• Journal of IKEEE
• /
• v.24 no.3
• /
• pp.777-782
• /
• 2020

In this paper, we propose CNN-based HMI system using Doppler radar and voice sensor, and present hardware design and implementation results. To overcome the limitation of single sensor monitoring, the proposed HMI system combines data from two sensors to improve performance. The proposed system exhibits improved performance by 3.5% and 12% compared to a single radar and voice sensor-based classifier in noisy environment. In addition, hardware to accelerate the complex computational unit of CNN is implemented and verified on the FPGA test system. As a result of performance evaluation, the proposed HMI acceleration platform can be processed with 95% reduction in computation time compared to a single software-based design.

• Journal of Korea Water Resources Association
• /
• v.48 no.12
• /
• pp.995-1009
• /
• 2015

Weather radar rainfall data has been recognized for making valuable contributions to short-term flood forecasting and management over the past decades. There are several advantages to better monitoring rainfall in ungauged area compared to ground-based rain gauges with which spatial patterns of the rainfall are not effectively identified. Hence, this study aims to develop a new scheme to forecast spatio-temporal rainfall field. The proposed model was based on an advection scheme to track wind patterns and velocity. The results showd a promising forecasting skill with quantitative and qualitative measures. It was confirmed that the forecasted rainfall may be effectively used an input data for a distributed hydrological model.

• Smart Structures and Systems
• /
• v.26 no.3
• /
• pp.277-290
• /
• 2020

Disasters, including earthquakes and landslides, have enormous economic and social losses besides their impact on environmental disruption. Iran, and particularly its Western part, is known as an earthquake susceptible area due to numerous strong ground motions. Studying ecological changes due to climate change can improve the public and expert sector's awareness and response to future disastrous events. Synthetic Aperture Radar (SAR) data and Interferometric Synthetic Aperture Radar (InSAR) technologies are appropriate tools for modeling and surface deformation modeling. This paper proposes an efficient approach to detect ground deformation changes using Sentinel-1A. The focal point of this research is to map the ground surface deformation modeling is presented using InSAR technology over Sarpol-e Zahab on 25^th November 2018 as a study case. For surface deformation modeling and detection of the ground movement due to earthquake SARPROZ in MATLAB programming language is used and discussed. Results show that there is a general ground movement due to the Sarpol-e Zahab earthquake between -7 millimeter to +18 millimeter in the study area. This research verified previous researches on the advanced image analysis techniques employed for mapping ground movement, where InSAR provides a reliable tool for assisting engineers and the decision-maker in choosing proper policies in a time of disasters. Based on the result, 574 out of 682 damaged buildings and infrastructures due to the 2017 Sarpol-e Zahab earthquake have moved from -2 to +17 mm due to the 2018 earthquake with a magnitude of 6.3 Richter. Results show that mountainous areas have suffered land subsidence, where urban areas had land uplift.

• Korean Journal of Geomatics
• /
• v.2 no.1
• /
• pp.37-46
• /
• 2002

Accurate classification of water area is an preliminary step to accurately analyze the flooded area and damages caused by flood. This step is especially useful for monitoring the region where annually repeating flood is a problem. The accurate estimation of flooded area can ultimately be utilized as a primary source of information for the policy decision. Although SAR (Synthetic Aperture Radar) imagery with its own energy source is sensitive to the water area, its shadow effect similar to the reflectance signature of the water area should be carefully checked before accurate classification. Especially when we want to identify small flood area with mountainous environment, the step for removing shadow effect turns out to be essential in order to accurately classify the water area from the SAR imagery. In this paper, the flood area was classified and monitored using multi-temporal RADARSAT SAR images of Ok-Chun and Bo-Eun located in Chung-Book Province taken in 12th (during the flood) and 19th (after the flood) of August, 1998. We applied several steps of geometric and radiometric calculations to the SAR imagery. First we reduced the speckle noise of two SAR images and then calculated the radar backscattering coefficient $(\sigma^0)$. After that we performed the ortho-rectification via satellite orbit modeling developed in this study using the ephemeris information of the satellite images and ground control points. We also corrected radiometric distortion caused by the terrain relief. Finally, the water area was identified from two images and the flood area is calculated accordingly. The identified flood area is analyzed by overlapping with the existing land use map.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.197-202
• /
• 2019

A small millimeter-wave tracking radar power supply must provide stable power with minimal ripple noise and the switching frequency noise of the DC-DC converter must have a real-time self-test capability through on-the-fly monitoring without causing false alarms and ghost In this study, we developed a multi-output switching power supply with output power of more than 80% (@ 100% load) and 10 output power by adopting + 28VDC input for application to small millimeter wave tracking radar, DC-DC converter is applied for large power output and multi-output flyback method is applied for the remaining small power output. The test results show that 85% efficiency efficiency is achieved under 100% load condition.

• 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.

