Adaptive Translational Motion Compensation Method for ISAR Imaging under Low SNR Based on Particle Swarm Optimization
Lei Liu,Feng Zhou,Mingliang Tao,Pange Sun,Zijing Zhang
DOI: https://doi.org/10.1109/jstars.2015.2491307
IF: 4.715
2015-01-01
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Abstract:Under low signal-to-noise ratio (SNR), the performance of conventional envelop-based range alignment methods for inverse synthetic aperture radar (ISAR) imaging degrades, resulting in the following phase adjustment or autofocus inapplicable. In this paper, a novel method for the translational motion compensation of ISAR imaging under low SNR is proposed. Translational motion is first modeled as a polynomial, and image quality evaluation metric (IQEM) such as image entropy, contrast, or peak value is utilized as the objective function to estimate the polynomial coefficient vector based on the particle swarm optimization (PSO). A PSO-based iteration process is presented to determine the polynomial order adaptively. Meanwhile, the computation burden of the proposed method is analyzed. In addition, a coarse estimation method of the polynomial coefficient vector is also discussed. Extensive experimental results verify the effectiveness and robustness of the proposed method.