• Atmosphere
• /
• v.23 no.1
• /
• pp.103-112
• /
• 2013

In this study, we developed a fog classification algorithm to classify fog type based on fog generation mechanism. For the analysis period of 1986-2005, 15,748 fog events had been reported from the 40 observational sites in South Korea. Thus, practically, it is almost impossible to individually classify the fog type of the whole fog events occurred in South Korea manually. In this study, the characteristics of fog during the research period were investigated and the fog classification flowchart were developed base on the analysis, and the fog classification algorithm was applied for the classification of fogs occurred at the observational sites. Finally, the classified fog-type and hindcasted fog occurance results obtained from the flowchart were evaluated for verification.

• Atmosphere
• /
• v.29 no.5
• /
• pp.639-658
• /
• 2019

The classification of fog type and the characteristics of fog based on fog events over South Korea were investigated using a 3-year (2015~2017) visibility meter data. One-minute visibility meter data were used to identify fog with present weather codes and surface observation data. The concept of fog events was adopted for the better definition of fog properties and more objective classification through the detailed investigation of life cycle of fog. Decision tree method was used to classify the fog types and the final fog types were radiation fog, advection fog, precipitation fog, cloud base lowering fog and morning evaporation fog. We enhanced objectivity in classifying the types of fog by adding the satellite and the buoy observations to the conventional usage of AWS and ceilometer data. Radiation fog, the most common type in South Korea, frequently occurs in inland during autumn. A considerable number of advection fogs occur in island area in summer, especially in July. Precipitation fog accounts for more than a quarter of the total fog events and frequently occurs in islands and coastal areas. Cloud base lowering fog, classified using ceilometer, occurs occasionally for all areas but the occurrence rate is relatively high in east and west coastal area. Morning evaporation fog type is rarely observed in inland. The occurrence rate of thick fog with visibility less than 100 meters is amount to 21% of total fog events. Although advection fog develops into thick fog frequently, radiation fog shows the minimum visibility, in some cases.

• Journal of Environmental Science International
• /
• v.32 no.12
• /
• pp.883-898
• /
• 2023

To investigate the occurrence characteristics and types of fog on the Korean Peninsula over the past three years (2020 to 2022), data from 96 synoptic meteorological observatories and 21 ocean buoys were collected and analyzed. We included precipitation fog, which occurs after precipitation events, and cloud-base lowering fog, which is caused by the development of lower-level clouds, with a total six subtypes of fog. In the case of cloud-base lowering fog, the occurrence frequency at 2.6% was not high at 2.6%, but the duration of low visibility below 200 m was very long at 6.9 hours. The seasonal frequency of fog is low in spring and winter, high in summer over islands and coastal areas, and high in autumn over inland areas. The frequency of inland fog, which is characterized by high radiation fog and dense fog, requires attention in terms of transportation safety, with an occurrence time of 0500 LST to 1000 LST. Therefore, systematic analysis of precipitation fog and cloud-base lowering, as well as radiation and advection fog, is required in the analysis of recognizing fog as a disaster and causing transportation disorders.

• Journal of Environmental Science International
• /
• v.24 no.5
• /
• pp.657-670
• /
• 2015

Visibility and Automatic Weather System(AWS) data near Nakdong river were analyzed to characterize fog formation during 2012-2013. The temperature was lower than its nearby city - Daegu, whereas the humidity was higher than the city. 157 fog events were observed in total during the 2 year period. About 65% of the events occurred in fall (September, October, and November) followed by winter, summer, and spring. 94 early morning fog events of longer than 30 minutes occurred when south westerly wind speed was lower than 2 m/s. During these events, the water temperature was highest followed by soil surface and air temperatures due to the advection of cold and humid air from nearby hill. The observed fog events were categorized using a fog-type classification algorithm, which used surface cooling, wind speed threshold, rate of change of air temperature and dew point temperature. As a result, frontal fog observed 6 times, radiation 4, advection 13, and evaporation 66. The evaporation fog in the study area lasted longer than other reports. It is due to the interactions of cold air drainage flow and warm surface in addition to the evaporation from the water surface. In particular, more than 60% of the evaporation fog events were accompanied with cold air flows over the wet and warm surface. Therefore, it is needed for the identification of the inland fog mechanism to evaluate the impacts of nearby topography and land cover as well as water body.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.13 no.1
• /
• pp.67-72
• /
• 2013

Cloud detection and analysis from satellite images has been a topic of research in many atmospheric and environmental studies; however, it still is a challenging task for many reasons. In this paper, we propose a new method for cloud-type classification using fuzzy logic. Knowing that visible-light images of clouds contain thickness related information, while infrared images haves height-related information, we propose a two-layered fuzzy logic based on the input source to provide us with a relatively clear-cut threshold in classification. Traditional noise-removal methods that use reflection/release characteristics of infrared images often produce false positive cloud areas, such as fog thereby it negatively affecting the classification accuracy. In this study, we used the color information from source images to extract the region of interest while avoiding false positives. The structure of fuzzy inference was also changed, because we utilized three types of source images: visible-light, infrared, and near-infrared images. When a cloud appears in both the visible-light image and the infrared image, the fuzzy membership function has a different form. Therefore we designed two sets of fuzzy inference rules and related classification rules. In our experiment, the proposed method was verified to be efficient and more accurate than the previous fuzzy logic attempt that used infrared image features.

• International journal of advanced smart convergence
• /
• v.10 no.4
• /
• pp.263-272
• /
• 2021

At a time when securing driving safety is the most important in the development and commercialization of autonomous vehicles, AI and big data-based algorithms are being studied to enhance and optimize the recognition and detection performance of various static and dynamic vehicles. However, there are many research cases to recognize it as the same vehicle by utilizing the unique advantages of radar and cameras, but they do not use deep learning image processing technology or detect only short distances as the same target due to radar performance problems. Radars can recognize vehicles without errors in situations such as night and fog, but it is not accurate even if the type of object is determined through RCS values, so accurate classification of the object through images such as cameras is required. Therefore, we propose a fusion-based vehicle recognition method that configures data sets that can be collected by radar device and camera device, calculates errors in the data sets, and recognizes them as the same target.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.10
• /
• pp.4865-4885
• /
• 2019

We propose a FPGA based design that performs real-time power-efficient analysis of heterogeneous sensor data using adaptive ANN on edge gateway of smart military wearables. In this work, four independent ANN classifiers are developed with optimum topologies. Out of which human activity, BP and toxic gas classifier are multiclass and ECG classifier is binary. These classifiers are later integrated into a single adaptive ANN hardware with a select line(s) that switches the hardware architecture as per the sensor type. Five versions of adaptive ANN with different precisions have been synthesized into IP cores. These IP cores are implemented and tested on Xilinx Artix-7 FPGA using Microblaze test system and LabVIEW based sensor simulators. The hardware analysis shows that the adaptive ANN even with 8-bit precision is the most efficient IP core in terms of hardware resource utilization and power consumption without compromising much on classification accuracy. This IP core requires only 31 microseconds for classification by consuming only 12 milliwatts of power. The proposed adaptive ANN design saves 61% to 97% of different FPGA resources and 44% of power as compared with the independent implementations. In addition, 96.87% to 98.75% of data throughput reduction is achieved by this edge gateway.