• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.5
• /
• pp.619-627
• /
• 2021

Recent advance in low cost and light-weight drones has extended their application areas in both military and private sectors. Accordingly surveillance program against unfriendly drones has become an important issue. Drone detection and classification technique has long been emphasized in order to prevent attacks or accidents by commercial drones in urban areas. Most commercial drones have small sizes and low reflection and hence typical sensors that use acoustic, infrared, or radar signals exhibit limited performances. Recently, artificial intelligence algorithm has been actively exploited to enhance radar image identification performance. In this paper, we adopt machined learning algorithm for high resolution radar imaging in drone detection and classification applications. For this purpose, simulation is carried out against commercial drone models and compared with experimental data obtained through high resolution radar field test.

• Journal of Satellite, Information and Communications
• /
• v.12 no.3
• /
• pp.63-68
• /
• 2017

SAR(Synthetic Aperture Radar) uses electromagnetic signals to acquire ground information and has been used for wide coverage reconnaissance missions regardless of weather conditions. However SAR is known to be vulnerable to interference signals by other communication devices or radar instruments and may suffer from undesirable performance degradations and image quality. In this paper, a modified Eigen-subspace based filter is proposed that can be easily applied to SAR images affected by interference signals. The method of constructing Eigen-subspace based filter is briefly described and various simulations are performed to show the performance of the interference mitigation process. The suppression filter is applied to a ALOS PALSAR raw data affected by interfering signals in order to verify its superiority over the Notch filter.

• The Magazine of the IEIE
• /
• v.34 no.11
• /
• pp.61-74
• /
• 2007

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.540-543
• /
• 2020

기존의 초해상도 딥러닝 기법은 모델의 깊이가 깊어지면서, 좋은 성능을 내지만 점점 더 복잡해지고 있고, 실제로 사용하는데 있어 많은 시간을 요구한다. 이를 해결하기 위해, 우리는 딥러닝 모델의 가중치를 양자화 하여 추론시간을 줄이고자 한다. 초해상도 모델은 feature extraction, non-linear mapping, reconstruction 세 부분으로 나누어져 있으며, 레이어 사이에 많은 skip-connection 이 존재하는 특징이 있다. 따라서 양자화 시 최종 성능 하락에 미치는 영향력이 레이어 별로 다르며, 이를 감안하여 강화학습으로 레이어 별 최적 bit 를 찾아 성능 하락을 최소화한다. 본 논문에서는 Skip-connection 이 많이 존재하는 MSRN 을 사용하였으며, 결과에서 feature extraction, reconstruction 부분과 블록 내 특정 위치의 레이어가 항상 높은 bit 를 가짐을 알 수 있다. 기존에 영상 분류에 한정되어 사용되었던 혼합 bit 양자화를 사용하여 초해상도 딥러닝 기법의 모델 사이즈를 줄인 최초의 논문이며, 제안 방법은 모바일 등 제한된 환경에 적용 가능할 것으로 생각된다.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_2
• /
• pp.793-810
• /
• 2022

Synthetic Aperture Radar (SAR) is considered to be suitable for near real-time inundation monitoring. The distinctly different intensity between water and land makes it adequate for waterbody detection, but the intrinsic speckle noise and variable intensity of SAR images decrease the accuracy of waterbody detection. In this study, we suggest two modules, named 'morphology module' and 'edge-enhanced module', which are the combinations of pooling layers and convolutional layers, improving the accuracy of waterbody detection. The morphology module is composed of min-pooling layers and max-pooling layers, which shows the effect of morphological transformation. The edge-enhanced module is composed of convolution layers, which has the fixed weights of the traditional edge detection algorithm. After comparing the accuracy of various versions of each module for U-Net, we found that the optimal combination is the case that the morphology module of min-pooling and successive layers of min-pooling and max-pooling, and the edge-enhanced module of Scharr filter were the inputs of conv9. This morphologic and edge-enhanced U-Net improved the F1-score by 9.81% than the original U-Net. Qualitative inspection showed that our model has capability of detecting small-sized waterbody and detailed edge of water, which are the distinct advancement of the model presented in this research, compared to the original U-Net.

