• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.3
• /
• pp.673-676
• /
• 2017

In this paper, we propose a day and night fog detection algorithm that is not affected by lighting conditions. First, we present the definitions and the extraction methods of fog features in daytime and nighttime environments, respectively. We then propose the fog detection algorithm using a neural network from the fog features. A set of experiments has been conducted with images taken at various environments, and the average recall of the proposed algorithm is 97.5%.

• Journal of Korean Society for Atmospheric Environment
• /
• v.12 no.4
• /
• pp.399-406
• /
• 1996

Acidic fog is catastrophic to aviation and potentially affect materials, vegetation, crops and public health. This paper was carried out to investigate the chemical features of fog sample at Mt. Sobaek (mean sea level : 1, 340m) from June to August 1995. Each sample was analyzed for pH, electrical conductivity and major ions (anion : $Cl^N)_3^-, SO_4^{2-}, cation : Na^+, NH_4^+, K^+, Mg^{2+}, Ca^{2+}$) by ion chromatography. The quality analysis of fog sample data was performed based on ion balance and electrical conductivity method. The wind directions are subdivided into the northerly and southerly wind according to the wind direction data at the Sobaek-san meteorological observation station. Statistical analyses were performed on the complete set of results in order to obtain a description of fog sample. All the statistical treatment was carried out using the SPSS/PC + software package. The major ion concentration of fog samples was higher for the northwesterly wind cases than sourtheasterly wind cases. The pH of fog sample varied between 2.95 and 6.08. The average pH and electrical conductivity of total sample (n=210) were 4.39 and 113.0 $\mu$S/cm, respectively. It may be noted that in nearly all the cases, the dominant major ions in the fog sample at Mt. Sobaek were $SO_4^{2-}, NO_3^-, H^+ and NH_4^+$.

• 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.

• Journal of Korea Multimedia Society
• /
• v.24 no.10
• /
• pp.1319-1325
• /
• 2021

It is important to remove fog for accurate object recognition and detection during preprocessing because images taken in foggy adverse weather suffer from poor quality of images due to scattering and absorption of light, resulting in poor performance of various vision-based applications. This paper proposes an end-to-end deep learning-based single image de-fogging method using U-Net architecture. The loss function used in the algorithm is a loss function based on Mahalanobis distance with fog features, which solves the problem of domain shifts, and demonstrates superior performance by comparing qualitative and quantitative numerical evaluations with conventional methods. We also design it to generate fog through the VGG19 loss function and use it as the next training dataset.

• Journal of the Korean Society of Safety
• /
• v.26 no.4
• /
• pp.127-134
• /
• 2011

A traffic accident which happens in Expressway during dense fog is more likely to cause the sequential accidents and high death rate. So, the preventive measures shall be taken at dangerous areas to enhance the efficiency of roads and minimize the accidents and the resultant damages. So, it is necessary to find out the characteristics of freeway zone which has high risk of fog occurrence and to establish the comprehensive safety strategy on installation and operation of the safety equipment. In this study, I developed a fog forecasting model by using the freeway fog data. This model can be used as the fog forecasting model in dealing with fog problems when new road is planned. The model was developed by using a statistical analysis technique or the regression analysis, focusing on the variables such as geographical features and regional conditions, distances to water sources and the area of water source. I have segmented the models by classifying the area into inland area and coastal area. The distance to water source and area of the water source located around the freeway were found to be main factors causing fog.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.6
• /
• pp.819-824
• /
• 2021

As technologies related to image processing such as autonomous driving and CCTV develop, fog removal algorithms using a single image are being studied to improve the problem of image distortion. As a method of predicting fog density, there is a method of estimating the depth of an image by generating a depth map, and various fog features may be used as training data of the depth map. In addition, it is essential to implement a hardware capable of processing high-definition images in real time in order to apply the fog removal algorithm to actual technologies. In this paper, we implement NLCV (Normalize Local Coefficient of Variation), a feature of fog based on coefficient of variation, in hardware. The proposed hardware is an FPGA implementation of Xilinx's xczu7ev-2ffvc1156 as a target device. As a result of synthesis through the Vivado program, it has a maximum operating frequency of 479.616MHz and shows that real-time processing is possible in 4K UHD environment.

