• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.8
• /
• pp.3567-3582
• /
• 2020

Human activities are often affected by weather conditions. Automatic weather recognition is meaningful to traffic alerting, driving assistance, and intelligent traffic. With the boost of deep learning and AI, deep convolutional neural networks (CNN) are utilized to identify weather situations. In this paper, a three-channel convolutional neural network (3C-CNN) model is proposed on the basis of ResNet50.The model extracts global weather features from the whole image through the ResNet50 branch, and extracts the sky and ground features from the top and bottom regions by two CNN5 branches. Then the global features and the local features are merged by the Concat function. Finally, the weather image is classified by Softmax classifier and the identification result is output. In addition, a medium-scale dataset containing 6,185 outdoor weather images named WeatherDataset-6 is established. 3C-CNN is used to train and test both on the Two-class Weather Images and WeatherDataset-6. The experimental results show that 3C-CNN achieves best on both datasets, with the average recognition accuracy up to 94.35% and 95.81% respectively, which is superior to other classic convolutional neural networks such as AlexNet, VGG16, and ResNet50. It is prospected that our method can also work well for images taken at night with further improvement.

• Advances in Energy Research
• /
• v.6 no.2
• /
• pp.91-101
• /
• 2019

This paper proposes an efficient data-driven approach to build models for predicting energy consumption in buildings. Data used in this research is collected by installing humidity and temperature sensors at different locations in a building. In addition to this, weather data from nearby weather station is also included in the dataset to study the impact of weather conditions on energy consumption. One of the main emphasize of this research is to make feature selection independent of domain knowledge. Therefore, to extract useful features from data, two different approaches are tested: one is feature selection through principal component analysis and second is relative importance-based feature selection in original domain. The regression model used in this research is gradient boosting regression and its optimal parameters are chosen through a two staged coarse-fine search approach. In order to evaluate the performance of model, different performance evaluation metrics like r2-score and root mean squared error are used. Results have shown that best performance is achieved, when relative importance-based feature selection is used with gradient boosting regressor. Results of proposed technique has also outperformed the results of support vector machines and neural network-based approaches tested on the same dataset.

• Journal of the Korean Solar Energy Society
• /
• v.38 no.6
• /
• pp.27-36
• /
• 2018

Typical Meteorological Year Dataset is necessary for the renewable energy feasibility study. Since National Renewable Energy Laboratory has been built Typical Meteorological Year Dataset in 1978, gridded datasets taken from numerical weather prediction or satellite imagery are employed to produce Typical Meteorological Year Dataset. In general, Typical Meteorological Year Dataset is generated by using long-term in-situ observations. However, solar insolation is not usually measured at synoptic observing stations and therefore it is limited to build the Typical Meteorological Year Dataset with only in-situ observation. This study attempts to build the Typical Meteorological Year Dataset with satellite derived solar insolation as an alternative and then we evaluate the Typical Meteorological Year Dataset made by using satellite derived solar irradiance at Daejeon ground station. The solar irradiance is underestimated when satellite imagery is employed.

• The Bulletin of The Korean Astronomical Society
• /
• v.35 no.1
• /
• pp.33.2-33.2
• /
• 2010

We have investigated a relationship among the solar proton events (SPEs), coronal mass ejections (CMEs) and solar flares during the solar cycle 23 (1997-2006). Using 63 SPE dataset, we found that SPE rise time, duration time, and decrease times depend on CME speed and SPE peak intensity depends on the CME earthward direction parameter as well as CME speed and x-ray flare intensity. While inspecting the relation between SPE peak intensity and the CME earthward direction parameter, we found that there are two groups: first group consists of large 6 SPEs (> 10,000 pfu at >10 MeV proton channel of GOES satellite) and shows a very good correlation (cc=0.65) between SPE peak intensity and CME earthward direction parameter. The second group has a relatively weak SPE peak intensity and shows poor correlation between SPE peak intensity and the CME earthward direction parameter (cc=0.01). By investigating characteristics of 6 SPEs in the first group, we found that there are special common conditions of the extremely large proton events (group 1); (1) all the SPEs are associated with very fast halo CME (>1400km/s), (2) they are almost located at disk region, (3) they also accompany large flare (>M7), (4) all they are preceded by another wide CMEs, and (5) they all show helmet streamer nearby the main CME. In this presentation, we will give details of the energy spectra of the 6 SPE events from the ERNE/HED aboard the Solar and Heliospheric Observatory (SOHO), and onset time comparison among the SPE, flare, type II burst, and CME.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_3
• /
• pp.1711-1720
• /
• 2020

