Biograph: Prof. Kyung-Tae Kim received the B.S. (1994), M.S. (1996), and Ph.D. (1999) degrees from Pohang University of Science and Technology (POSTECH), Pohang, South Korea, all in Electrical Engineering. From 2002 to 2010, he was a faculty member with the Department of Electronic Engineering, Yeungnam University. Since 2011, he has been with the Department of Electrical Engineering, POSTECH, Pohang, South Korea, and he is currently a Professor. During 2012-2017, He served as the Director of the Sensor Target Recognition Laboratory, sponsored by the Defense Acquisition Program Administration and the Agency for Defense Development. Currently, he is the Director of both the Unmanned Surveillance and Reconnaissance Technology (USRT) Research Center, and Radar & ElectroMagnetics Signal Processing (REMS) Laboratory at POSTECH. He is an author of about 200 papers on journals and conference proceedings, and has been the recipient of several outstanding research awards and best paper awards from the Korea Institute of Electromagnetic Engineering and Science (KIEES), and international conferences. He is a member of the IEEE and of the KIEES. He is currently carrying out several research projects funded by Korean government and several industries. His research interests are mainly in the field of radar signal processing and system modeling: SAR/ISAR imaging, target recognition, direction of arrival estimation, micro-Doppler analysis, automotive radars, digital beamforming, electronic warfare, and lectromagnetic scattering.
Title: Radar target recognition via ISAR images
Abstract: Inverse Synthetic Aperture Radar (ISAR) images show a two-dimensional (2-D) spatial distribution of scattering features of a target. There are several important issues in forming robust and high resolution ISAR images: translational and rotational motion compensation for focused ISAR images, selection of time-frame and length for robust image formation, and cross-range scaling for proper image interpretation. Over the past ten years, many researchers have been dealing with these issues via various signal processing tools and techniques. However, the applications of ISAR images have not been satisfactorily addressed. In particular, some researchers made an attempt to recognize target types using ISAR images via their own distinctive ways, but it seems that those attempts did not based on any solid theoretical background and could not suggest any unified framework for target recognition via ISAR images. The goal of target recognition via ISAR images can be achieved, only when the knowledge on scattering phenomenology, ISAR image formation, pattern recognition, and machine learning is effectively combined and merged to identify a specific ISAR image of a target. In this tutorial, various endeavors for past 20 years in my laboratory to enhance the target recognition performance using ISAR images will be summarized and reviewed. First, several issues of ISAR images to be addressed for target recognition will be reviewed: scattering phenomenology, cross-range scaling, variable cross-range resolution. Then, the process of feature extraction from ISAR images is discussed to enhance target recognition capability, compared to the direct use of ISAR image itself, i.e., template matching. Finally, the classifier architectures for ISAR images, developed in my laboratory, are presented, and they are extended to bi-static and multi-static ISAR images via information fusion strategy.