RF seemed to show the most robust performance to categorize between fog and non-fog situations during the training and evaluation period of 2017-2019. While the LGB performed better than in predicting fog occurrences than the others, its false alarm ratio was the highest (0.695) among the three models. The predictability of the three models considerably declined when applying them for an independent period of 2020, potentially due to the distinctively enhanced air quality in the year under the global lockdown. Nonetheless, even in 2020, the three models were all able to produce fog risk information consistent with the spatial variation of observed fog occurrences. This work suggests that the tree-based machine learning models could be used as tools to find locations with relatively high fog risks.

Study on the Prediction of Surface Color Change of Cultural Properties Materials by Fog Occurrence (안개 발생에 따른 문화재 표면의 색 변화 예측 연구)

  • Han, Ye Bin;Park, Sang Hyeon;Yu, Ji A;Chung, Yong Jae
    • Journal of Conservation Science
    • /
    • v.32 no.4
    • /
    • pp.491-500
    • /
    • 2016
  • Fog is atmospheric in which tiny drops of water vapor are suspended in the air near the ground. Its form, occurrence, etc., change according to the temperature, relative humidity, wind and geographical features of the space around it. In particular, fog tends to occur near a source of water because of temperature and relative humidity difference. These days, climate change is increasingly affecting the occurrence of fog. Therefore the purpose of this study was to investigate how fog affects materials that are part of our cultural properties through outdoor exposure tests and artificial degradation. The degradation evaluation of materials as a function of fog occurrence frequency, showed that the color of metals changed noticeably, whereas dyed silk and Dancheong showed degradation on the surface and color differences but no particular tendencies. Therefore, damage prediction by color differences as a function of fog occurrence frequency was based on metal samples, which showed constant color differences. Through a comparison of the predictive value and color difference by outdoor exposure, the accuracy and applicability of the damage prediction formula was confirmed. If a more complex damage prediction formula is created, it is expected that prediction of the degree of material damage in the field would be possible.

Development of a Mid-/Long-term Prediction Algorithm for Traffic Speed Under Foggy Weather Conditions (안개시 도시고속도로 통행속도 중장기 예측 알고리즘 개발)

  • JEONG, Eunbi;OH, Cheol;KIM, Youngho
    • Journal of Korean Society of Transportation
    • /
    • v.33 no.3
    • /
    • pp.256-267
    • /
    • 2015
  • The intelligent transportation systems allow us to have valuable opportunities for collecting wide-area coverage traffic data. The significant efforts have been made in many countries to provide the reliable traffic conditions information such as travel time. This study analyzes the impacts of the fog weather conditions on the traffic stream. Also, a strategy for predicting the long-term traffic speeds is developed under foggy weather conditions. The results show that the average of speed reductions are 2.92kph and 5.36kph under the slight and heavy fog respectively. The best prediction performance is achieved when the previous 45 pattern cases data is used, and the 14.11% of mean absolute percentage error(MAPE) is obtained. The outcomes of this study support the development of more reliable traffic information for providing advanced traffic information service.

Comparison of Machine Learning Model Performance based on Observation Methods using Naked-eye and Visibility-meter (머신러닝을 이용한 안개 예측 시 목측과 시정계 계측 방법에 따른 모델 성능 차이 비교)

  • Changhyoun Park;Soon-hwan Lee
    • Journal of the Korean earth science society
    • /
    • v.44 no.2
    • /
    • pp.105-118
    • /
    • 2023
  • In this study, we predicted the presence of fog with a one-hour delay using the XGBoost DART machine learning algorithm for Andong, which had the highest occurrence of fog among inland stations from 2016 to 2020. We used six datasets: meteorological data, agricultural observation data, additional derived data, and their expanded data. The weather phenomenon numbers obtained through naked-eye observations and the visibility distances measured by visibility meters were classified as fog [1] or no-fog [0]. We set up twelve machine learning modeling experiments and used data from 2021 for model validation. We mainly evaluated model performance using recall and AUC-ROC, considering the harmful effects of fog on society and local communities. The combination of oversampled meteorological data features and the target induced by weather phenomenon numbers showed the best performance. This result highlights the importance of naked-eye observations in predicting fog using machine learning algorithms.