Particulate matter (PM) that has been artificially generated during the recent of rapid industrialization and urbanization moves and disperses according to weather conditions, and adversely affects the human skin and respiratory systems. The purpose of this study is to predict the PM10 concentration in Seoul using meteorological factors as input dataset for multiple linear regression (MLR), support vector machine (SVM), and random forest (RF) models, and compared and evaluated the performance of the models. First, the PM10 concentration data obtained at 39 air quality monitoring sites (AQMS) in Seoul were divided into training and validation dataset (8:2 ratio). The nine meteorological factors (mean, maximum, and minimum temperature, precipitation, average and maximum wind speed, wind direction, yellow dust, and relative humidity), obtained by the automatic weather system (AWS), were composed to input dataset of models. The coefficients of determination (R2) between the observed PM10 concentration and that predicted by the MLR, SVM, and RF models was 0.260, 0.772, and 0.793, respectively, and the RF model best predicted the PM10 concentration. Among the AQMS used for model validation, Gwanak-gu and Gangnam-daero AQMS are relatively close to AWS, and the SVM and RF models were highly accurate according to the model validations. The Jongno-gu AQMS is relatively far from the AWS, but since PM10 concentration for the two adjacent AQMS were used for model training, both models presented high accuracy. By contrast, Yongsan-gu AQMS was relatively far from AQMS and AWS, both models performed poorly.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1451-1466
• /
• 2023

In the context of maritime emergencies, the utilization of drones has rapidly increased, with a particular focus on their application in search and rescue operations. Deep learning models utilizing drone images for the rapid detection of distressed vessels and other maritime drift objects are gaining attention. However, effective training of such models necessitates a substantial amount of diverse training data that considers various weather conditions and vessel states. The lack of such data can lead to a degradation in the performance of trained models. This study aims to enhance the performance of deep learning models for distress ship detection by developing a maritime environment simulator to augment the dataset. The simulator allows for the configuration of various weather conditions, vessel states such as sinking or capsizing, and specifications and characteristics of drones and sensors. Training the deep learning model with the dataset generated through simulation resulted in improved detection performance, including accuracy and recall, when compared to models trained solely on actual drone image datasets. In particular, the accuracy of distress ship detection in adverse weather conditions, such as rain or fog, increased by approximately 2-5%, with a significant reduction in the rate of undetected instances. These results demonstrate the practical and effective contribution of the developed simulator in simulating diverse scenarios for model training. Furthermore, the distress ship detection deep learning model based on this approach is expected to be efficiently applied in maritime search and rescue operations.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.6
• /
• pp.1063-1068
• /
• 2016

Anomalous propagation echo is a kind of abnormal radar signal occurred by irregularly refracted radar beam caused by temperature or humidity. The echo frequently appears in ground-based weather radar due to its observation principle and disturb weather forecasting process. In order to improve accuracy of weather forecasting, it is important to analyze radar data precisely. Therefore, there are several ongoing researches about identifying the anomalous propagation echo with data mining techniques. This paper conducts researches about implementation of classification method which can separate the anomalous propagation echo in the raw radar data using naive Bayes classifier with various kinds of observation results. Considering that collected data has a class imbalanced problem, this paper includes SMOTE method. It is confirmed that the fine classification results are derived by the suggested classifier with balanced dataset using actual appearance cases of the echo.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.6
• /
• pp.706-714
• /
• 2012

