• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.3
• /
• pp.101-109
• /
• 2021

Wave observations using a marine X-band radar are conducted by analyzing the backscattered radar signal from sea surfaces. Wave parameters are extracted using Modulation Transfer Function obtained from 3D wave number and frequency spectra which are calculated by 3D FFT of time series of sea surface images (42 images per minute). The accuracy of estimation of the significant wave height is, therefore, critically dependent on the quality of radar images. Wave observations during Typhoon Maysak and Haishen in the summer of 2020 show large errors in the estimation of the significant wave heights. It is because of the deteriorated radar images due to raindrops falling on the sea surface. This paper presents the algorithm developed to increase the accuracy of wave heights estimation from radar images by adopting convolution neural network(CNN) which automatically classify radar images into rain and non-rain cases. Then, an algorithm for deriving the Hs is proposed by creating different ANN models and selectively applying them according to the rain or non-rain cases. The developed algorithm applied to heavy rain cases during typhoons and showed critically improved results.

• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.189-200
• /
• 2022

Ground subsidence on urban roads is a social issue that can lead to human and property damages. Therefore, it is crucial to detect underground cavities in advance and repair them. Underground cavity detection is mainly performed using ground penetrating radar (GPR) surveys. This process is time-consuming, as a massive amount of GPR data needs to be interpreted, and the results vary depending on the skills and subjectivity of experts. To address these problems, researchers have studied automation and quantification techniques for GPR data interpretation, and recent studies have focused on deep learning-based interpretation techniques. In this study, we described a hyperbolic event detection process based on deep learning for GPR data interpretation. To demonstrate this process, we implemented a series of algorithms introduced in the preexisting research step by step. First, a deep learning-based YOLOv3 object detection model was applied to automatically detect hyperbolic signals. Subsequently, only hyperbolic signals were extracted using the column-connection clustering (C3) algorithm. Finally, the horizontal locations of the underground cavities were determined using regression analysis. The hyperbolic event detection using the YOLOv3 object detection technique achieved 84% precision and a recall score of 92% based on AP50. The predicted horizontal locations of the four underground cavities were approximately 0.12 ~ 0.36 m away from their actual locations. Thus, we confirmed that the existing deep learning-based interpretation technique is reliable with regard to detecting the hyperbolic patterns indicating underground cavities.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.8
• /
• pp.253-263
• /
• 2013

Frequency Scanning Interferometry(FSI) system, one of the most promising optical surface measurement techniques, generally results in superior optical performance comparing with other 3-dimensional measuring methods as its hardware structure is fixed in operation and only the light frequency is scanned in a specific spectral band without vertical scanning of the target surface or the objective lens. FSI system collects a set of images of interference fringe by changing the frequency of light source. After that, it transforms intensity data of acquired image into frequency information, and calculates the height profile of target objects with the help of frequency analysis based on Fast Fourier Transform(FFT). However, it still suffers from optical noise on target surfaces and relatively long processing time due to the number of images acquired in frequency scanning phase. 1) a Polarization-based Frequency Scanning Interferometry(PFSI) is proposed for optical noise robustness. It consists of tunable laser for light source, ${\lambda}/4$ plate in front of reference mirror, ${\lambda}/4$ plate in front of target object, polarizing beam splitter, polarizer in front of image sensor, polarizer in front of the fiber coupled light source, ${\lambda}/2$ plate between PBS and polarizer of the light source. Using the proposed system, we can solve the problem of fringe image with low contrast by using polarization technique. Also, we can control light distribution of object beam and reference beam. 2) the signal processing acceleration method is proposed for PFSI, based on parallel processing architecture, which consists of parallel processing hardware and software such as Graphic Processing Unit(GPU) and Compute Unified Device Architecture(CUDA). As a result, the processing time reaches into tact time level of real-time processing. Finally, the proposed system is evaluated in terms of accuracy and processing speed through a series of experiment and the obtained results show the effectiveness of the proposed system and method.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.30 no.2
• /
• pp.188-202
• /
• 1997

The seasonal variability of the sea surface winds over the last Sea of Korea (Japan Sea) is investigated by means of empirical orthogonal function (EOF) analysis. The combined representation of fields of three climatic variables by empirical orthogonal functions is discussed. The eigenvectors are derived from daily sea level pressure, wind speed and 10-day mean sea surface temperature (SST) during 15 years $(1978\~1992)$. The spatial patterns of the mean pressure are characterized by the high pressure in the western part and the low pressure in the eastern part. The spatial distribution of the standard deviation (SD) of pressure are characterized by max SD of 6.6 mb near the Vladivostok, and minima along the coast of the Japan. In Vladivostok, the maxima of SD of SST and south-north wind (WV) were also occurred. The representation of fields of individual meteorological variables by EOF shows that the first mode of the west-east wind (WU) explain over $47.3\%$ of the variance and the second mode of WU represents $30\%$. Especially, the first mode of the WV explain $70.9\%$ of the variance and their time series coefficients show 1-cpy, 0.5-cpy frequency spectrum. The spatial distribution of the first mode eigenvectors of SST are characterized by maximum near Vladivostok. The combined representation of fields of several variables (pressure, wind, SST) reveals that the first mode magnitudes of the variance of the combined eigenvectors (WU-PR) are increased. By means of this result, the 1-year peak and the 6-months peak are remarkable. In the three combined patterns (wind, pressure, SST), the second mode of the eigenvector (wind) is affected by the SST. Their time coefficients of the first mode show noticeable 1-year peak. The spectral analysis of the second mode shows broad seasonal signal with the period of 4-months and a significant peak of variability at 3-month period.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_4
• /
• pp.1503-1517
• /
• 2018

Mt. Baekdu is a stratovolcano located at the border between China and North Korea and is known to have formed through its differentiation stage after the Oligocene epoch in the Cenozoic era. There has been a growing interest in the magma re-activity of Mt. Baekdu volcano since 2010. Several research projects have been conducted by government such as Korea Meteorological Administration and Korea Institute of Geoscience and Mineral Resources. Because, however, the Mt. Baekdu volcano is located far from South Korea, it is quite difficult to collect in-situ observations by terrestrial equipment. Remote sensing is a science to analyze and interpret information without direct physical contact with a target object. Various types of platform such as automobile, unmanned aerial vehicle, aircraft and satellite can be used for carrying a payload. In the past several decades, numerous volcanic studies have been conducted by remotely sensed observations using wide spectrum of wavelength channels in electromagnetic waves. In particular, radar remote sensing has been widely used for volcano monitoring in that microwave channel can gather surface's information without less limitation like day and night or weather condition. Radar interferometric technique which utilized phase information of radar signal enables to estimate surface displacement such as volcano, earthquake, ground subsidence or glacial movement, etc. In 2018, long-term research project for collaborative observation for Mt. Baekdu volcano between Korea and China were selected by Korea government. A volcanic specialized research center has been established by the selected project. The purpose of this paper is to introduce about remote sensing techniques for volcano monitoring and to review selected studies with remote sensing techniques to monitor Mt. Baekdu volcano. The acquisition status of the archived observations of six synthetic aperture radar satellites which are in orbit now was investigated for application of radar interferometry to monitor Mt. Baekdu volcano. We will conduct a time-series analysis using collected synthetic aperture radar images.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.