Journal of Physics Conference Series · 2018 · 14 citations · 9 references
ComputingComputer EngineeringComputer ScienceComputer GraphicsParallel ComputingUnmanned Aircraft SystemsDigital Image ProcessingGraphics Processing Unit ArchitectureGraphics Processing Unit ComputingGraphics Processing UnitsHardware AccelerationImage ProcessingImage Processing On GpusPolarimetric Radar ProcessingReal-time OperationSynthetic Aperture RadarSynthetic Aperture Radar DataSynthetic Aperture Radar ImagingSynthetic Aperture Radar Remote SensingUnmanned Aerial Vehicle Imaging
High-resolution miniature unmanned aerial vehicle (UAV) synthetic aperture radar (SAR) imaging requires not only effective and high-precision imaging algorithms, but also an efficient processing platform with real-time capability. In this paper, a graphics processing unit (GPU) based parallel scheme is presented to solve those problems. The polar format algorithm (PFA) is accelerated by utilizing the principle of chirp scaling (PCS) to avoid the two-dimensional (2-D) interpolation. Moreover, an efficient 2-D autofocus processing scheme is applied to compensate for the serious motion error caused by air turbulence. The specific realization on GPU and the main optimization method are investigated. The GPU based scheme is now able to process 4096×8192 complex image in a single precision within 4.23 seconds. The experimental results validates the effectiveness and efficiency of the scheme.
9
UAVSAR: A New NASA Airborne SAR System for Science and Technology Research
P. A. Rosen, S. Hensley, Kevin Wheeler et al. · 2006 · 132 citations
Natural Sciences, Aerospace Engineering, Aircraft Design +16
Multi-Subaperture PGA for SAR Autofocusing
Daiyin Zhu, Rui Jiang, Xinhua Mao et al. · IEEE Transactions on Aerospace and Electronic Systems · 2013 · 111 citations
Natural Sciences, Geoinformatics, Digital Elevation Modeling +11