The Korea Meteorological Administration provided the recent four-years records of weather dataset for our very short-term heavy rainfall prediction. We divided the dataset into three parts: train, validation and test set. Through feature selection, we select only important features among 72 features to avoid significant increase of solution space that arises when growing exponentially with the dimensionality. We used a differential evolution algorithm and two classifiers as the fitness function of evolutionary computation to select more accurate feature subset. One of the classifiers is Support Vector Machine (SVM) that shows high performance, and the other is k-Nearest Neighbor (k-NN) that is fast in general. The test results of SVM were more prominent than those of k-NN in our experiments. Also we processed the weather data using undersampling and normalization techniques. The test results of our differential evolution algorithm performed about five times better than those using all features and about 1.36 times better than those using a genetic algorithm, which is the best known. Running times when using a genetic algorithm were about twenty times longer than those when using a differential evolution algorithm.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.6 no.1
• /
• pp.61-69
• /
• 2004

A long-term growth simulation was performed at 99 land units in Yeoncheon county to test the potential adaptability of each land unit for growing soybean cultivars. The land units for soybean cultivation(CZU), each represented by a geographically referenced land patch, were selected based on land use, soil characteristics, and minimum arable land area. Monthly climatic normals for daily maximum and minimum temperature, precipitation, number of rain days and solar radiation were extracted for each CZU from digital climate models(DCM). The DCM grid cells falling within a same CZU were aggregated to make spatially explicit climatic normals relevant to the CZU. A daily weather dataset for 30 years was randomly generated from the monthly climatic normals of each CZU. Growth and development parameters of CROPGRO-soybean model suitable for 2 domestic soybean cultivars were derived from long-term field observations. Three foreign cultivars with well established parameters were also added to this study, representing maturity groups 3, 4, and 5. Each treatment was simulated with the randomly generated 30 years' daily weather data(from planting to physiological maturity) for 99 land units in Yeoncheon to simulate the growth and yield responses to the inter-annual climate variation. The same model was run with input data from the Crop Experiment Station in Suwon to obtain a 30 year normal performance of each cultivar, which was used as a "reference" for evaluation. Results were analyzed with respect to spatial and temporal variation in yield and maturity, and used to evaluate the suitability of each land unit for growing a specific cultivar. A computer program(MAPSOY) was written to help utilize the results in a decision-making procedure for agrotechnology transfer. transfer.

• Journal of Internet Computing and Services
• /
• v.22 no.6
• /
• pp.25-32
• /
• 2021

In line with future changes in the marine environment, Aids to Navigation has been used in various fields and their use is increasing. The term "Aids to Navigation" means an aid to navigation prescribed by Ordinance of the Ministry of Oceans and Fisheries which shows navigating ships the position and direction of the ships, position of obstacles, etc. through lights, shapes, colors, sound, radio waves, etc. Also now the use of Aids to Navigation is transforming into a means of identifying and recording the marine weather environment by mounting various sensors and cameras. However, Aids to Navigation are mainly lost due to collisions with ships, and in particular, safety accidents occur because of poor observation visibility due to sea fog. The inflow of sea fog poses risks to ports and sea transportation, and it is not easy to predict sea fog because of the large difference in the possibility of occurrence depending on time and region. In addition, it is difficult to manage individually due to the features of Aids to Navigation distributed throughout the sea. To solve this problem, this paper aims to identify the marine weather environment by estimating sea fog level approximately with images taken by cameras mounted on Aids to Navigation and to resolve safety accidents caused by weather. Instead of optical and temperature sensors that are difficult to install and expensive to measure sea fog level, sea fog level is measured through the use of general images of cameras mounted on Aids to Navigation. Furthermore, as a prior study for real-time sea fog level estimation in various seas, the sea fog level criteria are presented using the Haze Model and Dark Channel Prior. A specific threshold value is set in the image through Dark Channel Prior(DCP), and based on this, the number of pixels without sea fog is found in the entire image to estimate the sea fog level. Experimental results demonstrate the possibility of estimating the sea fog level using synthetic haze image dataset and real haze image dataset.