• The Journal of the Acoustical Society of Korea
• /
• v.39 no.4
• /
• pp.292-302
• /
• 2020

Machine learning algorithms have made immense contributions in various fields including sonar and radar applications. Recently developed Cycle-Consistency Generative Adversarial Network (CycleGAN), a variant of GAN has been successfully used for unpaired image-to-image translation. We present a modified CycleGAN for translation of underwater ship engine sounds with high perceptual quality. The proposed network is composed of an improved generator model trained to translate underwater audio from one vessel type to other, an improved discriminator to identify the data as real or fake and a modified cycle-consistency loss function. The quantitative and qualitative analysis of the proposed CycleGAN are performed on publicly available underwater dataset ShipsEar by evaluating and comparing Mel-cepstral distortion, pitch contour matching, nearest neighbor comparison and mean opinion score with existing algorithms. The analysis results of the proposed network demonstrate the effectiveness of the proposed network.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.3
• /
• pp.52-58
• /
• 2009

If a moving target has a linear characteristics, the Kalman filter can estimate relatively accurate the location of a target, but this performance depends on how the dynamic status characteristics of the target is accurately modeled. In many practical problems of tracking a maneuvering target, a simple kinematic model can fairly accurately describe the target dynamics for a wide class of maneuvers. However, since the target can exhibit a wide range of dynamic characteristics, no fixed SKF(Simple Kalman filter) can be matched to estimate, to the required accuracy, the states of the target for every specific maneuver. In this paper, a new AKF(Active Kalman filter) is proposed to solve this problem The process noise covariance level of the Kalman filter is adjusted at each time step according to the study result which uses the neural network algorithm. It is demonstrated by means of a computer simulation that the tracking capability of the proposed AKF(Active Kalman filter) is better than that of the SKF(Simple Kalman Filter).

• Korean Journal of Remote Sensing
• /
• v.27 no.5
• /
• pp.521-534
• /
• 2011

The high-resolution X-band SAR systems such as COSMO-SkyMED and TerraSAR-X have been launched recently. In addition KOMPSAT-5 will be launched in the early of 2012. In this study we developed the new method for persistent scatterer candidate (PSC) selection and network construction, which is more suitable for PSInSAR analysis using multi-temporal X-band SAR data. PSC selection consists in two main steps: first, selection of initial PSCs based on amplitude dispersion index, mean amplitude, mean coherence. second, selection of final PSCs based on temporal coherence directly estimated from network analysis of initial PSCs. To increase the stability of network the Multi- TIN and complex network for non-urban area were addressed as well. The proposed algorithm was applied to twenty-one TerraSAR-X SAR of New Orleans. As a result many PSs were successfully extracted even in non-urban area. This research can be used as the practical application of KOMPSAT-5 for surface displacement monitoring using X-band PSInSAR.

• Proceedings of the KSRS Conference
• /
• 2007.03a
• /
• pp.153-157
• /
• 2007

본 논문은 지표면 현상의 관측에 날씨의 영향을 거의 받지 않는 마이크로파 L-밴드(1.95 GHz)와 C-밴드(5.3 GHz) scatterometer 시스템을 이용하여 농업과학기술원 내의 논에서 자라는 추청벼를 대상으로 2006년 5월 29일부터 10월 9일까지 생육에 따른 군락의 후방산란계수를 관측한 데이터와 작물의 생육과의 관계를 살펴보고 또한，측정 시스템의 개요，측정 시스템의 보정 방법들을 기술하고자 한다. Scatterometer 시스템의 송 수신기로 HP 8753D 벡터 네트워크 분석기를 사용하며，타워 위에 안테나를 설치하여 3.4 m의 높이에서 측정하도록 하였다. L-밴 드와 C-밴드 scatterometer는 VV-, VH-, HV-, HH-편파를 측정하여 fully polarimetric한 데이터를 얻도록 설계된 레이더시스템으로 입사각을 $30^{\circ}{\sim}60^{\circ}$에서 $10^{\circ}$간격으로 각각 30개의 독립적인 샘플을 측정하여 통계적으로 후방산란계수를 얻었다. 타워에서 발생하는 전파 잡음과 안테나 패턴의 부엽에 의한 지면에서의 수직반사(coherent 성분) 전파를 제거하기 위해 네트워크 분석기의 time gating 기능을 사용하며，55 cm 크기의 trihedral 전파반사기를 보정용 반사기로 사용하고， STCT(single target calibration technique) 방법을 이용하여 시스템을 보정하였다. 측정 결과를 분석하여 주파수， 입사각도， 편파의 변화에 대한 벼의 후방산란 특성과 벼의 생육상태과의 관계를 살펴보았다. L-밴드와 C-밴드 모두 벼의 생육과 밀접한 결과를 나타내었으나，입사각이 작을 때는 C-밴드와의 상관이 높게 나타났고 입사각이 커질수록 L-밴드와의 상관이 높게 나타났다. 편파는 L-밴드 와 C-밴드 모두 hh 편파가，입사각은 50도에서 가장 생육의 변이를 잘 설명하는 것으로 나타났다. 생육 데이터 모두를 이용한 경우보다는 유수형성기 또는 출수기 등 벼 생육의 질적인 변화를 보이는 시기에 따라 나누어 분석하는 것이 변화추이를 더 잘 설명하는 것으로 나타났다.