• Journal of IKEEE
• /
• v.25 no.1
• /
• pp.83-87
• /
• 2021

In harsh environments such as fog or fine dust, the cameras' detection ability for object recognition may significantly decrease. In order to accurately obtain important information even in bad weather, fog removal algorithms are necessarily required. Research has been conducted in various ways, such as computer vision/data-based fog removal technology. In those techniques, estimating the amount of fog through the input image's depth information is an important procedure. In this paper, a linear model is presented under the assumption that the image dark channel dictionary, saturation ∗ value, and sharpness characteristics are linearly related to depth information. The proposed method of haze removal through a linear model shows the superiority of algorithm performance in quantitative numerical evaluation.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.6_2
• /
• pp.1025-1032
• /
• 2020

In this paper, we propose a method to classify road weather conditions into rain, fog, and sun using a SVM (Support Vector Machine) classifier after extracting weather features from images acquired in real time using an optical sensor installed on a roadside post. A multi-dimensional weather feature vector consisting of factors such as image sharpeness, image entropy, Michelson contrast, MSCN (Mean Subtraction and Contrast Normalization), dark channel prior, image colorfulness, and local binary pattern as global features of weather-related images was extracted from road images, and then a road weather classifier was created by performing machine learning on 700 sun images, 2,000 rain images, and 1,000 fog images. Finally, the classification performance was tested for 140 sun images, 510 rain images, and 240 fog images. Overall classification performance is assessed to be applicable in real road services and can be enhanced further with optimization along with year-round data collection and training.

• Journal of Korean Society for Atmospheric Environment
• /
• v.21 no.3
• /
• pp.303-313
• /
• 2005

In this study, meteorological characteristics concerning the occurrence of fog are analyzed using 4-years $(2000\~2003)$ data at Chuncheon and the probability of prediction is investigated. From the analysis of meteorological characteristics, the fog at Chuncheon occurred before sunrise time and disappeared after that time and lasted for $2\~4$ hours. When fog occurred, on the whole, wind direction was blew the northerly and wind speed was below 2.1m/s. Especially, about $42\%$ of foggy day fell on the calm $(0\~0.2\;ms^{-1})$ conditions. The difference between air temperature and dew point temperature near the surface were mainly less than $2^{\circ}C$. For the lack of water surface temperature, the water surface temperature was calculated by using Water Quality River Reservoir System (WQRRS) and then it was used as the surface boundary condition of MM5. The numerical experiment was carried out for 2 days from 1300 LST on 14 October 2003 to 1300 LST on 16 October 2003 and fog was simulated at dawn on 15 and 16 October 2003. Simulated air temperature and dew point temperature indicate the similar tendency to observation and the simulated difference between air temperature and dew point temperature has also the similar tendency within $2^{\circ}C$. Thus, the occurrence of fog is well simulated in the terms of the difference between air temperature and dew point temperature. Horizontal distribution of the difference between air temperature and dew point temperature from the numerical experiment indicates occurrence, dissipation and lasting time of fog at Chuncheon. In Chuncheon, there is close correlation between the frequency of fog day and outflow from Soyang reservoir and high frequency of occurrence due to the difference between air and cold outlet water temperature.

• Asia-Pacific Journal of Atmospheric Sciences
• /
• v.54 no.4
• /
• pp.527-544
• /
• 2018

This paper presents a nighttime sea fog detection algorithm incorporating unsupervised learning technique. The algorithm is based on data sets that combine brightness temperatures from the $3.7{\mu}m$ and $10.8{\mu}m$ channels of the meteorological imager (MI) onboard the Communication, Ocean and Meteorological Satellite (COMS), with sea surface temperature from the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA). Previous algorithms generally employed threshold values including the brightness temperature difference between the near infrared and infrared. The threshold values were previously determined from climatological analysis or model simulation. Although this method using predetermined thresholds is very simple and effective in detecting low cloud, it has difficulty in distinguishing fog from stratus because they share similar characteristics of particle size and altitude. In order to improve this, the unsupervised learning approach, which allows a more effective interpretation from the insufficient information, has been utilized. The unsupervised learning method employed in this paper is the expectation-maximization (EM) algorithm that is widely used in incomplete data problems. It identifies distinguishing features of the data by organizing and optimizing the data. This allows for the application of optimal threshold values for fog detection by considering the characteristics of a specific domain. The algorithm has been evaluated using the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) vertical profile products, which showed promising results within a local domain with probability of detection (POD) of 0.753 and critical success index (CSI) of 0.477, respectively.