Implementation of Lucky Imaging Algorithm Based on FPGA

Lucky imaging is a kind of high-resolution image reconstruction techniques used to eliminate the influence of atmospheric turbulence on astronomical images. Its algorithm is usually implemented on the CPU-based platform, and cannot meet the requirement of real-time high-resolution imaging. Taking advantages of FPGA parallel processing, this paper presents a new lucky imaging algorithm based on FPGA, and builds an experimental system which uses FPGA to complete all the lucky imaging processes of the image selection, registration, stacking. The obtained image reconstructed by the FPGA system is exactly the same with the image processed by the traditional CPU platform. Nevertheless, it is important that the processing speed of the system is 19 times faster than that of the CPU platform. The implementation of the algorithm on FPGA provides us with a feasible scheme to make lucky imaging technology real-time or quasi real-time.