Nonuniformity correction algorithm with nonlinear model for infrared focalplane arrays

摘要

For the detector in infrared focal plane arrays (IRFPA) with a large dynamic range of response, a nonlin-ear model of response curve of the detector in IRFPA is introduced in this paper. With the model, theKalman-filter nonuniformity correction (NUC) algorithm with linear model, developed by Torres andHayat, is extended. In the extended algorithm, the raw image is translated into a linearized one firstlyby directly employing a logarithm-based transformation. Then the linearized image is corrected bythe Kalman-filter NUC algorithm with linear model, and the corrected linearized image is obtained.Finally the uniformity image of the original one is achieved by fulfilling an exponent transformationto the corrected linearized image. The presented algorithm not only inherits the advantage of the origi-nal algorithm that resolves the problem of the temporal drift in the gain and the bias in each detector byupdating NUC parameters with information of the current scene, but also reduce the influence of thedetector nonlinear response to the NUC performance, so it is suitable for IRFPA under large response-range. The NUC ability of the presented algorithm is demonstrated by experiments with real infraredimage sequences.