Abstract: To reduce the effect that the detector nonlinear response imposes on the nonuniformity correction, a nonuniformity correction algorithm based on the Kalman-filter with a piecewise model is presented. This paper introduces a piecewise linear model of a detector response curve. And the Kalman-filter based nonuniformity correction algorithm is improved, which adapts to infrared focal plane arrays with nonlinearity of the response characteristic. Compared with the original algorithm, it can solve not only the problems of detector offset and gain drift with time, but also the influence of the detector nonlinear response to the nonuniformity correction performance to some extent. Therefore it achieves a better effect of the nonuniformity correction than the original algorithm. The nonuniformity correction capability of the improved algorithm is validated with real infrared image sequences.

Key words: nonuniformity correction, Kalman filter, nonlinear model, infrared focal plane arrays