http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
유지희(Jihee Yoo),박지훈(Jihoon Park),배민지(Minji Bae),최현주(Hyunjoo Choi),서인호(Inho Seo),김경근(Kyeongkeun Kim) 한국국방우주학회 2023 한국국방우주학회지 Vol.1 No.1
본 논문에서는 위성에 탑재되는 SAR(Synthetic Aperture Radar) 센서의 군사적 관점에서의 활용에 대해 설명하고 정찰위성에서 수집하는 SAR 영상을 이용하여 관심 군사표적을 자동으로 탐지하고 식별함으로써 우리 군의 작전능력 향상에 기여할 수 있는 방법에 대해 기술한다. 센서 및 위성 관련 하드웨어 기술이 발전함에 따라 SAR 영상의 품질 및 획득 주기가 향상되고 있는데 이러한 고해상도 SAR 영상을 최대한 활용하기 위해서 필요한 표적 시뮬레이션 기술, 데이터베이스 구축 및 여러 센서들로부터 획득되는 데이터의 운용방안에 대해 제시한다. The paper explains the military use of SAR(Synthetic Aperture Radar) sensors mounted on satellites from a military point of view and describes how to contribute to the improvement of our military’s operational capabilities by automatically detecting and recognizing military targets of interest using SAR images collected from reconnaissance satellites. As sensor- and satellite-related hardware technologies develop, the quality and acquisition cycle of SAR images are improving. The database construction techniques involving the target simulation techniques are presented to make a good use of these high resolution SAR images and the need for a system is suggested which systematically archives, maintains and distributes a huge amount of data collected from various sensors.
Siamese 네트워크 기반 SAR 표적영상 간 유사도 분석
박지훈 한국군사과학기술학회 2022 한국군사과학기술학회지 Vol.25 No.5
Different from the field of electro-optical(EO) image analysis, there has been less interest in similarity metrics between synthetic aperture radar(SAR) target images. A reliable and objective similarity analysis for SAR target images is expected to enable the verification of the SAR measurement process or provide the guidelines of target CAD modeling that can be used for simulating realistic SAR target images. For this purpose, this paper presents a similarity analysis method based on the siamese network that quantifies the subjective assessment through the distance learning of similar and dissimilar SAR target image pairs. The proposed method is applied to MSTAR SAR target images of slightly different depression angles and the resultant metrics are compared and analyzed with qualitative evaluation. Since the image similarity is somewhat related to recognition performance, the capacity of the proposed method for target recognition is further checked experimentally with the confusion matrix.
합성 데이터를 이용한 SAR 지상표적의 딥러닝 탐지/분류 성능분석
박지훈 한국군사과학기술학회 2024 한국군사과학기술학회지 Vol.27 No.2
Based on the recently developed deep learning technology, many studies have been conducted on deep learningnetworks 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 ofwork carried out on the deep learning network aimed at detecting and classifying SAR ground targets and trainedwith the synthetic dataset generated from electromagnetic scattering simulations. In this respect, this paper presentsthe deep learning network trained with the synthetic dataset and applies it to detecting and classifying real SARground targets. With experiment results, this paper also analyzes the network performance according to thecomposition ratio between the real measured data and the synthetic data involved in network training. Finally, thesummary and limitations are discussed to give information on the future research direction.