• Journal of Ocean Engineering and Technology
• /
• v.38 no.3
• /
• pp.103-114
• /
• 2024

Remote sensing wave observation data are crucial when analyzing ocean waves, the main external force of coastal disasters. Nevertheless, it has limitations in accuracy when used in low-wind environments. Therefore, this study collected the raw data from MIROS Wave and Current Radar (MWR) and wave radar at the Ieodo Ocean Research Station (IORS) and applied the optimal filter by combining filters provided by MIROS software. The data were validated by a comparison with South Jeju ocean buoy data. The results showed it maintained accuracy for significant wave height, but errors were observed in significant wave periods and extreme waves. Hence, this study used an artificial neural network (ANN) to improve these errors. The ANN was generalized by separating the data into training and test datasets through stratified sampling, and the optimal model structure was derived by adjusting the hyperparameters. The application of ANN effectively improved the accuracy in significant wave periods and high wave conditions. Consequently, this study reproduced past wave data by enhancing the reliability of the MWR, contributing to understanding wave generation and propagation in storm conditions, and improving the accuracy of wave prediction. On the other hand, errors persisted under high wave conditions because of wave shadow effects, necessitating more data collection and future research.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1159-1172
• /
• 2020

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2846-2866
• /
• 2022

Caenorhabditis elegans exhibits sophisticated chemotaxis behavior through two parallel strategies, klinokinesis and klinotaxis, executed entirely by a small nervous circuit. It is therefore suitable for inspiring fast and energy-efficient solutions for autonomous navigation. As a random search strategy, the Lévy walk is optimal for diverse animals when foraging without external chemical cues. In this study, by combining these biological strategies for the first time, we propose a spiking neural network model for search and contour tracking of specific concentrations of environmental variables. Specifically, we first design a klinotaxis module using spiking neurons. This module works in conjunction with a klinokinesis module, allowing rapid searches for the concentration setpoint and subsequent contour tracking with small deviations. Second, we build a random exploration module. It generates a Lévy walk in the absence of concentration gradients, increasing the chance of encountering gradients. Third, considering local extrema traps, we develop a termination module combined with an escape module to initiate or terminate the escape in a timely manner. Experimental results demonstrate that the proposed model integrating these modules can switch strategies autonomously according to the information from a single sensor and control steering through output spikes, enabling the model worm to efficiently navigate across various scenarios.

• Journal of Digital Contents Society
• /
• v.19 no.7
• /
• pp.1333-1339
• /
• 2018

Recently, as disasters caused by weather such as heavy rains have increased, interests in forecasting weather and disasters using radars have been increasing, and related studies have also been actively performed. As the Ministry of Environment(ME) has established and operated a radar network on a national scale, utilization of radars has been emphasized. However, persons in charge and researchers, who want to use the data from radars need to understand characteristics of the radar data and are also experiencing a lot of trials and errors when converting and calibrating the radar data from Universal Format(UF) files. Hence, this study developed a Radar Display and Analysis Program(RaDAP) based on Graphic User Interface(GUI) using the Java Programming Language in order for UF-type radar data to be generated in an ASCII-formatted image file and text file. The developed program can derive desired radar rainfall data and minimize the time required to perform its analysis. Therefore, it is expected that this program will contribute to enhancing the utilization of radar data in various fields.

• Journal of Korea Multimedia Society
• /
• v.25 no.1
• /
• pp.1-9
• /
• 2022

Ship detection in synthetic aperture radar (SAR) images is an important application in marine monitoring for the military and civilian domains. Over the past decade, object detection has achieved significant progress with the development of convolutional neural networks (CNNs) and lot of labeled databases. However, due to difficulty in collecting and labeling SAR images, it is still a challenging task to solve SAR ship detection CNNs. To overcome the problem, some methods have employed conventional data augmentation techniques such as flipping, cropping, and affine transformation, but it is insufficient to achieve robust performance to handle a wide variety of types of ships. In this paper, we present a novel and effective approach for deep SAR ship detection, that exploits label-rich Electro-Optical (EO) images. The proposed method consists of two components: a data augmentation network and a ship detection network. First, we train the data augmentation network based on conditional generative adversarial network (cGAN), which aims to generate additional SAR images from EO images. Since it is trained using unpaired EO and SAR images, we impose the cycle-consistency loss to preserve the structural information while translating the characteristics of the images. After training the data augmentation network, we leverage the augmented dataset constituted with real and translated SAR images to train the ship detection network. The experimental results include qualitative evaluation of the translated SAR images and the comparison of detection performance of the networks, trained with non-augmented and augmented dataset, which demonstrates the effectiveness of the proposed framework.

