FPGA-based Lucky Imaging Algorithm in Frequency Domain

Lucky imaging is a post-processing image restoration technique, which can effectively reduce the impact of atmospheric turbulence on image quality. The traditional frequency-domain lucky imaging algorithm is better than the traditional space-domain lucky imaging algorithm in imaging results. It mainly consists of three parts:image preprocessing, data selection and data superposition. This paper presented a FPGA-based lucky imaging algorithm in frequency domain, and built an experimental system on FPGA. This system implements two-dimensional fast Fourier transform (2D-FFT) on astronomical image data by converting two-dimensional to two one-dimensional processing methods; And the sum of the square of the real part and the square of the imaginary part of the frequency data is used to replace the amplitude for data selection. The obtained high-resolution image reconstructed by the FPGA system is basically the same with the image processed by the CPU platform. The experimental system can process 2000 frames of 128 × 128 pixels image in frequency domain, what's more, the processing speed of the system on FPGA is 13 times faster than that on CPU. It accelerates the traditional frequency-domain lucky imaging algorithm while provides a feasible technical scheme for the real-time frequency-domain lucky imaging techniques in the future.