4 November 2010 The algorithm analysis on non-uniformity correction based on LMS adaptive filtering
Author Affiliations +
Abstract
The traditional least mean square (LMS) algorithm has the performance of good adaptivity to noise, but there are several disadvantages in the traditional LMS algorithm, such as the defect in desired value of pending pixels, undetermined original coefficients, which result in slow convergence speed and long convergence period. Method to solve the desired value of pending pixel has improved based on these problems, also, the correction gain and offset coefficients worked out by the method of two-point temperature non-uniformity correction (NUC) as the original coefficients, which has improved the convergence speed. The simulation with real infrared images has proved that the new LMS algorithm has the advantages of better correction effect. Finally, the algorithm is implemented on the hardware structure of FPGA+DSP.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Dongjun Zhan, Dongjun Zhan, Qun Wang, Qun Wang, Chensheng Wang, Chensheng Wang, Huawang Chen, Huawang Chen, } "The algorithm analysis on non-uniformity correction based on LMS adaptive filtering", Proc. SPIE 7854, Infrared, Millimeter Wave, and Terahertz Technologies, 78541N (4 November 2010); doi: 10.1117/12.869529; https://doi.org/10.1117/12.869529
PROCEEDINGS
10 PAGES


SHARE
Back to Top