Abstract: 

With the rapid development of high-resolution radar technology and artificial intelligence, deep learning-based target detection, target recognition, land use/land cover classification, change detection, etc., have become a research hotspot in the field of radar information processing and image interpretation. The efficacies of these methods have been demonstrated in many radar applications. Large-scale, diversified and high-quality radar data (HRRP/SAR/ISAR/PolSAR, etc.) are the core basis for achieving good generalization of deep learning-based interpretation methods. However, due to the limitation of observation angles, non-cooperation of targets, and high cost of outfield experiments, radar sample quantity and diversity are insufficient in some application scenarios. It seriously restricts the research of deep learning-based methods and the training of network models. In order to solve this problem, the researchers conduct radar data simulation based on electromagnetic scattering modeling, electromagnetic calculation, target/clutter scattering characterization and other means. In addition, in recent years, methods such as autoencoder (AE), generative adversarial network (GAN) and Diffusion Model have been applied to radar sample generation, and related research has been widely concerned and developed rapidly. This session will cover the latest research on radar scattering characterization, radar data simulation, radar sample generation, radar image enhancement. The scenarios include aerial, space, maritime, and ground targets along with background clutter. The topics include, but are not limited to:

 Radar data simulation based on electromagnetic scattering modeling and calculation

 Target/clutter characterization and data simulation

 Radar sample generation based on deep learning methods

 Radar image enhancement, including despeckling, super resolution, etc

 Conversion from optical image to radar image

 Conversion of radar image cross different view/sensor

 Quality assessment of generated radar data

Session Chairs: 

Assoc. Prof. Wei Wang (National University of Defense Technology, China), 

Prof. Ferdinando Nunziata (Sapienza University of Rome, Italy), 

Prof. Yanhua Wang (Beijing Institute of Technology, China)