• Journal of Advanced Navigation Technology
• /
• v.18 no.4
• /
• pp.325-333
• /
• 2014

An ADS-B system, which is a critical technology in surveillance area of the CNS/ATM, can replace or compensate a conventional radar based surveillance system through the communications among aircrafts. An ADS-B ground system which is to use the ADS-B on the ground air traffic management system consists of various subsystem devices such as ground stations that communicate with the aircrafts, and ADS-B/TIS-B/FIS-B servers. The ADS-B ground system has a form of distributed system and is interconnected through the network. Therefore, a system which can monitor and control the status of the multiple subsystem devices of the ADS-B ground system is essential. In this paper, we designed and implemented a remote status monitor and control system for the ADS-B system that can monitor and control the subsystem devices of the ADS-B system in remote place via SNMP protocol.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.1
• /
• pp.99-106
• /
• 2015

In this study, we propose Radial basis function Neural Network(RBFNN) using Recursive Weighted Least Square Estimation(RWLSE) to effectively deal with big data class meteorological radar data. In the condition part of the RBFNN, Fuzzy C-Means(FCM) clustering is used to obtain fitness values taking into account characteristics of input data, and connection weights are defined as linear polynomial function in the conclusion part. The coefficients of the polynomial function are estimated by using RWLSE in order to cope with big data. As recursive learning technique, RWLSE which is based on WLSE is carried out to efficiently process big data. This study is experimented with both widely used some Machine Learning (ML) dataset and big data obtained from meteorological radar to evaluate the performance of the proposed classifier. The meteorological radar data as big data consists of precipitation echo and non-precipitation echo, and the proposed classifier is used to efficiently classify these echoes.

• Proceedings of KOSOMES biannual meeting
• /
• 2005.05a
• /
• pp.209-213
• /
• 2005

The current AIS network and. system are run on a full scale with 22 ground stations and. 11 operational systems, completing a nation-wide, integrated network However, currently it needs to manage sea traffic by linking AIS to VIS which 1vs a limited service area due to restricted radar detection zones in harbors or coastal areas. Accordingly this study analyzes the current status of the AIS system and. proposes technological and. operational improvement plan of the current AIS system through investigating the actual conditions of the AIS system and. its operations.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.3
• /
• pp.219-230
• /
• 2022

In implementing a robust automatic target recognition(ATR) system with synthetic aperture radar(SAR) imagery, one of the most important issues is accurate classification of target variants, which are the same targets with different serial numbers, configurations and versions, etc. In this paper, a deep learning network with channel attention modules is proposed to cope with the recognition problem for target variants based on the previous research findings that the channel attention mechanism selectively emphasizes the useful features for target recognition. Different from other existing attention methods, this paper employs the channel attention modules without dimensionality reduction along the channel direction from which direct correspondence between feature map channels can be preserved and the features valuable for recognizing SAR target variants can be effectively derived. Experiments with the public benchmark dataset demonstrate that the proposed scheme is superior to the network with other existing channel attention modules.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.2
• /
• pp.147-155
• /
• 2024

Based on the recently developed deep learning technology, many studies have been conducted on deep learning networks that simultaneously detect and classify targets of interest in synthetic aperture radar(SAR) images. Although numerous research results have been derived mainly with the open SAR ship datasets, there is a lack of work carried out on the deep learning network aimed at detecting and classifying SAR ground targets and trained with the synthetic dataset generated from electromagnetic scattering simulations. In this respect, this paper presents the deep learning network trained with the synthetic dataset and applies it to detecting and classifying real SAR ground targets. With experiment results, this paper also analyzes the network performance according to the composition ratio between the real measured data and the synthetic data involved in network training. Finally, the summary and limitations are discussed to give information on the future research direction.

• Journal of Navigation and Port Research
• /
• v.43 no.6
• /
• pp.403-412
• /
• 2019

The characteristics of the manoeuvring motion of a ship are dependent on the ship type, as well as draft or speed in the same ship. In recent years, the number of extra-large vessels has increased significantly, which can cause enormous material and environmental damage in the event of a marine accident. Thus, the importance of ship maneuvering is increasing. The IMO has forced the officers to be trained in simulators through the STCW 95 amendment. However, FMSS is costly and difficult to access and the PC-based simulator has the disadvantage that only one person can engage in simulation. The purpose of this study was to solve the shortcomings of the FMSS and PC-based simulators by enabling multiple people to use their PCs to simulate based on a network. The simulator is implemented through the analysis and numerical calculation of the Nomoto model, Radar function mounting, data transfer protocol design, and GUI building. To verify the simulator, the simulation results were compared and analyzed with the test results of T.S. HANBADA according to the criteria of the Korean Register of Shipping(KR) and IMO standards for ship maneuverability. As a result, It showed a relative error of 0%~ 32.1% with an average of 13.7%, and it satisfied the IMO criteria for ship maneuverability.