• Journal of the Korean earth science society
• /
• v.38 no.7
• /
• pp.496-510
• /
• 2017

This study presents a new nationwide quantitative precipitation estimation (QPE) based on the hybrid surface rainfall (HSR) technique using the weather radar network of Korea Meteorological Administration (KMA). This new nationwide HSR is characterized by the synthesis of reflectivity at the hybrid surface that is not affected by ground clutter, beam blockage, non-meteorological echoes, and bright band. The nationwide HSR is classified into static (STATIC) and dynamic HSR (DYNAMIC) mosaic depending on employing a quality control process, which is based on the fuzzy logic approach for single-polarization radar and the spatial texture technique for dual-polarization radar. The STATIC and DYNAMIC were evaluated by comparing with official and operational radar rainfall mosaic (MOSAIC) of KMA for 10 rainfall events from May to October 2014. The correlation coefficients within the block region of STATIC, DYNAMIC and MOSAIC are 0.52, 0.78, and 0.69, respectively, and their mean relative errors are 34.08, 30.08, and 40.71%.

• Journal of Wetlands Research
• /
• v.12 no.1
• /
• pp.51-61
• /
• 2010

Changes in climate have largely increased concentrated heavy rainfall, which in turn is causing enormous damages to humans and properties. Therefore, it is important to understand the spatial-temporal features of rainfall. In this study, RADAR rainfall was used to calculate gridded areal rainfall which reflects the spatial-temporal variability. In addition, Kalman-filter method, a stochastical technique, was used to combine ground rainfall network with RADAR rainfall network to calculate areal rainfall. Thiessen polygon method, Inverse distance weighting method, and Kriging method were used for calculating areal rainfall, and the calculated data was compared with adjusted areal RADAR rainfall measured using the Kalman-filter method. The result showed that RADAR rainfall adjusted with Kalman-filter method well-reproduced the distribution of raw RADAR rainfall which has a similar spatial distribution as the actual rainfall distribution. The adjusted RADAR rainfall also showed a similar rainfall volume as the volume shown in rain gauge data. Anseong-Cheon basin was used as a study area and the RADAR rainfall adjusted with Kalman-filter method was applied in $Vflo^{TM}$ model, a physical-based distributed model, and ModClark model, a semi-distributed model. As a result, $Vflo^{TM}$ model simulated peak time and peak value similar to that of observed hydrograph. ModClark model showed good results for total runoff volume. However, for verifying the parameter, $Vflo^{TM}$ model showed better reproduction of observed hydrograph than ModClark model. These results confirmed that flood runoff simulation is applicable in domestic settings(in South Korea) if highly accurate areal rainfall is calculated by combining gauge rainfall and RADAR rainfall data and the simulation is performed in link to the distributed hydrological model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.11
• /
• pp.2292-2300
• /
• 2011

Exact and rapid acquirement of ocean environment information is going to become more of an indispensable source of naval operations. Ocean surface measurements using High Frequency (HF) radar, which covers about 10-220km and has spatial resolution of 0.3-12km, have being operated in our country. It remotely observe and transmit realtime sea surface currents and waves. In the near future, the HF radar systems will be established along the whole coastal area. A performance of network of HF radar will support various marine and naval activities. Operational uses of HF radar for enhancing naval operation ability are suggested.

• Journal of electromagnetic engineering and science
• /
• v.11 no.3
• /
• pp.213-219
• /
• 2011

Simulative target generators are needed for testing and calibrating various radar systems. The generator in this study discriminates the transmitting frequency from a radar and simulates parameters like target range, range rate, and atmospheric attenuation using the digital RF memory technique. The simulative target echo is then sent to the radar for testing and evaluation. This paper proposes a novel architecture for controlling the digital RF memory so it continually writes ADC data to the memory and reads it for the DAC with increasing one step address in order to control the delay of target range in a simple way. The target echo is programmed according to various preprogrammed scenarios and is generated in real time using a wireless local area network (LAN). To analyze the detected and generated target information easily, the system times for the radar and simulative target generator are synchronized using a global positioning system (GPS).

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.341-348
• /
• 2011

Exact and rapid acquirement of ocean environment information is going to become more of an indispensable source of naval operations. Ocean surface measurements using High Frequency (HF) radar, which covers about 10-220km and has spatial resolution of 0.3-12km, have being operated in our country. It remotely observe and transmit realtime sea surface currents and waves. In the near future, the HF radar systems will be established along the whole coastal area. A performance of network of HF radar will support various marine and naval activities. Operational uses of HF radar for enhancing naval operation ability are suggested.