• 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.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.195-202
• /
• 2005

An adaptive segmentation algorithm is used to efficiently target decisions in local non-stationary images. Until now, several adaptive approaches have been proposed as a method of segmentation. However, they can't be directly used for radar target detection because a radar signal has different characteristics from general images. Generally, a histogram of radar signal shows that targets have a relatively small number of frequency functions compared to the background and distribution of background, which have several shapes as the environment changes. In this paper, we propose an adaptive segmentation algorithm using a histogram chord which is a right-down line from maximum pick of frequency function. The proposed method provides thresholds which are optimum for several radar environments because the used chord for threshold search is not significantly effected by interference conditions. Simulation results show that the proposed method is superior to the traditional algorithms, global threshold method and distribution median method, with respect to detection performance.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.28 no.8
• /
• pp.636-645
• /
• 2017

A scanning radar is exploited widely such as for ground surveillance, disaster rescue, and etc. However, the range resolution is limited by transmitted bandwidth and cross-range resolution is limited by beam width. In this paper, we propose a method for super-resolution radar imaging. If the distribution of reflectivity is sparse, the distribution is called sparse signal. That is, the problem could be formulated as compressive sensing problem. In this paper, 2D super-resolution radar image is generated via reweighted ${\ell}_1-Minimization$. In the simulation results, we compared the images obtained by the proposed method with those of the conventional Orthogonal Matching Pursuit(OMP) and Synthetic Aperture Radar(SAR).

• The Korean Journal of Nuclear Medicine Technology
• /
• v.14 no.1
• /
• pp.101-104
• /
• 2010

Purpose: Raynaud scan is divided to flow, blood pool and local-delay image. Usually, we evaluate comparison through blood pool and local-delay image. We will evaluate about usability when comparative observe blood image and local-delay image in Raynaud scan that used $^{201}Tl$ as making flow image to one sheet of images. Materials and Methods: We have selected 29 Raynaud phenomenon patients aged 14~68 years who visited department of vascular surgery between Feb. 2008 and Aug. 2009. An intravenous injection $^{201}Tl$ of 111 MBq (3 mCi) to opposite side diagonal line limbs above an internal auditing department. Equipment used Philips gamma camera forte A-Z, and collimator used LEHR. Matrix size set up to each $64{\times}64$, $128{\times}128$, $256{\times}256$ and zoom factor used to full field. Protocol of dynamic is 2 second to 155 frames. Blood pool and delay count to 300 second. We set up ROI by a foundation to data acquired in PEGASYS processing program. Each results were analyzed with the SPSS 12.0 statistical software. Results: Each averages of count ratio (Rt / Lt) to have been given at composite image, a blood pool image, delay images analyzed at Raynaud phenomenon patients is $1.25{\pm}0.39$, $1.20{\pm}0.33$, $1.11{\pm}0.17$. The sample analysis results of blood pool image and delay image contented itself with p<0.029. Also, there don't have been each difference, and blood pool image, delay image regarding composite image was able to know. Conclusion: We were able to give help for comparison to evaluate a blood pool image and a local delay image at the Raynaud scan which used $^{201}Tl$ while making a flow image to one sheet image. Identification to be visual too was possible. If you are proceeded a researcher that there was further depth, you are more appropriate for, and you may get useful information.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.24 no.2
• /
• pp.207-213
• /
• 2013

This study describes an omega-K algorithm for focusing bistatic synthetic aperture radar(SAR) data using bistatic-to-monostatic conversion. Bistatic SAR system considered in this study consists of a transmitting antenna and a physical array of several receiving antennas. The length of the physical array is identical to the SAR synthetic aperture. Unlike the monostatic case, an omega-K algorithm for the bistatic case is difficult to obtain the exact equation in the 2D wavenumber domain. The key of the proposed algorithm is converting the bistatic data into a monostatic one. The effectiveness of the proposed algorithm is proved by simulation and real measurement data.