Development of SVR model for Visibility Forecasting by using Feature Selection based on Genetic Algorithm (유전 알고리즘 기반의 특징선택을 이용한 SVR 모델의 시정 예측 모델 개발)

  • Lim, Sung-Joon;Ahn, Kwang-Deuk;Ha, Jong-Chul;Lim, Eun-Ha;Lee, Yong Hee;Oh, Sung-Kwun
    • Proceedings of the KIEE Conference
    • /
    • 2015.07a
    • /
    • pp.1353-1354
    • /
    • 2015
  • 본 연구에서는 관측자료 기반의 안개 예보를 수행하기 위해 특징선택을 이용한 SVR 회귀분석 기반 시정 예측 가이던스를 개발하였다. 예측에 필요인자를 사전에 선택하는 유전알고리즘 기반의 최적화 방법을 적용하여, 관측된 여러 기상인자의 입력인자 중 실제 시정을 예측하기 위한 입력인자를 선택하여 준다. 지점별 안개발생에 필요한 입력인자 및 예측 모델을 구성하여 통합적인 예측 모델이 아닌 각 지점에 최적화된 정보를 제공할 수 있도록 예측을 수행한다. 자료의 수집 특성상 3시간 간격으로 3시간 예보를 위한 시정을 예측하고, 예측 모델의 검증을 위해 현업의 수치모델 기반의 시정예측 정보와의 비교를 통해 실제 안개 시점에 대해 비교 분석하였고 그 결과를 통해 긍정적인 효과를 보였다. 예측모델을 적용하여 지도에 예측시정 정보를 제공하는 표출 시스템을 통해 실시간 가이던스를 제공하고자 연구를 수행하였다.

  • PDF

Local contrast and Transmission Based Fog Degree Measurement in Single Image (Local Contrast와 빛 전달량 기반 Single Image의 안개 정도 측정 방법)

  • Lee, Geun-min;Kim, Wonha
    • Journal of Broadcast Engineering
    • /
    • v.22 no.3
    • /
    • pp.375-380
    • /
    • 2017
  • This paper has proposed a single image based fog degree quantification method by measuring both transmission and local contrast. The proposed method estimates the foggy expected regions from transmission, and then assesses the size of regions of which transmission values are foggy expected ones and the range of local contrast value on such regions. Compared with fog degree gauged by the scattering coefficient measurement sensor, the proposed method quantifies the fog degree with more than 95% accuracy for images containing various objects and environments. We also developed a technique that measures the local contrast values in process of measuring transmission values. So, the proposed method does not increase complexity compared to the existing transmission method.

Fog Forecasting by Using Numerical Weather Prediction Model (수치모델을 이용한 안개 예측 사례 연구)

  • 김영아;오희진;서태건
    • Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
    • /
    • 2002.11a
    • /
    • pp.85-88
    • /
    • 2002
  • 기상학적으로 안개는 지상에서 발생하는 응결 현상으로, 시정이 1km 이하일 때로 정의된다. 안개 발생은 기후 인자의 영향을 많이 받는다. 따라서 각 지역마다의 발생 특성을 따로 통계해야 할 필요가 있다. 특히 항공 교통의 장애가 되는 위험 요소로서의 역할이 중시되어 각 비행장마다 발생 특성이 따로 통계 분석되고 이용되어 왔다.(중략)

  • PDF

A Study on Prediction System of Sea Fogs in the East Sea (동해의 해무 예측 시스템 연구)

  • 서장원;오희진;안중배;윤용훈
    • The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
    • /
    • v.8 no.2
    • /
    • pp.121-131
    • /
    • 2003
  • We have found that the east coast of Korea has had few sea fogs on January, February, November and December for the past 20 years by the analysis of monthly fog frequency and duration time. These phenomena appear to relate to the topographical characteristics of which the Taebaek Mountains descends toward the east to bar the radiation fog. On the other hand, the cause of occurring the spring and summer fog which has 90% of the whole frequency is divided into three cases. The first is the steam fog caused by the advection of the northeast cold air current on the East Sea due to the extension of Okhotsk High. The second is the advection fog caused by cooling and saturation of warm airmass advected on cold sea surface. And the last is the frontal fog caused by the supply of enough vapor due to the movement of low-pressure system and the advection of cold air behind a cold front. While, we simulate the sea fog for the period of the case studies by implementing fog prediction system(DUT-METRI) that makes it possible to forecast the fog in the vertical section of neighborhood of the East Sea and to predict the sea surface wind, relative humidity, ceiling height, visibility etc. Finally we verified this result by satellite image.