• The Journal of Engineering Geology
• /
• v.32 no.2
• /
• pp.271-280
• /
• 2022

Landfast sea ice forms near coastlines in polar regions. Continuous monitoring of this sea ice is important, as it plays a key role in the marine ecosystem and affects the operation of nearby research stations. This study detected landfast sea ice around Jang Bogo research station in East Antarctica by stacking interferometric coherence images of Sentinel-1 synthetic aperture radar (SAR) data with 6-, 12- and 18-day temporal baselines. A total of 50 landfast sea ice maps were generated covering July 2017 to June 2018. The time series revealed regional differences in the timing of the maximum extent as well as growth rate of landfast sea ice. Overall, detecting landfast sea ice using interferometric SAR coherence seems promisingly feasible; however, limitations remain owing to low backscattering coefficients from new and smooth sea ice surfaces and subtle movements of sea ice in contact with the Campbell Glacier Tongue.

• Korean Journal of Remote Sensing
• /
• v.40 no.3
• /
• pp.285-293
• /
• 2024

This study explores the use of Sentinel-1 Synthetic Aperture Radar (SAR), processed through Google Earth Engine (GEE), to monitor changes in the areas of Antarctic ice shelves. Focusing on the Campbell Glacier Tongue (CGT) and Drygalski Ice Tongue (DIT),the research utilizes GEE's cloud computing capabilities to handle and analyze large datasets. The study employs Otsu's method for image binarization to distinguish ice shelves from the ocean and mitigates detection errors by averaging monthly images and extracting main regions. Results indicate that the CGT area decreased by approximately 26% from January 2016 to January 2024, primarily due to calving events,while DIT showed a slight increase overall,with notable reduction in recent years. Validation against Sentinel-2 optical images demonstrates high accuracy,underscoring the effectiveness of SAR and GEE for continuous, long-term monitoring of Antarctic ice shelves.

• Korean Journal of Remote Sensing
• /
• v.29 no.1
• /
• pp.11-20
• /
• 2013

Space-borne Synthetic Aperture Radar(SAR) has been one of the most effective tools for monitoring quantitative oceanographic physical parameters. The Doppler information recorded in single-channel SAR raw data can be useful in estimating moving velocity of water mass in ocean. The Doppler shift is caused by the relative motion between SAR sensor and the water mass of ocean surface. Thus, the moving velocity can be extracted by measuring the Doppler anomaly between extracted Doppler centroid and predicted Doppler centroid. The predicted Doppler centroid, defined as the Doppler centroid assuming that the target is not moving, is calculated based on the geometric parameters of a satellite, such as the satellite's orbit, look angle, and attitude with regard to the rotating Earth. While the estimated Doppler shift, corresponding to the actual Doppler centroid in the situation of real SAR data acquisition, can be extracted directly from raw SAR signal data, which usually calculated by applying the Average Cross Correlation Coefficient(ACCC). The moving velocity was further refined to obtain ocean surface current by subtracting the phase velocity of Bragg-resonant capillary waves. These methods were applied to Envisat ASAR raw data acquired in the East Sea, and the extracted ocean surface currents were compared with the current measured by HF-radar.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.49 no.3
• /
• pp.301-310
• /
• 2013

To suggest a standard concerning with the arrangement of bridge equipment, the authors conducted the video observations with 3CCD (charge coupled device) cameras installed on the ceil of the bridge for monitoring the working activities of two bridge teams (the skipper/mate1 and the skipper/mate2) in a Korean coastal large trawler(gross tonnage: 139) for five days from July 30th. 2010 and analyzed of the data. Work elements coded by the work activities were input on the sheet of work analysis by the time unit of 1 sec according to the time occurred. A single work element among the work activities for every 5 minutes was denoted as the number of occurrence. The frequency of equipment usage was limited only in the usage of the equipment. In the case of the navigation and the towing net two ranks were integrated and analyzed. On the other hand, in the case of the casting net and the hauling net, two processes were integrated to as one and then analyzed separately as two ranks. As the results, 15 elements of work was carried out between two bridge teams for the observation; lookout, radar, GPS plotter, fish finder, net monitor, fishing deck, RPM indicator, rudder angle indicator, compass card, for maneuver; steering, ship speed control, trawl winch operation and external communications, paper works and others. It was found that the work load of the skipper per 5 minutes accordance with the navigation, the casting net, the towing net and the hauling net are 20.5 times, 11.9 times, 38.0 times and 9.5 times respectively, the mates are 65.2 times, 66.5 times, 85.7 times and 59.1 times respectively. The radar was shown the highest frequency of the equipment usage and the next was the fish finder, the GPS plotter and the external communications in the case of the navigation. In the case of the towing net the frequency of usage was high the ranking as the radar, the net monitor, the fish finder, the GPS plotter, the steering system and the external communications. In the case of the integrated process both of the casting and hauling net the trawl winch was shown the highest frequency to the skipper and the next was the GPS plotter and the radar, and the steering system was shown the highest frequency to the mate and the next was the radar, the ship speed control system, the GPS plotter, the net monitor and the fish finder.