Real-Time Camera-Independent Image Processing System For Long-Range Tactical Imaging Applications

We propose to identify the algorithms, architectures and performance evaluation criteria for real-time, high-resolution, long-range imaging through atmospheric turbulence. The most promising algorithms will be implemented on a hardware acceleration platform. The end-goal for phase 1 is to develop a complete system design ready to be implemented in hardware for real-time video processing of 2Kx2K video imagery with frame rates of 400HZ. The key to achieving this end-goal is an integrated and experienced team consisting of hardware platforms experts, imaging specialists and system-level designers. Working together, this team can simultaneously develop and integrate the image processing algorithms for turbulence mitigation, build the necessary hardware platforms, and optimize the system further for better serving military needs. For phase 1 option we propose to investigate the use of high-speed, multi-channel deformable mirrors hardware and stochastic optimization algorithms to improve imaging performance beyond what is possible without optical wavefront aberration correction.